Adventures in software engineering : plugging HCI & acessibility gaps with open source solutions
- Authors: Lansley, Alastair
- Date: 2020
- Type: Text , Thesis , PhD
- Full Text:
- Description: There has been a great deal of research undertaken in the field of Human-Computer Interfaces (HCI), input devices, and output modalities in recent years. From touch-based and voice control input mechanisms such as those found on modern smart-devices to the use of touch-free input through video-stream/image analysis (including depth streams and skeletal mapping) and the inclusion of gaze tracking, head tracking, virtual reality and beyond - the availability and variety of these I/O (Input/Output) mechanisms has increased tremendously and progressed both into our living rooms and into our lives in general. With regard to modern desktop computers and videogame consoles, at present many of these technologies are at a relatively immature stage of development - their use often limited to simple adjuncts to the staple input mechanisms of mouse, keyboard, or joystick / joypad inputs. In effect, we have these new input devices - but we're not quite sure how best to use them yet; that is, where their various strengths and weaknesses lie, and how or if they can be used to conveniently and reliably drive or augment applications in our everyday lives. In addition, much of this technology is provided by proprietary hardware and software, providing limited options for customisation or adaptation to better meet the needs of specific users. Therefore, this project investigated the development of open source software solutions to address various aspects of innovative user I/O in a flexible manner. Towards this end, a number of original software applications have been developed which incorporate functionality aimed at enhancing the current state of the art in these areas and making that software freely available for use by any who may find it beneficial.
- Description: Doctor of Philosophy
- Authors: Lansley, Alastair
- Date: 2020
- Type: Text , Thesis , PhD
- Full Text:
- Description: There has been a great deal of research undertaken in the field of Human-Computer Interfaces (HCI), input devices, and output modalities in recent years. From touch-based and voice control input mechanisms such as those found on modern smart-devices to the use of touch-free input through video-stream/image analysis (including depth streams and skeletal mapping) and the inclusion of gaze tracking, head tracking, virtual reality and beyond - the availability and variety of these I/O (Input/Output) mechanisms has increased tremendously and progressed both into our living rooms and into our lives in general. With regard to modern desktop computers and videogame consoles, at present many of these technologies are at a relatively immature stage of development - their use often limited to simple adjuncts to the staple input mechanisms of mouse, keyboard, or joystick / joypad inputs. In effect, we have these new input devices - but we're not quite sure how best to use them yet; that is, where their various strengths and weaknesses lie, and how or if they can be used to conveniently and reliably drive or augment applications in our everyday lives. In addition, much of this technology is provided by proprietary hardware and software, providing limited options for customisation or adaptation to better meet the needs of specific users. Therefore, this project investigated the development of open source software solutions to address various aspects of innovative user I/O in a flexible manner. Towards this end, a number of original software applications have been developed which incorporate functionality aimed at enhancing the current state of the art in these areas and making that software freely available for use by any who may find it beneficial.
- Description: Doctor of Philosophy
Positive attitude change to school - Narrative inquiry into adolescent students' lived experiences
- Authors: Wojtaszek, Sylwia
- Date: 2020
- Type: Text , Thesis , PhD
- Full Text:
- Description: This thesis reports on a qualitative research study that investigated adolescent students‟ experiences of positive attitude change to school. The literature review situates the research of students‟ attitude changes to school within the affective component of the multidimensional construct of student engagement and identifies this field and the phenomenon of positive attitude change to school as underresearched and under-theorised. Narrative inquiry methodology was applied in order to provide a detailed description of students‟ lived experiences and generate knowledge to fill the existing gap of how such an experience manifests itself. Eight students, who self-reported to have experienced positive attitude change to school, shared stories through in-depth semi-structured interviews of how the attitude change came about, who or what influenced it, and what meaning they attached to it in relation to its impact on their engagement and wellbeing. Students‟ attitudes to school are predominantly examined through quantitative research, whereas this study provided a unique and nuanced insight into attitude change based on the qualitative paradigm and a social constructionist view of the experience from the students‟ vantage point. Students‟ narrative accounts are compared and contrasted with each other to identify five resonant threads associated with the experience of positive attitude change to school. Data analysis suggests that positive attitude change to school has a significant impact on student engagement in learning and student wellbeing through its embodiment of perceived positive emotions associated with being at school. It consequently illustrates the relevance of broadening the understanding of such an experience to address the critical issue of disengagement in adolescent students. Key findings indicate that students develop a negative attitude to school when personal problems remain unresolved or have been insufficiently addressed within the school environment; no “helping hand” was there to assist these ambitious students who were struggling to engage in learning due to their experience of negative emotions at school. This research study has revealed that a negative attitude to school does not necessarily equate to a negative attitude to learning. Students‟ perception of the available support, both from the teachers and the services offered at school, is a critical factor in the transformation of their attitudes to school. Further, the students who participated in this study did not themselves feel that they were equipped with the required knowledge and skills to manage their personal problems effectively in order to maintain their engagement in learning. Only after having “hit rock bottom” and having sought help from outside the school environment were the students able to apply a different perspective to their circumstances that was associated with positive attitude change to school. From this research study it can be concluded that a student‟s positive attitude to school is a requirement for successful social and academic outcomes, and it is an educational goal in itself regarding the notion of developing lifelong learners. Personal problems and their impact on student engagement and wellbeing need to be acknowledged and catered for within the school environment. School support services must proactively extend a helping hand to students who have a negative attitude to school. Further, students need to develop selfefficacy regarding their personal wellbeing so that they become confident to act autonomously in solving their situations at school that are characterised by the difficult negative emotions that they are experiencing. Students‟ attitudes to school and the complexity of the multidimensional construct of student engagement need to be considered in the development of initiatives to address adolescent student disengagement and in the development of student wellbeing frameworks.
- Description: Doctor of Philosophy
- Authors: Wojtaszek, Sylwia
- Date: 2020
- Type: Text , Thesis , PhD
- Full Text:
- Description: This thesis reports on a qualitative research study that investigated adolescent students‟ experiences of positive attitude change to school. The literature review situates the research of students‟ attitude changes to school within the affective component of the multidimensional construct of student engagement and identifies this field and the phenomenon of positive attitude change to school as underresearched and under-theorised. Narrative inquiry methodology was applied in order to provide a detailed description of students‟ lived experiences and generate knowledge to fill the existing gap of how such an experience manifests itself. Eight students, who self-reported to have experienced positive attitude change to school, shared stories through in-depth semi-structured interviews of how the attitude change came about, who or what influenced it, and what meaning they attached to it in relation to its impact on their engagement and wellbeing. Students‟ attitudes to school are predominantly examined through quantitative research, whereas this study provided a unique and nuanced insight into attitude change based on the qualitative paradigm and a social constructionist view of the experience from the students‟ vantage point. Students‟ narrative accounts are compared and contrasted with each other to identify five resonant threads associated with the experience of positive attitude change to school. Data analysis suggests that positive attitude change to school has a significant impact on student engagement in learning and student wellbeing through its embodiment of perceived positive emotions associated with being at school. It consequently illustrates the relevance of broadening the understanding of such an experience to address the critical issue of disengagement in adolescent students. Key findings indicate that students develop a negative attitude to school when personal problems remain unresolved or have been insufficiently addressed within the school environment; no “helping hand” was there to assist these ambitious students who were struggling to engage in learning due to their experience of negative emotions at school. This research study has revealed that a negative attitude to school does not necessarily equate to a negative attitude to learning. Students‟ perception of the available support, both from the teachers and the services offered at school, is a critical factor in the transformation of their attitudes to school. Further, the students who participated in this study did not themselves feel that they were equipped with the required knowledge and skills to manage their personal problems effectively in order to maintain their engagement in learning. Only after having “hit rock bottom” and having sought help from outside the school environment were the students able to apply a different perspective to their circumstances that was associated with positive attitude change to school. From this research study it can be concluded that a student‟s positive attitude to school is a requirement for successful social and academic outcomes, and it is an educational goal in itself regarding the notion of developing lifelong learners. Personal problems and their impact on student engagement and wellbeing need to be acknowledged and catered for within the school environment. School support services must proactively extend a helping hand to students who have a negative attitude to school. Further, students need to develop selfefficacy regarding their personal wellbeing so that they become confident to act autonomously in solving their situations at school that are characterised by the difficult negative emotions that they are experiencing. Students‟ attitudes to school and the complexity of the multidimensional construct of student engagement need to be considered in the development of initiatives to address adolescent student disengagement and in the development of student wellbeing frameworks.
- Description: Doctor of Philosophy
An investigation of Surf Life Saving Australia's Junior Development Program for Nippers
- Authors: Higgerson, Amanda
- Date: 2020
- Type: Text , Thesis , Masters
- Full Text:
- Description: The primary aim of Surf Life Saving Australia (SLSA), through their State and Territory bodies is to provide safe coastal environments for Australian beach goers. With over 312 affiliated clubs across Australia, SLSA provides an opportunity for primary school aged children (Nippers) to experience water safety and aquatic education in an open water environment. Nippers in the Under 8-Under 13 years age groups are involved in a participation-based education program known as the Junior Development Program (JDP), which aims to develop skills required for future lifesaving (Giles & Slade, 2012). The overall aim of this research was to identify and explore the barriers and enablers to involvement in SLSA’s JDP from the perspective of the participants, their parents/carers and those involved in the delivery of the program. Utilising a multiphase mixed method design, 341 Nippers from eight surf lifesaving clubs in New South Wales completed a questionnaire and 30 interviews were conducted – 19 with parents/carers from seven of the eight participating lifesaving clubs, and 11 Age Managers from six of the eight clubs. The findings indicated that the Nippers perceived most lifesaving skills were important and were happy/very happy to be learning about safety of themselves and how to assist others in the surf. Most concerns related to unknowns in terms of sea creatures and adverse weather conditions like storms and lightening. Interviewees believed Nipper participation was based on interpersonal and intrapersonal reasons, with socialisation a positive aspect of the program. In line with common fears reported by Nippers, interviewees reported unfavourable environmental conditions, including big surf, cold weather, weed and stingers, as barriers to participation. The findings and opinions from those involved in the JDP provide insights into barriers and enablers to participation in the program, which is intended to provide important guidance for future delivery of the JDP across Australia.
