An analysis of human engagement behaviour using descriptors from human feedback, eye tracking, and saliency modelling
- Authors: Podder, Pallab , Paul, Manoranjan , Debnath, Tanmoy , Murshed, Manzur
- Date: 2015
- Type: Text , Conference proceedings
- Relation: 2015 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2015, Adelaide, 23-25th Nov 2015 in Digital Image Computing: Techniques and Applications (DICTA), 2015 International Conference
- Relation: http://purl.org/au-research/grants/arc/DP130103670
- Full Text: false
- Reviewed:
- Description: In this paper an analysis of human engagement behaviour with video is presented based on real life experiments. An engagement model could be employed in classroom education, enhancing programming skills, reading etc. Two groups of people, independent of one another, watched eighteen video clips separately at different times. The first group's participants' eye gaze locations, right and left pupil sizes, and eye blinking patterns are recorded by a state of the art Tobii eye tracker. The second group of people who are video experts opined about the most significant attention points of the videos. A well-known bottom-up visual saliency model, Graph-Based Visual Saliency (GBVS), is also utilized to create salient points for the videos. Taking into consideration all the above mentioned descriptors the introduced behaviour analysis demonstrates the level of participants' concentration with the videos.
Predicting and controlling the dynamics of infectious diseases
- Authors: Evans, Robin , Mammadov, Musa
- Date: 2015
- Type: Text , Conference proceedings
- Relation: 54th IEEE Conference on Decision and Control, CDC 2015; Osaka, Japan; 15th-18th December 2015; Published in Proceedings of the IEEE Conference on Decision and Control; p. 5378-5383
- Full Text:
- Description: This paper introduces a new optimal control model to describe and control the dynamics of infectious diseases. In the present model, the average time to isolation (i.e. hospitalization) of infectious population is the main time-dependent parameter that defines the spread of infection. All the preventive measures aim to decrease the average time to isolation under given constraints. The model suggested allows one to generate a small number of possible future scenarios and to determine corresponding trajectories of infected population in different regions. Then, this information is used to find an optimal distribution of bed capabilities across countries/regions according to each scenario. © 2015 IEEE.
Predictive analytics for detecting sensor failure using autoregressive integrated moving average model
- Authors: Thiyagarajan, Karthick , Kodagoda, Sarath , Van Nguyen, Linh
- Date: 2017
- Type: Text , Conference proceedings
- Relation: 2017 12th IEEE Conference on Industrial Electronics and Applications (ICIEA); Siem Reap, Cambodia; 18-20 June 2017 p. 1926-1931
- Full Text: false
- Reviewed:
- Description: Sensors play a vital role in monitoring the important parameters of critical infrastructure. Failure of such sensors causes destabilization to the entire system. In this regard, this paper proposes a predictive analytics solution for detecting the failure of a sensor that measures surface temperature from an urban sewer. The proposed approach incorporates a forecasting technique based on the past time series of sparse data using an autoregressive integrated moving average (ARIMA) model. Based on the 95% forecast interval and continuity of faulty data, a criterion was set to detect anomalies and to issue a warning for sensor failure. The forecasted and faulty data were assumed Gaussian distributed. By using the probability density of the distribution, the mean and variance were computed for faulty data to examine the abnormality in the variance value of each day to detect the sensor failure. The experimental results on the sewer temperature data are appealing.
Gaussian mixture marginal distributions for modelling remaining metallic pipe wall thickness
- Authors: Nguyen, Linh , Miro, Jaime Valls , Shi, Lei , Vidal-Calleja, Teresa
- Date: 2019
- Type: Text , Conference proceedings
- Relation: 2019 IEEE International Conference on Cybernetics and Intelligent Systems (CIS) and IEEE Conference on Robotics, Automation and Mechatronics (RAM);Bangkok, Thailand; 18-20 November 2019 p. 257-262
- Full Text: false
- Reviewed:
- Description: Rapidly estimating the remaining wall thickness (RWT) is paramount for the non-destructive condition assessment evaluation of large critical metallic pipelines. A robotic vehicle with embedded magnetism-based sensors has been developed to traverse the inside of a pipeline and conduct inspections at the location of a break. However, its sensing speed is constrained by the magnetic principle of operation, thus slowing down the overall operation in seeking dense RWT mapping. To ameliorate this drawback, this work proposes the partial scanning of the pipe and then employing Gaussian Processes (GPs) to infer RWT at the unseen pipe sections. Since GP prediction assumes to have normally distributed input data - which does correspond with real RWT measurements - Gaussian mixture (GM) models are proven in this work as fitting marginal distributions to effectively capture the probability of any RWT value in the inspected data. The effectiveness of the proposed approach is validated from real-world data collected in collaboration with a water utility from a cast iron water main pipeline in Sydney, Australia.