- Description: Masters by Research
- Authors: Higgerson, Amanda
- Date: 2020
- Type: Text , Thesis , Masters
- Full Text:
- Description: The primary aim of Surf Life Saving Australia (SLSA), through their State and Territory bodies is to provide safe coastal environments for Australian beach goers. With over 312 affiliated clubs across Australia, SLSA provides an opportunity for primary school aged children (Nippers) to experience water safety and aquatic education in an open water environment. Nippers in the Under 8-Under 13 years age groups are involved in a participation-based education program known as the Junior Development Program (JDP), which aims to develop skills required for future lifesaving (Giles & Slade, 2012). The overall aim of this research was to identify and explore the barriers and enablers to involvement in SLSA’s JDP from the perspective of the participants, their parents/carers and those involved in the delivery of the program. Utilising a multiphase mixed method design, 341 Nippers from eight surf lifesaving clubs in New South Wales completed a questionnaire and 30 interviews were conducted – 19 with parents/carers from seven of the eight participating lifesaving clubs, and 11 Age Managers from six of the eight clubs. The findings indicated that the Nippers perceived most lifesaving skills were important and were happy/very happy to be learning about safety of themselves and how to assist others in the surf. Most concerns related to unknowns in terms of sea creatures and adverse weather conditions like storms and lightening. Interviewees believed Nipper participation was based on interpersonal and intrapersonal reasons, with socialisation a positive aspect of the program. In line with common fears reported by Nippers, interviewees reported unfavourable environmental conditions, including big surf, cold weather, weed and stingers, as barriers to participation. The findings and opinions from those involved in the JDP provide insights into barriers and enablers to participation in the program, which is intended to provide important guidance for future delivery of the JDP across Australia.
- Description: Masters by Research
A good sheep run. Letters from New South Wales in Scottish newspapers between 1820 and 1850 with potential to influence decisions on emigration
- Authors: Hannaford, Graham
- Date: 2020
- Type: Text , Thesis , PhD
- Full Text:
- Description: The primary aim of this thesis is to contribute to ongoing historical research into migration to and settlement in Australia by Scots. It achieves this by identifying and examining letters sent from the colonies in New South Wales which were printed in historic Scottish newspapers between 1820 and 1850. In examining the material, this thesis argues that the letters had potential to influence emigration decisions by Scots. The study shows some of the ways in which New South Wales was reported in the Scottish press and compares those reports with conditions in Scotland at the time. The comparisons and analyses of the letters, with consideration of their authors and likely readers as well as the newspapers in which they were printed demonstrate that the letters did have potential to influence emigration decisions. Its particular contribution to knowledge arises from demonstrating how mostly private letters which became publicly available through publication in newspapers had potential to influence emigrants’ decisions about moving to Australia. Rather than claiming direct evidence of the publication of particular letters as having influenced emigration, it shows how reporting of conditions in Australia when set into a context of contemporary events and conditions in Scotland had potential to influence decisions. It is grounded in the body of historical research about colonial Australia and sits within this Australian historiographical context. Given the motivations and attractions of Scots to colonial Australia this thesis also engages with techniques and theoretical approaches associated with Scottish diaspora studies, an area of research that often emphasises other Scottish migration patterns to Canada, New Zealand and the USA. When considered together both of these historiographical approaches lend themselves to primary source material analysis and a methodological approach that this doctoral study uses to examine the motivations of Scots who migrated to colonial Australia.
- Description: Doctor of Philosophy
- Authors: Hannaford, Graham
- Date: 2020
- Type: Text , Thesis , PhD
- Full Text:
- Description: The primary aim of this thesis is to contribute to ongoing historical research into migration to and settlement in Australia by Scots. It achieves this by identifying and examining letters sent from the colonies in New South Wales which were printed in historic Scottish newspapers between 1820 and 1850. In examining the material, this thesis argues that the letters had potential to influence emigration decisions by Scots. The study shows some of the ways in which New South Wales was reported in the Scottish press and compares those reports with conditions in Scotland at the time. The comparisons and analyses of the letters, with consideration of their authors and likely readers as well as the newspapers in which they were printed demonstrate that the letters did have potential to influence emigration decisions. Its particular contribution to knowledge arises from demonstrating how mostly private letters which became publicly available through publication in newspapers had potential to influence emigrants’ decisions about moving to Australia. Rather than claiming direct evidence of the publication of particular letters as having influenced emigration, it shows how reporting of conditions in Australia when set into a context of contemporary events and conditions in Scotland had potential to influence decisions. It is grounded in the body of historical research about colonial Australia and sits within this Australian historiographical context. Given the motivations and attractions of Scots to colonial Australia this thesis also engages with techniques and theoretical approaches associated with Scottish diaspora studies, an area of research that often emphasises other Scottish migration patterns to Canada, New Zealand and the USA. When considered together both of these historiographical approaches lend themselves to primary source material analysis and a methodological approach that this doctoral study uses to examine the motivations of Scots who migrated to colonial Australia.
- Description: Doctor of Philosophy
Investigating the configuration of a flight training device for visual flight rules navigation
- Authors: Harvey, William
- Date: 2020
- Type: Text , Thesis , PhD
- Full Text:
- Description: The acquisition of pilot navigational skills utilising visual landmarks is a crucial skill that is required as part of Visual Flight Rules navigation towards obtaining a Private Pilot License. Due to the high cost of pilot training simulators, industry has identified a need for research in efficient utilisation of low-end, low cost personal compute flight simulators to assist in developing pilot skills. Analysis of the effectiveness of the use of such personal computer simulators depend on proper configuration determined by measurable errors to define simulator fidelity. To date, research has shown that the configuration of these simulators appears to have been done in an ad-hoc fashion and not in a scientific fashion. Therefore, the problem that needed to be solved was how to effectively configure such simulators. This thesis research attempted to solve this problem and present the process for effectively configuring a personal computer simulator, or flight training device, capable of successful Visual Flight Rules navigation. The simulator was configurated utilising a process that followed an interpretation of the Design Science research method, and an error correction model to determine the errors in the simulator configuration. This was done by comparing two probability distributions to measure the maximum error variable distance in order to configure a simulator suitable for the acquisition of Visual Flight Rules navigation piloting skills required for obtaining a Private Pilot Licence in Australia. This error identification method was then used to indicate simulator configuration efficiency and fidelity in order to achieve a minimum suitable configuration and setup. Further application of the findings of this research could potentially lead to the configuration of different types of non-aviation simulators, in particular Part-Task-Trainers and other training devices, including Virtual Reality Augmented Reality devices utilising various types of platforms such as Windows, Apple, and Android.
- Description: Doctor of Philosophy
- Authors: Harvey, William
- Date: 2020
- Type: Text , Thesis , PhD
- Full Text:
- Description: The acquisition of pilot navigational skills utilising visual landmarks is a crucial skill that is required as part of Visual Flight Rules navigation towards obtaining a Private Pilot License. Due to the high cost of pilot training simulators, industry has identified a need for research in efficient utilisation of low-end, low cost personal compute flight simulators to assist in developing pilot skills. Analysis of the effectiveness of the use of such personal computer simulators depend on proper configuration determined by measurable errors to define simulator fidelity. To date, research has shown that the configuration of these simulators appears to have been done in an ad-hoc fashion and not in a scientific fashion. Therefore, the problem that needed to be solved was how to effectively configure such simulators. This thesis research attempted to solve this problem and present the process for effectively configuring a personal computer simulator, or flight training device, capable of successful Visual Flight Rules navigation. The simulator was configurated utilising a process that followed an interpretation of the Design Science research method, and an error correction model to determine the errors in the simulator configuration. This was done by comparing two probability distributions to measure the maximum error variable distance in order to configure a simulator suitable for the acquisition of Visual Flight Rules navigation piloting skills required for obtaining a Private Pilot Licence in Australia. This error identification method was then used to indicate simulator configuration efficiency and fidelity in order to achieve a minimum suitable configuration and setup. Further application of the findings of this research could potentially lead to the configuration of different types of non-aviation simulators, in particular Part-Task-Trainers and other training devices, including Virtual Reality Augmented Reality devices utilising various types of platforms such as Windows, Apple, and Android.
- Description: Doctor of Philosophy
Beyond the princess, the priestess and the galactic kitchen sink: Reformulation of feminine roles in certain work of Lois McMaster Bujold
- Authors: Herington, Caitlin
- Date: 2018
- Type: Text , Thesis , PhD
- Full Text:
- Description: In this thesis I examine the Science Fiction and Fantasy works of Lois McMaster Bujold in the Vorkosigan Series and Chalion Series, in particular the way she reformulates women’s roles and identities in society through the characters presented in these novels. I use the term Speculative Fiction as an umbrella term that encompasses both Science Fiction and Fantasy as modes of speculation, in that they both rely on extrapolation and estrangement as narrative features. My main proposition is that Bujold is an important transitional figure in speculative fiction between second and third wave feminist thinking. Although her work mimics some distinctive features of speculative fiction that utilise patriarchal structures and traditional gender norms, it is not limited by them. As a result, Bujold conveys a more complex and insightful understanding of gender. The research method of this thesis is the close reading of a range of sample texts from Bujold’s Vorkosigan Series and Chalion Series which feature female protagonists. I seek to explore the discussion of gender relations and reformulation that occurs within them in the context of both speculative and feminist criticism. Bujold’s exploration of the identities and social roles of women in these fictional worlds is complex and challenging, using a range of approaches from simple reversal, to hybridity of gender, to more complex partial positions. This thesis argues that she takes an implicitly feminist approach, focussing on female experiences and examining the modes of social control and exercise of power within patriarchal social structures as they impact on women. Science Fiction and Fantasy often seem to reiterate traditional patriarchal hierarchies. Validating gender norms that conform to social expectations rather than challenging them. Bujold is presented in this thesis as utilising established norms and tropes such that her texts are easily identified as examples of Science Fiction and Fantasy, but in other ways her reformulations present radical challenges to cultural expectations of gender. This thesis reveals that social critique and reformulation of gender roles is possible and powerful in both Science Fiction and Fantasy by examining the work of a significant author whose work has lacked critical attention until recently. Although numerous studies have examined the way gender has been treated in Science Fiction and Fantasy, the unique contribution of this thesis is to examine an author previously under-studied and to consider the patterns of these reformulations as expressed in Bujold’s works.
- Description: Doctor of Philosophy
- Authors: Herington, Caitlin
- Date: 2018
- Type: Text , Thesis , PhD
- Full Text:
- Description: In this thesis I examine the Science Fiction and Fantasy works of Lois McMaster Bujold in the Vorkosigan Series and Chalion Series, in particular the way she reformulates women’s roles and identities in society through the characters presented in these novels. I use the term Speculative Fiction as an umbrella term that encompasses both Science Fiction and Fantasy as modes of speculation, in that they both rely on extrapolation and estrangement as narrative features. My main proposition is that Bujold is an important transitional figure in speculative fiction between second and third wave feminist thinking. Although her work mimics some distinctive features of speculative fiction that utilise patriarchal structures and traditional gender norms, it is not limited by them. As a result, Bujold conveys a more complex and insightful understanding of gender. The research method of this thesis is the close reading of a range of sample texts from Bujold’s Vorkosigan Series and Chalion Series which feature female protagonists. I seek to explore the discussion of gender relations and reformulation that occurs within them in the context of both speculative and feminist criticism. Bujold’s exploration of the identities and social roles of women in these fictional worlds is complex and challenging, using a range of approaches from simple reversal, to hybridity of gender, to more complex partial positions. This thesis argues that she takes an implicitly feminist approach, focussing on female experiences and examining the modes of social control and exercise of power within patriarchal social structures as they impact on women. Science Fiction and Fantasy often seem to reiterate traditional patriarchal hierarchies. Validating gender norms that conform to social expectations rather than challenging them. Bujold is presented in this thesis as utilising established norms and tropes such that her texts are easily identified as examples of Science Fiction and Fantasy, but in other ways her reformulations present radical challenges to cultural expectations of gender. This thesis reveals that social critique and reformulation of gender roles is possible and powerful in both Science Fiction and Fantasy by examining the work of a significant author whose work has lacked critical attention until recently. Although numerous studies have examined the way gender has been treated in Science Fiction and Fantasy, the unique contribution of this thesis is to examine an author previously under-studied and to consider the patterns of these reformulations as expressed in Bujold’s works.
- Description: Doctor of Philosophy
Canonical Dual Algorithms for Global Optimization with Applications
- Authors: Zhou, Xiaojun
- Date: 2014
- Type: Text , Thesis
- Full Text:
- Description: Canonical duality theory provides a unified framework which can transform a nonconvex primal minimization problem to a canonical dual maximization problem over a convex domain without duality gap. But the global optimality is guaranteed by a certain positive definite condition and such condition is not always satisfied. The goal of this thesis aims to explore possible techniques that can be used to solve global optimization problems based on the canonical duality theory. Firstly, an algorithmic framework for canonical duality theory is established, which shows that the canonical dual algorithms can be developed in four aspects under the positive definite condition explicitly or implicitly, namely, (i) minimizing the primal problem, (ii) maximizing the canonical dual problem, (iii) solving a nonlinear equation caused by total complementary function, and (iv) solving a nonlinear equation caused by canonical dual function. Secondly, we show that if there exists a critical point of the canonical dual problem in the positive definite domain, by solving an equivalent semidefinite programming (SDP) problem, the corresponding global solution to the primal problem can be obtained easily via off-the-shelf software packages. A specific canonical dual algorithm is given for each problem, including sum of fourth-order polynomials minimization, nonconvex quadratically constrained quadratic program (QCQP), and boolean quadratic program (BQP). Thirdly, we propose a canonical primal-dual algorithm framework based on the total complementary function. Convergence analysis is discussed from the perspective of variational inequalities (VIs) and contraction methods. Specific canonical primal-dual algorithms for sum of fourth-order polynomials minimization is given as well. And a real-world application to the sensor network localization problem is illustrated. Next, a canonical sequential reduction approach is proposed to recover the approximate or global solution for the BQP problem. By fixing some previously known components, the original problem can be reduced sequentially to a lower dimension one. This approach is successfully applied to the well-known maxcut problem. Finally, we discuss the canonical dual approach applied to continuous time constrained optimal control. And it shows that the optimal control law for the n-dimensional constrained linear quadratic regulator can be achieved precisely via one-dimensional canonical dual variable, and for the optimal control problem with concave cost functional, an approximate solution can be obtained by introducing a linear perturbation term.
- Description: PhD
- Authors: Zhou, Xiaojun
- Date: 2014
- Type: Text , Thesis
- Full Text:
- Description: Canonical duality theory provides a unified framework which can transform a nonconvex primal minimization problem to a canonical dual maximization problem over a convex domain without duality gap. But the global optimality is guaranteed by a certain positive definite condition and such condition is not always satisfied. The goal of this thesis aims to explore possible techniques that can be used to solve global optimization problems based on the canonical duality theory. Firstly, an algorithmic framework for canonical duality theory is established, which shows that the canonical dual algorithms can be developed in four aspects under the positive definite condition explicitly or implicitly, namely, (i) minimizing the primal problem, (ii) maximizing the canonical dual problem, (iii) solving a nonlinear equation caused by total complementary function, and (iv) solving a nonlinear equation caused by canonical dual function. Secondly, we show that if there exists a critical point of the canonical dual problem in the positive definite domain, by solving an equivalent semidefinite programming (SDP) problem, the corresponding global solution to the primal problem can be obtained easily via off-the-shelf software packages. A specific canonical dual algorithm is given for each problem, including sum of fourth-order polynomials minimization, nonconvex quadratically constrained quadratic program (QCQP), and boolean quadratic program (BQP). Thirdly, we propose a canonical primal-dual algorithm framework based on the total complementary function. Convergence analysis is discussed from the perspective of variational inequalities (VIs) and contraction methods. Specific canonical primal-dual algorithms for sum of fourth-order polynomials minimization is given as well. And a real-world application to the sensor network localization problem is illustrated. Next, a canonical sequential reduction approach is proposed to recover the approximate or global solution for the BQP problem. By fixing some previously known components, the original problem can be reduced sequentially to a lower dimension one. This approach is successfully applied to the well-known maxcut problem. Finally, we discuss the canonical dual approach applied to continuous time constrained optimal control. And it shows that the optimal control law for the n-dimensional constrained linear quadratic regulator can be achieved precisely via one-dimensional canonical dual variable, and for the optimal control problem with concave cost functional, an approximate solution can be obtained by introducing a linear perturbation term.
- Description: PhD
Broadening the concept of school: how a re-configuration of school must be inclusive of students who are "put at" a disadvantage
- Authors: Peters, Edward (Keith)
- Date: 2013
- Type: Text , Thesis , PhD
- Full Text:
- Description: This thesis critically examines how students enrolled in state-funded schools can be ‘put at’ a disadvantage. I do this through examining two techniques of exclusion that stimulate student disconnection from school: first, the ways in which standards-based and performativity-driven learning outcomes are implemented to shape how student success is determined. Secondly, I examine how discourses around power control curriculum and student identity instil monological learning structures that normalises standards-based learning outcomes. Challenging this, I allow the voices of my co-researchers, the students in my thesis, to speak back to confront these school-based policies that allow disconnection to occur. Examining school policy and student voice at the point where they intersect allowed me to undertake an evaluation of how schools adversely affect students, and what students say they want from their experience of schooling. The final theme I develop is based on student and agency worker voice and what they say school-based learning should become. Relational learning and learning that develops students ethically emerged as fundamental strengths of what enriching learning transactions should look like. I argue that creating relational learning spaces develop challenging environments that can lead students to ethically understand their identity within complex social and cultural lifestyles. I argue that the ways in which schools are organised to administer time and space must be radically overhauled if this is to be achieved.
- Description: Doctor of Philosophy
- Authors: Peters, Edward (Keith)
- Date: 2013
- Type: Text , Thesis , PhD
- Full Text:
- Description: This thesis critically examines how students enrolled in state-funded schools can be ‘put at’ a disadvantage. I do this through examining two techniques of exclusion that stimulate student disconnection from school: first, the ways in which standards-based and performativity-driven learning outcomes are implemented to shape how student success is determined. Secondly, I examine how discourses around power control curriculum and student identity instil monological learning structures that normalises standards-based learning outcomes. Challenging this, I allow the voices of my co-researchers, the students in my thesis, to speak back to confront these school-based policies that allow disconnection to occur. Examining school policy and student voice at the point where they intersect allowed me to undertake an evaluation of how schools adversely affect students, and what students say they want from their experience of schooling. The final theme I develop is based on student and agency worker voice and what they say school-based learning should become. Relational learning and learning that develops students ethically emerged as fundamental strengths of what enriching learning transactions should look like. I argue that creating relational learning spaces develop challenging environments that can lead students to ethically understand their identity within complex social and cultural lifestyles. I argue that the ways in which schools are organised to administer time and space must be radically overhauled if this is to be achieved.
- Description: Doctor of Philosophy
An investigation into the agronomic factors affecting sustainability, surface hardness and rotational traction on community-level football grounds during drought conditions.
- Authors: Ford, Phillip
- Date: 2013
- Type: Text , Thesis , PhD
- Full Text:
- Description: Excessive hardness and rotational traction are the factors most associated with non-contact, ground-related injury risk in football. Irrigation has the greatest influence on surface hardness, and also in determining the turfgrass species that can be sustained. However, irrigation is prohibited on many Australian community-level grounds during drought. This thesis investigated the influence of various agronomic factors on surface hardness and on rotational traction, with the aim of devising strategies to reduce ground-related injury risk in drought conditions. In replicated plots tested over a drought period, Clegg hardness values on a football ground peaked at 160 gravities when turfgrass coverage was present, compared to values in excess of 200 gravities when turf coverage was absent. If the Clegg threshold was set at 160 g for community-level football, many grounds would stay open for play provided they sustained turfgrass coverage. In southern mainland Australia during drought and without irrigation, only a drought-resistant species such as couchgrass (Cynodon dactylon and hybrids) could do this. However, couchgrass has been linked by past epidemiological evidence to a higher risk of knee injury than perennial ryegrass (Lolium perenne), which was attributed to couchgrass having a higher rotational traction. In data presented in this thesis, perennial ryegrass actually had significantly higher rotational stiffness than couchgrass on four out of eight assessment dates. Consequently, there appears no reason to favour perennial ryegrass and to recommend against the use of couchgrass for reasons concerning rotational traction. Furthermore, the largest range in rotational traction was between areas of full grasscover compared to denuded areas. The thesis concluded that the ability to sustain turfgrass coverage provided the most effective agronomic solution for moderating both hardness and rotational traction on community-level football fields, and that a turfgrass species should be selected primarily on its ability to best sustain that coverage.