Multi-classifier predictive maintenance strategy for a manufacturing plant
- Authors: Singh, Prashant , Agrawal, Sunil , Chakraborty, Ayon
- Date: 2021
- Type: Text , Conference proceedings
- Relation: 2021 International Conference on Maintenance and Intelligent Asset Management (ICMIAM), Ballarat, Australia, 12-15 December 2021 p. 1-4
- Full Text: false
- Reviewed:
- Description: Predictive Maintenance Management in an industry can play a pivotal role in asset management and revenue generation. This work proposes a data-driven-based multi classifier model for implementing predictive maintenance to simultaneously reduce the downtime and idle time of the machines in a manufacturing plant. A case study of the plant comprising of 100 machines has been done to identify the early prediction of failure, its nature, and the attributing cause. Gradient Boosting Tree Classifier and Random Forest Classifier machine learning algorithms have been used to develop the models for fault prediction. A comparative analysis of results obtained using these methods has also been done. Random Forest Classifier outperforms Gradient Boost tree classifier in all evaluation parameters - accuracy, precision and recall.
A unified model predictive voltage and current control for microgrids with distributed fuzzy cooperative secondary control
- Authors: Shan, Yinghao , Hu, Jiefeng , Chan, Ka , Islam, Syed
- Date: 2021
- Type: Text , Journal article
- Relation: IEEE Transactions on Industrial Informatics Vol. 17, no. 12 (DEC 2021), p. 8024-8034
- Full Text: false
- Reviewed:
- Description: A microgrid formed by a cluster of parallel distributed generation (DG) units is capable of operating in either islanded mode or grid-connected mode. Traditionally, by using model predictive control algorithms, these two operation modes can be achieved with two separate and different cost functions, which brings in control complexity and hence, compromises system reliability. In this article, a unified model predictive voltage and current control strategy is proposed for both islanded and grid-connected operations and their smooth transition. The cost function is kept unified with voltage and current taken into account without altering the control architecture. It can be used for high-quality voltage supply at the primary control level and for bidirectional power flow at the tertiary control level. In addition, by only using DGs' own and neighboring information, a distributed fuzzy cooperative algorithm is developed at the secondary layer to mitigate the voltage and frequency deviations inherent from the power droop. The fuzzy controller can optimize the secondary control coefficients for further voltage quality improvement. Comprehensive tests under various scenarios demonstrate the merits of the proposed control strategy over traditional methods.
Data-driven classifier for extreme outage prediction based on bayes decision theory
- Authors: Mohammadian, Mostafa , Aminifar, Farrokh , Amjady, Nima , Shahidehpour, Mohammad
- Date: 2021
- Type: Text , Journal article
- Relation: IEEE transactions on power systems Vol. 36, no. 6 (2021), p. 4906-4914
- Full Text: false
- Reviewed:
- Description: The growing concern over catastrophic weather events, mostly as a direct result of climate changes, has underscored the need for expanding traditional power system contingency analyses to handle the associated risks of extreme power outages. To enable power system operators to make timely decisions when facing extreme events, we explore in this paper the viability of a classifier which uses the machine learning approach based on the Bayes decision theory as a means of predicting power system component outages. However, owing to an excessively imbalanced and largely sparse power component outage datasets, the corresponding classifier learning is a challenging problem in the data mining community. In the proposed approach, we apply a resampling method to overcome the class imbalance problem. The proposed classifier provides an effective framework that not only minimizes outage prediction errors for power system components, but also considers the cost of each preventive action according to its implication in extreme events. The outcome of the proposed model can be used for introducing operation-oriented preventive measures that allow the rescheduling of generation resources for maximizing the power system resilience. The performance of the proposed classifier is examined through numerical simulations by utilizing the confusion matrix.
Nested bilevel optimization for dera operation strategy: A stochastic multiobjective igdt model with hybrid endogenous/exogenous scenarios
- Authors: Yazdaninejad, Mohsen , Amjady, Nima , Hatziargyriou, Nikos
- Date: 2021
- Type: Text , Journal article
- Relation: IEEE systems journal Vol. 15, no. 4 (2021), p. 5495-5506
- Full Text: false
- Reviewed:
- Description: An aggregation of distributed energy resources (DERs) can bring economic and technical benefits for the DER owners and system operator. However, the operation of DERs encounters various uncertainties, which can seriously impact the benefits of DER aggregation. This article presents a new operation optimization approach for an aggregator of DERs considering the unavailability of DERs (as discrete uncertainty sources) as well as forecast uncertainties of electricity prices, solar powers, and wind powers (as continuous uncertainty sources). The proposed approach for DER aggregator (DERA) operation optimization comprises stochastic multiobjective information-gap decision theory (IGDT) to model these discrete and continuous uncertain variables. Moreover, a hybrid endogenous/exogenous scenario generation method is incorporated into the proposed approach to enhance the efficiency of the stochastic programming part by producing decision-dependent scenario trees. The proposed approach is formulated as a nested bilevel optimization model. The proposed approach is compared with other DERA operation optimization models using an out-of-sample analysis method. The comparative results illustrate the superiority of the proposed stochastic multiobjective IGDT approach over various deterministic, stochastic, and IGDT methods. In addition, the high tractability of the proposed solution method is illustrated, while its linearization error for the stochastic multiobjective IGDT problem is well below 1%.