- Description: Doctor of Philosophy
- Authors: Ford, Phillip
- Date: 2013
- Type: Text , Thesis , PhD
- Full Text:
- Description: Excessive hardness and rotational traction are the factors most associated with non-contact, ground-related injury risk in football. Irrigation has the greatest influence on surface hardness, and also in determining the turfgrass species that can be sustained. However, irrigation is prohibited on many Australian community-level grounds during drought. This thesis investigated the influence of various agronomic factors on surface hardness and on rotational traction, with the aim of devising strategies to reduce ground-related injury risk in drought conditions. In replicated plots tested over a drought period, Clegg hardness values on a football ground peaked at 160 gravities when turfgrass coverage was present, compared to values in excess of 200 gravities when turf coverage was absent. If the Clegg threshold was set at 160 g for community-level football, many grounds would stay open for play provided they sustained turfgrass coverage. In southern mainland Australia during drought and without irrigation, only a drought-resistant species such as couchgrass (Cynodon dactylon and hybrids) could do this. However, couchgrass has been linked by past epidemiological evidence to a higher risk of knee injury than perennial ryegrass (Lolium perenne), which was attributed to couchgrass having a higher rotational traction. In data presented in this thesis, perennial ryegrass actually had significantly higher rotational stiffness than couchgrass on four out of eight assessment dates. Consequently, there appears no reason to favour perennial ryegrass and to recommend against the use of couchgrass for reasons concerning rotational traction. Furthermore, the largest range in rotational traction was between areas of full grasscover compared to denuded areas. The thesis concluded that the ability to sustain turfgrass coverage provided the most effective agronomic solution for moderating both hardness and rotational traction on community-level football fields, and that a turfgrass species should be selected primarily on its ability to best sustain that coverage.
- Description: Doctor of Philosophy
Efficacy of various complex and contrast set protocols acute responses and chronic muscle power development.
- Authors: Talpey, Scott
- Date: 2014
- Type: Text , Thesis
- Full Text:
- Description: PhD
- Description: "The complex and contrast methods are two separate, but related approaches to resistance training that present intriguing options for coaches who are seeking to enhance an athletes explosive lower body movement. This intrigue is caused by the post-activation potentiation (PAP) phenomenon, which provides the physiological underpinning for these training methods. Exploitation of the PAP response is believed to elicit acute responses, and ultimately chronic adaptations in an individuals explosive force producing ability to an extent that is greater than can be obtained through conventional training methods. However, evidence to support the use of these training methods is derived from investigations that have reported acute increases in performance within a single session and not over an extended period of time. Additionally, the complex method of training, in which sets of a heavy resistance exercise (conditioning action) are performed prior to sets of a lighter explosive exercise has not been investigated as extensively as the contrast method, in which heavy and light exercises are alternated in a set-by-set fashion, leading to question which method is more effective. Furthermore, it is currently not understood if a static or dynamic conditioning action is more effective in complex or contrast methods, and how a change in body position from seated to standing acutely effects explosive lower body movement. Therefore, four studies were designed to address these current gaps in the literature." "The first study sought to determine the reliability of the explosive force variables of jump height (JH), peak bar velocity (BV), peak force (PF), and peak power (PP) obtained from a countermovement jump (CMJ) test." "The second study of the thesis examined the acute responses to five different unconventional resistance training methods which attempted to capitalise on a PAP response and one conventional session where CMJs were performed without any influence from a previous conditioning action." "The third investigation of this thesis was a nine-week training study in which participants were randomly assigned to two training interventions, conventional (n = 11) or complex (n = 9)." "Due to the dramatic improvements observed in CMJ height, that were not observed in other variables, it was questioned whether the specific instructions provided to participants during training and testing throughout the study to jump for maximal height had any influence on these results. This question led to the design of a fourth study. Though not directly related to complex or contrast training, this investigation into the effects of instructions on CMJ variables has scope to alter the way this test is conducted both in research and in a practical setting.
- Authors: Talpey, Scott
- Date: 2014
- Type: Text , Thesis
- Full Text:
- Description: PhD
- Description: "The complex and contrast methods are two separate, but related approaches to resistance training that present intriguing options for coaches who are seeking to enhance an athletes explosive lower body movement. This intrigue is caused by the post-activation potentiation (PAP) phenomenon, which provides the physiological underpinning for these training methods. Exploitation of the PAP response is believed to elicit acute responses, and ultimately chronic adaptations in an individuals explosive force producing ability to an extent that is greater than can be obtained through conventional training methods. However, evidence to support the use of these training methods is derived from investigations that have reported acute increases in performance within a single session and not over an extended period of time. Additionally, the complex method of training, in which sets of a heavy resistance exercise (conditioning action) are performed prior to sets of a lighter explosive exercise has not been investigated as extensively as the contrast method, in which heavy and light exercises are alternated in a set-by-set fashion, leading to question which method is more effective. Furthermore, it is currently not understood if a static or dynamic conditioning action is more effective in complex or contrast methods, and how a change in body position from seated to standing acutely effects explosive lower body movement. Therefore, four studies were designed to address these current gaps in the literature." "The first study sought to determine the reliability of the explosive force variables of jump height (JH), peak bar velocity (BV), peak force (PF), and peak power (PP) obtained from a countermovement jump (CMJ) test." "The second study of the thesis examined the acute responses to five different unconventional resistance training methods which attempted to capitalise on a PAP response and one conventional session where CMJs were performed without any influence from a previous conditioning action." "The third investigation of this thesis was a nine-week training study in which participants were randomly assigned to two training interventions, conventional (n = 11) or complex (n = 9)." "Due to the dramatic improvements observed in CMJ height, that were not observed in other variables, it was questioned whether the specific instructions provided to participants during training and testing throughout the study to jump for maximal height had any influence on these results. This question led to the design of a fourth study. Though not directly related to complex or contrast training, this investigation into the effects of instructions on CMJ variables has scope to alter the way this test is conducted both in research and in a practical setting.
- Authors: Curry, Christina
- Date: 2012
- Type: Text , Thesis , PhD
- Full Text: false
- Description: With the movement to evolving classroom practices and pedagogies to enhance student-centered learning environments across all Key Learning Areas, there has been growing concern about how educators can produce high quality, intellectual learning experiences within physical education. To provide much-needed understanding of teachers' experiences of the implementation of a TGfU (Teaching Games for Understanding) teaching approach, this study aimed to identify the ways in which individual teachers, adopt, embrace or alternatively resist TGfU as an innovative pedagogy. - Taken from abstract.
- Description: Doctor of Philosophy
Why firms in China go green and how they market green?
- Authors: Song-Turner, Helen
- Date: 2014
- Type: Text , Thesis , PhD
- Full Text:
- Description: Since the late 1970s China has strived to build a globally competitive market economy based on a range of industry sectors, focusing on manufacturing and allied industries. By 2014 the size of China’s economy is second only to that of the United States of America. A challenge in this context and situation—where economic growth is still a key area of policy focus for the government—is dealing with associated environmental matters and issues. By the turn of the 21st Century, environmental issues were receiving an unprecedented degree of attention from the Chinese Government, and one of the subsets of this issue is the concept of green marketing. A question arises regarding the possibilities of successful growing firms in China applying the concepts, principles and practices which embody green marketing within a rubric and context of sustained economic development. To gain a deep understanding of ways in which the concepts, applications and conditions of green marketing actually work in China, this research explores why firms in China go green and how they market their green products and services in China using an emic approach that emphasizes an indigenous, within culture perspective of firms’ behaviour in the Chinese context. Related to the research question this research explores a range of possible motivating factors and ascertains features of key influential stakeholders which might well influence firms’ green marketing approaches and practices in China. These firms are exceptional in the way they have addressed green marketing. The research undertaken seeks to identify ways in which these factors translate in marketing terms within a green marketing paradigm. Given that green marketing is arguably where economic development and environmental matters tend to interact and intersect within a market economy, this research provides insights to how sustainable Chinese firms deal with this complex and important issue. The use of a case study approach, across a range of industry sectors and in a variety of locations in China, provides depth and realism to this research. The result of this research contribute significantly to the understanding of how Chinese managers perceive ecological sustainability and expand firm stakeholder theory by identifying key influential stakeholders in proactive green decisions. The ways in which firms define and identify key influential stakeholders in their green decisions are influenced by resource dependency, the institutional power of the stakeholders, and moderated by firm’s characteristics and life cycle stages. Firms operating in a strong government-driven setting derive four conceptual green motivations—philosophical and social responsibility, management of risk reduction, competition pressure, and special events— that offer insights on understanding firms’ green behaviours and green orientations in China.
- Authors: Song-Turner, Helen
- Date: 2014
- Type: Text , Thesis , PhD
- Full Text:
- Description: Since the late 1970s China has strived to build a globally competitive market economy based on a range of industry sectors, focusing on manufacturing and allied industries. By 2014 the size of China’s economy is second only to that of the United States of America. A challenge in this context and situation—where economic growth is still a key area of policy focus for the government—is dealing with associated environmental matters and issues. By the turn of the 21st Century, environmental issues were receiving an unprecedented degree of attention from the Chinese Government, and one of the subsets of this issue is the concept of green marketing. A question arises regarding the possibilities of successful growing firms in China applying the concepts, principles and practices which embody green marketing within a rubric and context of sustained economic development. To gain a deep understanding of ways in which the concepts, applications and conditions of green marketing actually work in China, this research explores why firms in China go green and how they market their green products and services in China using an emic approach that emphasizes an indigenous, within culture perspective of firms’ behaviour in the Chinese context. Related to the research question this research explores a range of possible motivating factors and ascertains features of key influential stakeholders which might well influence firms’ green marketing approaches and practices in China. These firms are exceptional in the way they have addressed green marketing. The research undertaken seeks to identify ways in which these factors translate in marketing terms within a green marketing paradigm. Given that green marketing is arguably where economic development and environmental matters tend to interact and intersect within a market economy, this research provides insights to how sustainable Chinese firms deal with this complex and important issue. The use of a case study approach, across a range of industry sectors and in a variety of locations in China, provides depth and realism to this research. The result of this research contribute significantly to the understanding of how Chinese managers perceive ecological sustainability and expand firm stakeholder theory by identifying key influential stakeholders in proactive green decisions. The ways in which firms define and identify key influential stakeholders in their green decisions are influenced by resource dependency, the institutional power of the stakeholders, and moderated by firm’s characteristics and life cycle stages. Firms operating in a strong government-driven setting derive four conceptual green motivations—philosophical and social responsibility, management of risk reduction, competition pressure, and special events— that offer insights on understanding firms’ green behaviours and green orientations in China.
- Authors: Teusner, Annabel
- Date: 2013
- Type: Text , Thesis , PhD
- Full Text: false
- Description: Medium sized workplaces are often grouped with small enterprises, yet they can have unique characteristics compared to more intensively studied small and large organisations, however, these differences are often not clearly recognised. The food processing industry is often part of the small and medium sized enterprises (SME) group and has a higher rate of accidents, injury and disease compared with many other industry groups. To improve safety performance in medium enterprises (within the food processing industry), there is a need to identify the perceived barriers to improving OHS amongst managers, supervisors, and process workers. This research provides a unique insider’s point of view of a food packing room within a food manufacturing plant in Melbourne. It is a single case design study, focusing on a specific medium sized enterprise, using insider research and an ethnographic approach to gain a greater understanding of the perceived barriers to improving OHS. The first part of this research focused on understanding the safety culture of the organisation, from different hierarchical perspectives, using Hudson’s Evolution of Safety Cultures model. The second part utilised the Health Belief model to identify the barriers to improving OHS from multiple perspectives of managers, supervisors and process workers. Following the analysis of the results, five barriers were identified: (i) fragmentation of group perspectives; (ii) poor communication between and within the hierarchical groups; (iii) limited OHS training and competency; (iv) inadequate management commitment to OHS; and (v) lack of empowerment. This created tensions amongst and between the groups. However, further examination of these findings suggested a deeper reason for the difficulty in improving the OHS, which was affecting the organisation. The most significant barrier to improving OHS at this workplace was the learned helplessness that had developed over a period of time and had filtered through all hierarchical groups. A reflection on the choice of methodology (insider research) and the importance of ethical considerations and the depth of meaning that was drawn from this research has been presented in this thesis. Learned helplessness has been discussed in the context of a medium sized enterprise and its significance to the contribution to OHS literature. Recommendations have been provided in terms of addressing learned helplessness within a medium sized enterprise, by changing the way individuals think, using learned optimism. Other changes that have been presented have included organisation design, facilitating improved communication and goal setting, and the investigation into the supporting role of supply chains toward OHS initiatives.
- Description: Doctor of Philosophy
Nonsmooth optimization models and algorithms for data clustering and visualization
- Authors: Mohebi, Ehsan
- Date: 2015
- Type: Text , Thesis
- Full Text:
- Description: Cluster analysis deals with the problem of organization of a collection of patterns into clusters based on a similarity measure. Various distance functions can be used to define this measure. Clustering problems with the similarity measure defined by the squared Euclidean distance have been studied extensively over the last five decades. However, problems with other Minkowski norms have attracted significantly less attention. The use of different similarity measures may help to identify different cluster structures of a data set. This in turn may help to significantly improve the decision making process. High dimensional data visualization is another important task in the field of data mining and pattern recognition. To date, the principal component analysis and the self-organizing maps techniques have been used to solve such problems. In this thesis we develop algorithms for solving clustering problems in large data sets using various similarity measures. Such similarity measures are based on the squared L
- Description: Doctor of Philosophy
- Authors: Mohebi, Ehsan
- Date: 2015
- Type: Text , Thesis
- Full Text:
- Description: Cluster analysis deals with the problem of organization of a collection of patterns into clusters based on a similarity measure. Various distance functions can be used to define this measure. Clustering problems with the similarity measure defined by the squared Euclidean distance have been studied extensively over the last five decades. However, problems with other Minkowski norms have attracted significantly less attention. The use of different similarity measures may help to identify different cluster structures of a data set. This in turn may help to significantly improve the decision making process. High dimensional data visualization is another important task in the field of data mining and pattern recognition. To date, the principal component analysis and the self-organizing maps techniques have been used to solve such problems. In this thesis we develop algorithms for solving clustering problems in large data sets using various similarity measures. Such similarity measures are based on the squared L
- Description: Doctor of Philosophy
Green IT: Sustainability by aligning business requirements with IT resource utilization
- Authors: Subburaj, Srikanth
- Date: 2012
- Type: Text , Thesis , Masters
- Full Text:
- Description: The term “green IT” is defined as “Optimal use of Information and Communication Technology (ICT) for managing the environment sustainability of enterprise operations throughout their life cycles.” Its objective is to create a positive impact on environment through net lowered emissions. The heart of green IT (Information Technology) is the concept of “Environmental Sustainability,” its initiatives are multifaceted to support enterprises “business as usual model,” in low-carbon economy. Many green IT initiatives have provided short–term financial benefits, based on which organizations are now focusing on increased energy efficiency. Since many of these approaches provide less financials benefits’ which are harder to implement but provides improved environmental performance. So the financial benefits and environmental performance should be the multiple objectives in the green IT study.
- Description: Master of Computing (by Research)
- Authors: Subburaj, Srikanth
- Date: 2012
- Type: Text , Thesis , Masters
- Full Text:
- Description: The term “green IT” is defined as “Optimal use of Information and Communication Technology (ICT) for managing the environment sustainability of enterprise operations throughout their life cycles.” Its objective is to create a positive impact on environment through net lowered emissions. The heart of green IT (Information Technology) is the concept of “Environmental Sustainability,” its initiatives are multifaceted to support enterprises “business as usual model,” in low-carbon economy. Many green IT initiatives have provided short–term financial benefits, based on which organizations are now focusing on increased energy efficiency. Since many of these approaches provide less financials benefits’ which are harder to implement but provides improved environmental performance. So the financial benefits and environmental performance should be the multiple objectives in the green IT study.
- Description: Master of Computing (by Research)
Student acquisition of social skills through teacher modelling
- Authors: Tocknell, Carol
- Date: 2015
- Type: Text , Thesis
- Full Text:
- Description: Doctor of Philosophy
- Description: Using the perspective of school leaders, teachers and students, this research questions whether Victorian Government schools provide opportunities, in the form of explicit teacher modelling, for the development of student social skills and social competence. The essence of this study was to determine whether students feel they can thrive at school and whether they believe there are opportunities for them to succeed. Two Victorian Government secondary schools were involved in the study and eight male and ten female students aged 12–14 years volunteered to be interviewed over the course of their first year in secondary school. Teachers and principals at the schools were also interviewed. The research found that students do not identify teachers as role models for social skills. However, school leaders believe that teachers are behaviour role models for students. Students and teachers were found to place value on positive teacher–student relationships. The variance of social-skills instruction in secondary schools and classrooms points to significant differences between intent and practice of such instruction within schools. It seems there are considerable differences in the perception of the Australian Curriculum‘s Personal and social capability domain within schools and between individual teachers‘ classrooms. This limits opportunity for students to develop social skills, as there appears to be no strategic whole-school social–emotional curriculum plan, few resources to implement such plans and no professional development for teachers in this area. This research describes the opportunities that are available for secondary students at Victorian Government schools to develop social skills and social competence. The research contributes to, and informs teacher pedagogy and contributes to knowledge on the value placed on social competence by students, teachers and school leaders. It also provides insight into the importance that students place on teacher interpersonal skills, the subsequent engagement of students with teachers and their sense of connectedness and belonging to their school.
- Authors: Tocknell, Carol
- Date: 2015
- Type: Text , Thesis
- Full Text:
- Description: Doctor of Philosophy
- Description: Using the perspective of school leaders, teachers and students, this research questions whether Victorian Government schools provide opportunities, in the form of explicit teacher modelling, for the development of student social skills and social competence. The essence of this study was to determine whether students feel they can thrive at school and whether they believe there are opportunities for them to succeed. Two Victorian Government secondary schools were involved in the study and eight male and ten female students aged 12–14 years volunteered to be interviewed over the course of their first year in secondary school. Teachers and principals at the schools were also interviewed. The research found that students do not identify teachers as role models for social skills. However, school leaders believe that teachers are behaviour role models for students. Students and teachers were found to place value on positive teacher–student relationships. The variance of social-skills instruction in secondary schools and classrooms points to significant differences between intent and practice of such instruction within schools. It seems there are considerable differences in the perception of the Australian Curriculum‘s Personal and social capability domain within schools and between individual teachers‘ classrooms. This limits opportunity for students to develop social skills, as there appears to be no strategic whole-school social–emotional curriculum plan, few resources to implement such plans and no professional development for teachers in this area. This research describes the opportunities that are available for secondary students at Victorian Government schools to develop social skills and social competence. The research contributes to, and informs teacher pedagogy and contributes to knowledge on the value placed on social competence by students, teachers and school leaders. It also provides insight into the importance that students place on teacher interpersonal skills, the subsequent engagement of students with teachers and their sense of connectedness and belonging to their school.
Hyperbolic smoothing in nonsmooth optimization and applications
- Authors: Al Nuaimat, Alia
- Date: 2014
- Type: Text , Thesis , PhD
- Full Text:
- Description: Nonsmooth nonconvex optimization problems arise in many applications including economics, business and data mining. In these applications objective functions are not necessarily differentiable or convex. Many algorithms have been proposed over the past three decades to solve such problems. In spite of the significant growth in this field, the development of efficient algorithms for solving this kind of problem is still a challenging task. The subgradient method is one of the simplest methods developed for solving these problems. Its convergence was proved only for convex objective functions. This method does not involve any subproblems, neither for finding search directions nor for computation of step lengths, which are fixed ahead of time. Bundle methods and their various modifications are among the most efficient methods for solving nonsmooth optimization problems. These methods involve a quadratic programming subproblem to find search directions. The size of the subproblem may increase significantly with the number of variables, which makes the bundle-type methods unsuitable for large scale nonsmooth optimization problems. The implementation of bundle-type methods, which require the use of the quadratic programming solvers, is not as easy as the implementation of the subgradient methods. Therefore it is beneficial to develop algorithms for nonsmooth nonconvex optimization which are easy to implement and more efficient than the subgradient methods. In this thesis, we develop two new algorithms for solving nonsmooth nonconvex optimization problems based on the use of the hyperbolic smoothing technique and apply them to solve the pumping cost minimization problem in water distribution. Both algorithms use smoothing techniques. The first algorithm is designed for solving finite minimax problems. In order to apply the hyperbolic smoothing we reformulate the objective function in the minimax problem and study the relationship between the original minimax and reformulated problems. We also study the main properties of the hyperbolic smoothing function. Based on these results an algorithm for solving the finite minimax problem is proposed and this algorithm is implemented in GAMS. We present preliminary results of numerical experiments with well-known nonsmooth optimization test problems. We also compare the proposed algorithm with the algorithm that uses the exponential smoothing function as well as with the algorithm based on nonlinear programming reformulation of the finite minimax problem. The second nonsmooth optimization algorithm we developed was used to demonstrate how smooth optimization methods can be applied to solve general nonsmooth (nonconvex) optimization problems. In order to do so we compute subgradients from some neighborhood of the current point and define a system of linear inequalities using these subgradients. Search directions are computed by solving this system. This system is solved by reducing it to the minimization of the convex piecewise linear function over the unit ball. Then the hyperbolic smoothing function is applied to approximate this minimization problem by a sequence of smooth problems which are solved by smooth optimization methods. Such an approach allows one to apply powerful smooth optimization algorithms for solving nonsmooth optimization problems and extend smoothing techniques for solving general nonsmooth nonconvex optimization problems. The convergence of the algorithm based on this approach is studied. The proposed algorithm was implemented in Fortran 95. Preliminary results of numerical experiments are reported and the proposed algorithm is compared with an other five nonsmooth optimization algorithms. We also implement the algorithm in GAMS and compare it with GAMS solvers using results of numerical experiments.