A new solar power prediction method based on feature clustering and hybrid-classification-regression forecasting
- Authors: Nejati, Maryam , Amjady, Nima
- Date: 2022
- Type: Text , Journal article
- Relation: IEEE transactions on sustainable energy Vol. 13, no. 2 (2022), p. 1188-1198
- Full Text: false
- Reviewed:
- Description: Solar generation systems are globally extending in terms of scale and number, which highlights the increasing importance of solar power forecast. In this paper, a day-ahead solar power prediction method is proposed including 1) a novel feature selecting/clustering approach based on relevancy and redundancy criteria and 2) an innovative hybrid-classification-regression forecasting engine. The proposed feature selecting/clustering approach filters out irrelevant features and partitions relevant features to two separate subsets to decrease the redundancy of features. Each of these two subsets is separately trained by one forecasting engine and the final solar power prediction of the proposed method is obtained by a relevancy-based combination of these two forecasts. The proposed forecasting engine classifies the historical data based on the learnability of its constituent regression models and assigns each class of training samples to one regression model. Each regression model predicts the outputs of the test samples that belong to its class. The effectiveness of the proposed solar power prediction method is illustrated by testing on two real-world solar farms.
Machine learning-based modelling for museum visitations prediction
- Authors: Yap, Norman , Gong, Mingwei , Naha, Ranesh , Mahanti, Aniket
- Date: 2020
- Type: Text , Conference proceedings
- Relation: 2020 International Symposium on Networks, Computers and Communications (ISNCC); Montreal, Canada; 20-22nd October, 2020, p.1-7
- Full Text: false
- Reviewed:
- Description: Cultural venues like museums increasingly seek to harness the value of data analytics to make data driven decisions related to exhibitions duration, marketing campaigns, resource planning, and revenue optimization. One key priority is the need to understand the influencing factors behind visitor attendance. Using data collected from a large museum, we investigated whether the weather has a significant impact on visitor attendance or that other factors are more important. We applied the Cross Industry Standard Process for Data Mining (CRISP-DM) methodology to perform the research, developed and built four different types of regression models using R and its machine learning packages to model visitor attendance. The models were trained and evaluated. Predictions of visitor attendance were then generated from each of the four models and forecast accuracy was measured. The extreme gradient boost model was the best model with the highest average forecast accuracy of 93% and lowest forecast variability when benchmarked against the actual visitor attendance from the test data set. The weather was not considered to be as significant in predicting visitor trends and numbers to the museum compared to factors like time of the day, day of the week and school holidays. However, it was still measured to have a slight impact as excluding weather variables resulted in a model with a poorer fit. Weather can potentially have a more marked impact on cultural attractions in more extreme weather environments and outdoor venues.
Educational big data : predictions, applications and challenges
- Authors: Bai, Xiaomei , Zhang, Fuli , Li, Jinzhou , Guo, Teng , Xia, Feng
- Date: 2021
- Type: Text , Journal article , Review
- Relation: Big Data Research Vol. 26, no. (2021), p.
- Full Text:
- Reviewed:
- Description: Educational big data is becoming a strategic educational asset, exceptionally significant in advancing educational reform. The term educational big data stems from the rapidly growing educational data development, including students' inherent attributes, learning behavior, and psychological state. Educational big data has many applications that can be used for educational administration, teaching innovation, and research management. The representative examples of such applications are student academic performance prediction, employment recommendation, and financial support for low-income students. Different empirical studies have shown that it is possible to predict student performance in the courses during the next term. Predictive research for the higher education stage has become an attractive area of study since it allowed us to predict student behavior. In this survey, we will review predictive research, its applications, and its challenges. We first introduce the significance and background of educational big data. Second, we review the students' academic performance prediction research, such as factors influencing students' academic performance, predicting models, evaluating indices. Third, we introduce the applications of educational big data such as prediction, recommendation, and evaluation. Finally, we investigate challenging research issues in this area. This discussion aims to provide a comprehensive overview of educational big data. © 2021 Elsevier Inc. **Please note that there are multiple authors for this article therefore only the name of the first 5 including Federation University Australia affiliate “Feng Xia” is provided in this record**
A new feature selection technique for load and price forecast of electrical power systems
- Authors: Abedinia, Oveis , Amjady, Nima , Zareipour, Hamidreza
- Date: 2017
- Type: Text , Journal article
- Relation: IEEE transactions on power systems Vol. 32, no. 1 (2017), p. 62-74
- Full Text: false
- Reviewed:
- Description: Load and price forecasts are necessary for optimal operation planning in competitive electricity markets. However, most of the load and price forecast methods suffer from lack of an efficient feature selection technique with the ability of modeling the nonlinearities and interacting features of the forecast processes. In this paper, a new feature selection method is presented. An important contribution of the proposed method is modeling interaction in addition to relevancy and redundancy, based on information-theoretic criteria, for feature selection. Another main contribution of the paper is proposing a hybrid filter-wrapper approach. The filter part selects a minimum subset of the most informative features by considering relevancy, redundancy, and interaction of the candidate inputs in a coordinated manner. The wrapper part fine-tunes the settings of the composite filter.