- Description: Doctor of Philosophy
- Authors: Al Nuaimat, Alia
- Date: 2014
- Type: Text , Thesis , PhD
- Full Text:
- Description: Nonsmooth nonconvex optimization problems arise in many applications including economics, business and data mining. In these applications objective functions are not necessarily differentiable or convex. Many algorithms have been proposed over the past three decades to solve such problems. In spite of the significant growth in this field, the development of efficient algorithms for solving this kind of problem is still a challenging task. The subgradient method is one of the simplest methods developed for solving these problems. Its convergence was proved only for convex objective functions. This method does not involve any subproblems, neither for finding search directions nor for computation of step lengths, which are fixed ahead of time. Bundle methods and their various modifications are among the most efficient methods for solving nonsmooth optimization problems. These methods involve a quadratic programming subproblem to find search directions. The size of the subproblem may increase significantly with the number of variables, which makes the bundle-type methods unsuitable for large scale nonsmooth optimization problems. The implementation of bundle-type methods, which require the use of the quadratic programming solvers, is not as easy as the implementation of the subgradient methods. Therefore it is beneficial to develop algorithms for nonsmooth nonconvex optimization which are easy to implement and more efficient than the subgradient methods. In this thesis, we develop two new algorithms for solving nonsmooth nonconvex optimization problems based on the use of the hyperbolic smoothing technique and apply them to solve the pumping cost minimization problem in water distribution. Both algorithms use smoothing techniques. The first algorithm is designed for solving finite minimax problems. In order to apply the hyperbolic smoothing we reformulate the objective function in the minimax problem and study the relationship between the original minimax and reformulated problems. We also study the main properties of the hyperbolic smoothing function. Based on these results an algorithm for solving the finite minimax problem is proposed and this algorithm is implemented in GAMS. We present preliminary results of numerical experiments with well-known nonsmooth optimization test problems. We also compare the proposed algorithm with the algorithm that uses the exponential smoothing function as well as with the algorithm based on nonlinear programming reformulation of the finite minimax problem. The second nonsmooth optimization algorithm we developed was used to demonstrate how smooth optimization methods can be applied to solve general nonsmooth (nonconvex) optimization problems. In order to do so we compute subgradients from some neighborhood of the current point and define a system of linear inequalities using these subgradients. Search directions are computed by solving this system. This system is solved by reducing it to the minimization of the convex piecewise linear function over the unit ball. Then the hyperbolic smoothing function is applied to approximate this minimization problem by a sequence of smooth problems which are solved by smooth optimization methods. Such an approach allows one to apply powerful smooth optimization algorithms for solving nonsmooth optimization problems and extend smoothing techniques for solving general nonsmooth nonconvex optimization problems. The convergence of the algorithm based on this approach is studied. The proposed algorithm was implemented in Fortran 95. Preliminary results of numerical experiments are reported and the proposed algorithm is compared with an other five nonsmooth optimization algorithms. We also implement the algorithm in GAMS and compare it with GAMS solvers using results of numerical experiments.
- Description: Doctor of Philosophy
The contribution of poker machines to reduced community wellbeing : A pre and post study
- Authors: Bell, Diana
- Date: 2018
- Type: Text , Thesis , Masters
- Full Text:
- Description: This thesis reports on a study of the impacts of poker machines on community wellbeing, using a pre and post survey method. The study used a variety of indicators to test community wellbeing and gambling attitudes and behaviours before and after a hotel venue with 40 poker machines opened in a new suburb in the designated growth area of Melbourne’s northern fringe. There was a higher proportion of respondents who met the criteria for ‘problem gambling’ after the poker machines were installed, compared to before, particularly when considered as a proportion of people who gambled on poker machines (5.3% compared to 3.6%). A proportion of respondents reported reduced levels of personal happiness, contentment and wellbeing as a result of the introduction of poker machines (16.5%, 12.3% and 16.1% respectively) and 41.5% reported there had been a detrimental impact on the community, in terms of social character. Mean scores on sense of community indexes and social cohesion showed a small decline in the post sample on every measure. Overall, the community reported reduced wellbeing on all measures after the introduction of poker machines. The significance of this study is that measures of community wellbeing and attitudes towards poker machines were measured before their introduction so that this baseline data could be compared with reported wellbeing 18 months after their installation within the suburban area. The substantial proportion of respondents who reported detrimental impact on social character, along with many negatively expressed opinions of poker machines, and a higher rate of problem gambling provide support for the notion that the introduction of poker machines at least contributed to the reduction in community wellbeing. This research provides some suggestions for the use of indicators for measuring the impact of poker machines on community wellbeing.
- Description: Masters by Research
- Authors: Bell, Diana
- Date: 2018
- Type: Text , Thesis , Masters
- Full Text:
- Description: This thesis reports on a study of the impacts of poker machines on community wellbeing, using a pre and post survey method. The study used a variety of indicators to test community wellbeing and gambling attitudes and behaviours before and after a hotel venue with 40 poker machines opened in a new suburb in the designated growth area of Melbourne’s northern fringe. There was a higher proportion of respondents who met the criteria for ‘problem gambling’ after the poker machines were installed, compared to before, particularly when considered as a proportion of people who gambled on poker machines (5.3% compared to 3.6%). A proportion of respondents reported reduced levels of personal happiness, contentment and wellbeing as a result of the introduction of poker machines (16.5%, 12.3% and 16.1% respectively) and 41.5% reported there had been a detrimental impact on the community, in terms of social character. Mean scores on sense of community indexes and social cohesion showed a small decline in the post sample on every measure. Overall, the community reported reduced wellbeing on all measures after the introduction of poker machines. The significance of this study is that measures of community wellbeing and attitudes towards poker machines were measured before their introduction so that this baseline data could be compared with reported wellbeing 18 months after their installation within the suburban area. The substantial proportion of respondents who reported detrimental impact on social character, along with many negatively expressed opinions of poker machines, and a higher rate of problem gambling provide support for the notion that the introduction of poker machines at least contributed to the reduction in community wellbeing. This research provides some suggestions for the use of indicators for measuring the impact of poker machines on community wellbeing.
- Description: Masters by Research
The determinants and development of fast bowling performance in cricket
- Authors: Feros, Simon
- Date: 2015
- Type: Text , Thesis , PhD
- Full Text:
- Description: This thesis sought to reveal the physical and kinematic determinants of pace bowling performance. After drawing on these determinants, a secondary aim was to investigate whether pace bowling performance could be enhanced with chronic resistance training and warm-up strategies. However, before the physical and kinematic determinants of pace bowling performance could be identified, and the effects of two training interventions and warm-ups on pace bowling performance, a new pace bowling test was created, and the test-retest reliability of its performance and kinematic measures were evaluated. Knowledge of a variables’ test-retest reliability is important for interpreting the validity of correlations, but also for the determination of a meaningful change following a training intervention. Only one published study to date has explored the test-retest reliability of a pace bowling assessment, and this test only measured bowling accuracy (1). Previous research has not comprehensively examined the relationships between physical qualities and pace bowling performance. Several important physical qualities (e.g., power, speed-acceleration, flexibility, repeat-sprint ability) have been excluded in correlational research, which may be crucial for optimal pace bowling performance. Furthermore, there is only one published training intervention study on pace bowling research (2). Consequently there is scant evidence for coaches to design training programs proven to enhance pace bowling performance. Baseball pitching studies have trialled the effects of heavy-ball throwing in the warm-up on subsequent throwing velocity and accuracy, but this approach has not been studied in cricket pace bowling, especially after several weeks of training. Therefore, four studies were conducted in this PhD project to address these deficiencies in the literature. The purpose of Study 1 (Chapter 3) was to ascertain the test-retest reliability of bowling performance measures (i.e., bowling speed, bowling accuracy, consistency of bowling speed, and consistency of bowling accuracy) and selected bowling kinematics (i.e., approach speed, step length, step-length phase duration, power phase duration, and knee extension angle at front-foot contact and at ball release) in a novel eight-over test, and for the first four overs of this test. The intraclass correlation coefficient (ICC), standard error of measurement (SEM), and coefficient of variation (CV) were used as measures of test-retest reliability (3). Following a three week familiarisation period of bowling, 13 participants completed a novel eight-over bowling test on two separate days with 4–7 days apart. The most reliable performance measures in the bowling test were peak bowling speed (ICC = 0.948–0.975, CV = 1.3–1.9%) and mean bowling speed (ICC = 0.981–0.987, CV = 1.0–1.3%). Perceived effort was partially reliable (ICC = 0.650– 0.659, CV = 3.8–3.9%). However, mean bowling accuracy (ICC = 0.491–0.685, CV = 12.5–16.8%) and consistency of bowling accuracy failed to meet the pre-set standard for acceptable reliability (ICC = 0.434–0.454, CV = 15.3–19.3%). All bowling kinematic variables except approach speed exhibited acceptable reliability (i.e., ICC > 0.8, CV < 10%). The first four overs of the bowling test exhibited slightly poorer test-retest reliability for all measures, compared to the entire eight-over test. There were no systematic biases (i.e., p > 0.05) detected with all variables between bowling tests, indicating there was no learning or fatigue effects. The smallest worthwhile change was established for all bowling performance and kinematic variables, by multiplying the SEM by 1.5 (4). It is recommended that the eight-over pace bowling test be used as a more comprehensive measure of consistency of bowling speed and consistency of bowling accuracy, as bowlers are more likely to be fatigued. However, if coaches seek to assess pace bowlers in shorter time, delimiting the test to the first four overs is recommended. Both versions of the pace bowling test are only capable of reliably measuring bowling performance outcomes such as peak and mean bowling speed, and perceived effort. The second study of this PhD project examined the relationships between selected physical qualities, bowling kinematics, and bowling performance measures. Another purpose of this novel study was to determine if delivery instructions (i.e., maximal-effort, match-intensity, slower-ball) influenced the strength of the relationships between physical qualities and bowling performance measures. Given that there were three delivery instructions in the bowling test, an objective of this study was to explore the relationship between bowling speed and bowling accuracy (i.e., speed-accuracy trade-off). Thirty-one participants completed an eight-over bowling test in the first session, and a series of physical tests, spread over two separate sessions. Each session was separated by four to seven days. Mean bowling speed (of all pooled deliveries) was significantly correlated to 1-RM pull-up strength (rs [24] = 0.55, p = 0.01) and 20-m sprint time (rs [30] = -0.37, p = 0.04), but the correlations marginally increased as delivery effort increased (i.e., maximal-effort ball). Greater hamstring flexibility was associated with a better consistency of bowling speed, but only for a match-intensity delivery (rs [29] = -0.49, p = 0.01). Repeat-sprint ability (i.e., percent decrement on 10 × 20-m sprints, on every 20 s) displayed a stronger correlation to consistency of bowling speed (rs [21] = -0.42, p = 0.06) than for mean bowling speed (rs [21] = 0.15, p = 0.53). Bench press strength was moderately related to bowling accuracy for a maximal-effort delivery (rs [26] = -0.42, p = 0.03), with weaker but non-significant (p > 0.05) correlations for match-intensity and slower-ball deliveries. Bowling accuracy was also significantly related to peak concentric countermovement jump power (rs [28] = -0.41, p = 0.03) and mean peak concentric countermovement jump power (rs [27] = -0.45, p = 0.02), with both physical qualities displaying stronger correlations as delivery effort increased. Greater reactive strength was negatively associated with mean bowling accuracy (rs [30] = 0.38, p = 0.04) and consistency of bowling accuracy (rs [30] = 0.43, p = 0.02) for maximal-effort deliveries only. Faster bowling speeds were correlated to a longer step length (rs [31] = 0.51, p < 0.01) and quicker power phase duration (rs [31] = -0.45, p = 0.01). A better consistency of bowling accuracy was associated with a faster approach speed (rs [31] = -0.36, p = 0.05) and greater knee flexion angle at ball release (rs [27] = -0.42, p = 0.03). No speedaccuracy trade-off was observed for the group (rs [31] = -0.28, p = 0.12), indicating that most bowlers could be instructed to train at maximal-effort without compromising bowling accuracy. Pull-up strength training and speed-acceleration training were chosen for the “evidence-based” training program (Study 3). Heavy-ball bowling was also considered as part of the evidence-based training program, as it is a specific form of training used previously, and because there was a shortage of significant relationships (p < 0.05) between physical qualities and bowling performance measures in Study 2. The third investigation of this PhD project compared the effects of an eight-week evidence-based training program or normal training program (not a control group) on pace bowling performance, approach speed, speed-acceleration, and pull-up strength. Participants were matched for bowling speed and then randomly split into two training groups, with six participants in each group. After an initial two-week familiarisation period of bowling training, sprint training, and pull-up training, participants completed two training sessions per week, and were tested before and after the training intervention. Testing comprised the four-over pace bowling test (Study 1), 20-m sprint test (Study 2), and 1-RM pull-up test (Study 2). In training, the volume of bowling and sprinting was constant between both groups; the only differences were that the evidence-based training group bowled with heavy balls (250 g and 300 g) as well as a regular ball (156 g), sprinted with a weighted-vest (15% and 20% body mass) and without a weighted-vest, and performed pull-up training. Participants were instructed to deliver each ball with maximal effort in training, as no speed-accuracy trade-off was observed for the sample in Study 2. The evidence-based training group bowled with poorer accuracy and consistency of accuracy, with only a small improvement in peak and mean bowling speed. Heavy-ball bowling may have had a negative transfer to regular-ball bowling. Although speculative, a longer evidence-based program may have significantly enhanced bowling speed. Coaches could use both training programs to develop performance but should be aware that bowling accuracy may suffer with the evidence-based program. The evidence-based training group displayed slower 20-m sprint times following training (0.08 ± 0.05 s). However, the normal training group was also slower (0.10 ± 0.09 s), indicating the potential for speed-acceleration improvement is compromised if speed training is performed immediately after bowling training; most likely due to residual fatigue. Consequently it is recommended that speed-acceleration training be conducted when bowlers are not fatigued, in a separate session, or at the beginning of a session. The evidence-based training group improved their 1-RM pull-up strength by 5.8 ± 6.8 kg (d = 0.68), compared to the normal training group of 0.2 ± 1.7 kg (d = 0.01). The difference between training groups is due to the fact that the normal training group were not prescribed pull-up training. As many participants could not complete the pull-up exercise due to insufficient strength, the dumbbell pullover may be a suitable alternative that is more specific to the motion of the bowling arm (i.e., extended arm). The fourth study of this PhD project explored the acute effects of a heavy-ball bowling warm-up on pace bowling performance, and determined if these acute effects could be enhanced or negated following an evidence-based training program. This study involved the same participants who completed the evidence-based training program in Study 3. These participants were required to perform two different bowling warm-ups (heavy-ball or regular-ball) in pre and post-test period, followed by the four-over pace bowling test (Study 1). In pre-test period, bowling accuracy was 8.8 ± 7.4 cm worse for the heavy-ball warm-up compared to the regular-ball warm-up (d = 1.19). In post-test period however, bowling accuracy was 5.5 ± 6.4 cm better in the heavy-ball warm-up compared to the regular-ball warm-up (d = -0.90). A similar trend was observed for consistency of bowling accuracy. These findings indicate that pace bowlers adapt to heavy-ball bowling, and bowl more accurately with a regular ball if they warm-up with a heavy ball first (but only after eight weeks of heavy-ball training). Coaches could employ a heavy-ball warm-up prior to training or a match, but only after eight weeks of evidence based training. It is hypothesised that a less biomechanically similar exercise to the pace bowling motion such as resisted push-ups / bench press throws could be more effective in eliciting potentiation by activating higher order motor units without negatively transferring to bowling performance. From the studies presented in this thesis, it is concluded that peak and mean bowling speed are the most reliable bowling performance measures, and all kinematic variables apart from approach speed possess excellent reliability. Furthermore, 1-RM pull-up strength and 20-m speed are significantly correlated to bowling speed. An evidence-based training program can develop peak and mean bowling speed, but the cost to bowling accuracy and consistency of bowling accuracy does not make this training program worthwhile in enhancing pace bowling performance. A heavy-ball warm-up impairs bowling accuracy and consistency of bowling accuracy compared to the regular-ball warm-up, but only prior to training with the heavier balls. Pace bowlers adapt to heavyball bowling after eight weeks of training, but must use the heavy balls in the warm-up to bowl more accurately with a regular ball, otherwise pace bowling performance is below optimal.
- Description: Doctor of Philosophy
- Authors: Feros, Simon
- Date: 2015
- Type: Text , Thesis , PhD
- Full Text:
- Description: This thesis sought to reveal the physical and kinematic determinants of pace bowling performance. After drawing on these determinants, a secondary aim was to investigate whether pace bowling performance could be enhanced with chronic resistance training and warm-up strategies. However, before the physical and kinematic determinants of pace bowling performance could be identified, and the effects of two training interventions and warm-ups on pace bowling performance, a new pace bowling test was created, and the test-retest reliability of its performance and kinematic measures were evaluated. Knowledge of a variables’ test-retest reliability is important for interpreting the validity of correlations, but also for the determination of a meaningful change following a training intervention. Only one published study to date has explored the test-retest reliability of a pace bowling assessment, and this test only measured bowling accuracy (1). Previous research has not comprehensively examined the relationships between physical qualities and pace bowling performance. Several important physical qualities (e.g., power, speed-acceleration, flexibility, repeat-sprint ability) have been excluded in correlational research, which may be crucial for optimal pace bowling performance. Furthermore, there is only one published training intervention study on pace bowling research (2). Consequently there is scant evidence for coaches to design training programs proven to enhance pace bowling performance. Baseball pitching studies have trialled the effects of heavy-ball throwing in the warm-up on subsequent throwing velocity and accuracy, but this approach has not been studied in cricket pace bowling, especially after several weeks of training. Therefore, four studies were conducted in this PhD project to address these deficiencies in the literature. The purpose of Study 1 (Chapter 3) was to ascertain the test-retest reliability of bowling performance measures (i.e., bowling speed, bowling accuracy, consistency of bowling speed, and consistency of bowling accuracy) and selected bowling kinematics (i.e., approach speed, step length, step-length phase duration, power phase duration, and knee extension angle at front-foot contact and at ball release) in a novel eight-over test, and for the first four overs of this test. The intraclass correlation coefficient (ICC), standard error of measurement (SEM), and coefficient of variation (CV) were used as measures of test-retest reliability (3). Following a three week familiarisation period of bowling, 13 participants completed a novel eight-over bowling test on two separate days with 4–7 days apart. The most reliable performance measures in the bowling test were peak bowling speed (ICC = 0.948–0.975, CV = 1.3–1.9%) and mean bowling speed (ICC = 0.981–0.987, CV = 1.0–1.3%). Perceived effort was partially reliable (ICC = 0.650– 0.659, CV = 3.8–3.9%). However, mean bowling accuracy (ICC = 0.491–0.685, CV = 12.5–16.8%) and consistency of bowling accuracy failed to meet the pre-set standard for acceptable reliability (ICC = 0.434–0.454, CV = 15.3–19.3%). All bowling kinematic variables except approach speed exhibited acceptable reliability (i.e., ICC > 0.8, CV < 10%). The first four overs of the bowling test exhibited slightly poorer test-retest reliability for all measures, compared to the entire eight-over test. There were no systematic biases (i.e., p > 0.05) detected with all variables between bowling tests, indicating there was no learning or fatigue effects. The smallest worthwhile change was established for all bowling performance and kinematic variables, by multiplying the SEM by 1.5 (4). It is recommended that the eight-over pace bowling test be used as a more comprehensive measure of consistency of bowling speed and consistency of bowling accuracy, as bowlers are more likely to be fatigued. However, if coaches seek to assess pace bowlers in shorter time, delimiting the test to the first four overs is recommended. Both versions of the pace bowling test are only capable of reliably measuring bowling performance outcomes such as peak and mean bowling speed, and perceived effort. The second study of this PhD project examined the relationships between selected physical qualities, bowling kinematics, and bowling performance measures. Another purpose of this novel study was to determine if delivery instructions (i.e., maximal-effort, match-intensity, slower-ball) influenced the strength of the relationships between physical qualities and bowling performance measures. Given that there were three delivery instructions in the bowling test, an objective of this study was to explore the relationship between bowling speed and bowling accuracy (i.e., speed-accuracy trade-off). Thirty-one participants completed an eight-over bowling test in the first session, and a series of physical tests, spread over two separate sessions. Each session was separated by four to seven days. Mean bowling speed (of all pooled deliveries) was significantly correlated to 1-RM pull-up strength (rs [24] = 0.55, p = 0.01) and 20-m sprint time (rs [30] = -0.37, p = 0.04), but the correlations marginally increased as delivery effort increased (i.e., maximal-effort ball). Greater hamstring flexibility was associated with a better consistency of bowling speed, but only for a match-intensity delivery (rs [29] = -0.49, p = 0.01). Repeat-sprint ability (i.e., percent decrement on 10 × 20-m sprints, on every 20 s) displayed a stronger correlation to consistency of bowling speed (rs [21] = -0.42, p = 0.06) than for mean bowling speed (rs [21] = 0.15, p = 0.53). Bench press strength was moderately related to bowling accuracy for a maximal-effort delivery (rs [26] = -0.42, p = 0.03), with weaker but non-significant (p > 0.05) correlations for match-intensity and slower-ball deliveries. Bowling accuracy was also significantly related to peak concentric countermovement jump power (rs [28] = -0.41, p = 0.03) and mean peak concentric countermovement jump power (rs [27] = -0.45, p = 0.02), with both physical qualities displaying stronger correlations as delivery effort increased. Greater reactive strength was negatively associated with mean bowling accuracy (rs [30] = 0.38, p = 0.04) and consistency of bowling accuracy (rs [30] = 0.43, p = 0.02) for maximal-effort deliveries only. Faster bowling speeds were correlated to a longer step length (rs [31] = 0.51, p < 0.01) and quicker power phase duration (rs [31] = -0.45, p = 0.01). A better consistency of bowling accuracy was associated with a faster approach speed (rs [31] = -0.36, p = 0.05) and greater knee flexion angle at ball release (rs [27] = -0.42, p = 0.03). No speedaccuracy trade-off was observed for the group (rs [31] = -0.28, p = 0.12), indicating that most bowlers could be instructed to train at maximal-effort without compromising bowling accuracy. Pull-up strength training and speed-acceleration training were chosen for the “evidence-based” training program (Study 3). Heavy-ball bowling was also considered as part of the evidence-based training program, as it is a specific form of training used previously, and because there was a shortage of significant relationships (p < 0.05) between physical qualities and bowling performance measures in Study 2. The third investigation of this PhD project compared the effects of an eight-week evidence-based training program or normal training program (not a control group) on pace bowling performance, approach speed, speed-acceleration, and pull-up strength. Participants were matched for bowling speed and then randomly split into two training groups, with six participants in each group. After an initial two-week familiarisation period of bowling training, sprint training, and pull-up training, participants completed two training sessions per week, and were tested before and after the training intervention. Testing comprised the four-over pace bowling test (Study 1), 20-m sprint test (Study 2), and 1-RM pull-up test (Study 2). In training, the volume of bowling and sprinting was constant between both groups; the only differences were that the evidence-based training group bowled with heavy balls (250 g and 300 g) as well as a regular ball (156 g), sprinted with a weighted-vest (15% and 20% body mass) and without a weighted-vest, and performed pull-up training. Participants were instructed to deliver each ball with maximal effort in training, as no speed-accuracy trade-off was observed for the sample in Study 2. The evidence-based training group bowled with poorer accuracy and consistency of accuracy, with only a small improvement in peak and mean bowling speed. Heavy-ball bowling may have had a negative transfer to regular-ball bowling. Although speculative, a longer evidence-based program may have significantly enhanced bowling speed. Coaches could use both training programs to develop performance but should be aware that bowling accuracy may suffer with the evidence-based program. The evidence-based training group displayed slower 20-m sprint times following training (0.08 ± 0.05 s). However, the normal training group was also slower (0.10 ± 0.09 s), indicating the potential for speed-acceleration improvement is compromised if speed training is performed immediately after bowling training; most likely due to residual fatigue. Consequently it is recommended that speed-acceleration training be conducted when bowlers are not fatigued, in a separate session, or at the beginning of a session. The evidence-based training group improved their 1-RM pull-up strength by 5.8 ± 6.8 kg (d = 0.68), compared to the normal training group of 0.2 ± 1.7 kg (d = 0.01). The difference between training groups is due to the fact that the normal training group were not prescribed pull-up training. As many participants could not complete the pull-up exercise due to insufficient strength, the dumbbell pullover may be a suitable alternative that is more specific to the motion of the bowling arm (i.e., extended arm). The fourth study of this PhD project explored the acute effects of a heavy-ball bowling warm-up on pace bowling performance, and determined if these acute effects could be enhanced or negated following an evidence-based training program. This study involved the same participants who completed the evidence-based training program in Study 3. These participants were required to perform two different bowling warm-ups (heavy-ball or regular-ball) in pre and post-test period, followed by the four-over pace bowling test (Study 1). In pre-test period, bowling accuracy was 8.8 ± 7.4 cm worse for the heavy-ball warm-up compared to the regular-ball warm-up (d = 1.19). In post-test period however, bowling accuracy was 5.5 ± 6.4 cm better in the heavy-ball warm-up compared to the regular-ball warm-up (d = -0.90). A similar trend was observed for consistency of bowling accuracy. These findings indicate that pace bowlers adapt to heavy-ball bowling, and bowl more accurately with a regular ball if they warm-up with a heavy ball first (but only after eight weeks of heavy-ball training). Coaches could employ a heavy-ball warm-up prior to training or a match, but only after eight weeks of evidence based training. It is hypothesised that a less biomechanically similar exercise to the pace bowling motion such as resisted push-ups / bench press throws could be more effective in eliciting potentiation by activating higher order motor units without negatively transferring to bowling performance. From the studies presented in this thesis, it is concluded that peak and mean bowling speed are the most reliable bowling performance measures, and all kinematic variables apart from approach speed possess excellent reliability. Furthermore, 1-RM pull-up strength and 20-m speed are significantly correlated to bowling speed. An evidence-based training program can develop peak and mean bowling speed, but the cost to bowling accuracy and consistency of bowling accuracy does not make this training program worthwhile in enhancing pace bowling performance. A heavy-ball warm-up impairs bowling accuracy and consistency of bowling accuracy compared to the regular-ball warm-up, but only prior to training with the heavier balls. Pace bowlers adapt to heavyball bowling after eight weeks of training, but must use the heavy balls in the warm-up to bowl more accurately with a regular ball, otherwise pace bowling performance is below optimal.
- Description: Doctor of Philosophy
Effective and efficient kernel-based image representations for classification and retrieval
- Authors: Karmakar, Priyabrata
- Date: 2018
- Type: Text , Thesis , PhD
- Full Text:
- Description: Image representation is a challenging task. In particular, in order to obtain better performances in different image processing applications such as video surveillance, autonomous driving, crime scene detection and automatic inspection, effective and efficient image representation is a fundamental need. The performance of these applications usually depends on how accurately images are classified into their corresponding groups or how precisely relevant images are retrieved from a database based on a query. Accuracy in image classification and precision in image retrieval depend on the effectiveness of image representation. Existing image representation methods have some limitations. For example, spatial pyramid matching, which is a popular method incorporating spatial information in image-level representation, has not been fully studied to date. In addition, the strengths of pyramid match kernel and spatial pyramid matching are not combined for better image matching. Kernel descriptors based on gradient, colour and shape overcome the limitations of histogram-based descriptors, but suffer from information loss, noise effects and high computational complexity. Furthermore, the combined performance of kernel descriptors has limitations related to computational complexity, higher dimensionality and lower effectiveness. Moreover, the potential of a global texture descriptor which is based on human visual perception has not been fully explored to date. Therefore, in this research project, kernel-based effective and efficient image representation methods are proposed to address the above limitations. An enhancement is made to spatial pyramid matching in terms of improved rotation invariance. This is done by investigating different partitioning schemes suitable to achieve rotation-invariant image representation and the proposal of a weight function for appropriate level contribution in image matching. In addition, the strengths of pyramid match kernel and spatial pyramid are combined to enhance matching accuracy between images. The existing kernel descriptors are modified and improved to achieve greater effectiveness, minimum noise effects, less dimensionality and lower computational complexity. A novel fusion approach is also proposed to combine the information related to all pixel attributes, before the descriptor extraction stage. Existing kernel descriptors are based only on gradient, colour and shape information. In this research project, a texture-based kernel descriptor is proposed by modifying an existing popular global texture descriptor. Finally, all the contributions are evaluated in an integrated system. The performances of the proposed methods are qualitatively and quantitatively evaluated on two to four different publicly available image databases. The experimental results show that the proposed methods are more effective and efficient in image representation than existing benchmark methods.
- Description: Doctor of Philosophy
- Authors: Karmakar, Priyabrata
- Date: 2018
- Type: Text , Thesis , PhD
- Full Text:
- Description: Image representation is a challenging task. In particular, in order to obtain better performances in different image processing applications such as video surveillance, autonomous driving, crime scene detection and automatic inspection, effective and efficient image representation is a fundamental need. The performance of these applications usually depends on how accurately images are classified into their corresponding groups or how precisely relevant images are retrieved from a database based on a query. Accuracy in image classification and precision in image retrieval depend on the effectiveness of image representation. Existing image representation methods have some limitations. For example, spatial pyramid matching, which is a popular method incorporating spatial information in image-level representation, has not been fully studied to date. In addition, the strengths of pyramid match kernel and spatial pyramid matching are not combined for better image matching. Kernel descriptors based on gradient, colour and shape overcome the limitations of histogram-based descriptors, but suffer from information loss, noise effects and high computational complexity. Furthermore, the combined performance of kernel descriptors has limitations related to computational complexity, higher dimensionality and lower effectiveness. Moreover, the potential of a global texture descriptor which is based on human visual perception has not been fully explored to date. Therefore, in this research project, kernel-based effective and efficient image representation methods are proposed to address the above limitations. An enhancement is made to spatial pyramid matching in terms of improved rotation invariance. This is done by investigating different partitioning schemes suitable to achieve rotation-invariant image representation and the proposal of a weight function for appropriate level contribution in image matching. In addition, the strengths of pyramid match kernel and spatial pyramid are combined to enhance matching accuracy between images. The existing kernel descriptors are modified and improved to achieve greater effectiveness, minimum noise effects, less dimensionality and lower computational complexity. A novel fusion approach is also proposed to combine the information related to all pixel attributes, before the descriptor extraction stage. Existing kernel descriptors are based only on gradient, colour and shape information. In this research project, a texture-based kernel descriptor is proposed by modifying an existing popular global texture descriptor. Finally, all the contributions are evaluated in an integrated system. The performances of the proposed methods are qualitatively and quantitatively evaluated on two to four different publicly available image databases. The experimental results show that the proposed methods are more effective and efficient in image representation than existing benchmark methods.
- Description: Doctor of Philosophy