On stability of M-stationary points in MPCCs
- Authors: Červinka, Michal , Outrata, Jiri , Pištěk, M
- Date: 2014
- Type: Text , Journal article
- Relation: Set-Valued and Variational Analysis Vol. 22, no. 3 (2014), p. 575-595
- Relation: http://purl.org/au-research/grants/arc/DP110102011
- Full Text: false
- Reviewed:
- Description: We consider parameterized Mathematical Programs with Complementarity Constraints arising, e.g., in modeling of deregulated electricity markets. Using the standard rules of the generalized differential calculus we analyze qualitative stability of solutions to the respective M-stationarity conditions. In particular, we provide characterizations and criteria for the isolated calmness and the Aubin properties of the stationarity map. To this end, we introduce the second-order limiting coderivative of mappings and provide formulas for this notion and for the graphical derivative of the limiting coderivative in the case of the normal cone mapping to ℝn Funding ARC- Australian Research Council
The effect of stabilization on the utilization of municipal sewage sludge as a soil amendment
- Authors: Černe, Marko , Palčić, Igor , Pasković, Igor , Major, Nikola , Romić, Marija , Filipović, Vilim , Igrc, Marina Diana , Perčin, Aleksandra , Goreta Ban, Smiljana , Zorko, Benjamin , Vodenik, Branko , Glavič Cindro, Denis , Milačič, Radmila , Heath, David John , Ban, Dean
- Date: 2019
- Type: Text , Journal article
- Relation: Waste Management Vol. 94, no. (2019), p. 27-38
- Full Text: false
- Reviewed:
- Description: Stabilization procedures affect nutrient and inorganic contaminant concentrations.•Anaerobic sludge has the highest levels of metals and radionuclides.•High P and N contents in sludge are a measure of its potential as a fertilizer.•Adsorption of metals and radionuclides to sludge biomass increases with nutrient content.•Soil amendment with sewage sludge is a promising strategy for nutrient recovery. This study assesses the potential use of different types of stabilized sewage sludge as a soil amendment by considering their physicochemical characteristics, nutritional status, and their trace metal and radionuclide content. The concentrations of trace metals and radionuclides were determined using ICP-OES and gamma-ray spectrometry, respectively. For determining nutritional status and chemical characterization, this study followed standard ISO-recommended procedures. Data analysis revealed that anaerobic sludge contains higher concentrations of Cr, Hg, and Ni compared to aerobic and non-biologically stabilized sludge. A similar observation was observed in the case of 226Ra, 210Pb, 228Ra, and 228Th. Furthermore, the high levels of P and N in aerobic sludge suggest that biologically stabilized sludge has the potential to be a good fertilizer. In addition, the study finds strong evidence that nutrients are involved in the adsorption of metals and radionuclides onto sludge biomass. Overall, eight of the nine studied sludge samples are safe for agricultural use since the concentrations of trace metals fall well below the limits set by Croatian legislation (NN 38/08). In addition, the levels of radionuclides do not pose a radiological risk. This means that soil conditioning with sewage sludge remains a viable strategy for nutrient recovery from municipal waste, although long-term impact assessments of repeated applications are necessary.
Effect of sewage sludge derived compost or biochar amendment on the phytoaccumulation of potentially toxic elements and radionuclides by Chinese cabbage
- Authors: Černe, Marko , Palčić, Igor , Major, Nikola , Pasković, Igor , Perković, Josipa , Užila, Zoran , Filipović, Vilim , Romić, Marija , Goreta Ban, Smiljana , Jaćimović, Radojko , Benedik, Ljudmila , Heath, David John , Ban, Dean
- Date: 2021
- Type: Text , Journal article
- Relation: Journal of environmental management Vol. 293, no. (2021), p. 112955-112955
- Full Text: false
- Reviewed:
- Description: This study set out to evaluate the effect of using sewage sludge-derived compost (SSC) or biochar (SSB) as a soil amendment on the phytoaccumulation of potentially toxic elements, PTE (Cd, Cr, Cu, Ni, Pb, Zn) and natural radionuclides (238U and 232Th) by Chinese cabbage (Brassica rapa L. subsp. pekinensis (Lour.) Hanelt) in terra rossa and rendzina soils, which are the two common soil types in Croatia. The experiment consisted of a greenhouse pot trial using a three-factor design where soil type, sludge post-stabilisation procedure and amendment rate (12 and 120 mgP/L) were the main factors. At harvest, the concentrations of analytes in the substrate, leaves and roots were measured, from which the edible tissue uptake (ETU) and concentration ratios (CR) were determined. Also, the average daily dose (ADD) and hazard quotient (HQ) were determined to assess the health risk, as well as soil contamination factor (CF). The results showed that neither adding SSC nor SSB affected the soil loading at the rates applied, suggesting a low risk of soil contamination (CF ≤ 1). The ETU of Cd, Cu, and Zn were 0.0061, 1.23, and 0.91 mg/plant from compost-amended soil and 0.0046, 0.78 and 0.65 mg/plant for biochar-amended soil, respectively. This difference suggests that their ETU was higher in compost-amended soils than in soils treated with biochar. The CR data indicate that the bioavailability of Cu (CR of 5.30) is highest at an amendment rate of 12 mgP/L, while for Zn (CR of 0.69), the highest bioaccumulation was observed with an amendment rate of 120 mgP/L. Translocation of Cr, Ni, Pb and 238U to the leaves was limited. Overall, the HQ (<1) for Cd, Cu and Zn in the edible parts confirmed that consuming Chinese cabbage does not threaten human health. Similarly, the daily intake of 232Th remained below the limit (3 μg) set by ICRP, suggesting no radiological risk. Finally, although the amendment rate, which was 10-times the amount stipulated in Croatian regulation and the CR ranged from 0.007 to 5.30, the precautionary principle is advised, and the long-term impact of sewage sludge derived compost or biochar on different plant groups (incl. root vegetables) at the field-scale is recommended. [Display omitted] •Sewage sludge compost or biochar amendment did not result in soil contamination.•Biochar amendment reduced Cd, Cu and Zn edible tissue uptake relative to compost use.•No evidence of phytoaccumulation of Cr, Ni, Pb and 238U in Chinese cabbage leaves.•Levels of Cd, Cu, Zn and 232Th in the edible tissues does not pose a health risk.
The new robust conic GPLM method with an application to finance : prediction of credit default
- Authors: Özmen, Ay , Weber, Gerhard-Wilhelm , Çavu , Defterli, Özlem
- Date: 2012
- Type: Text , Journal article
- Relation: Journal of Global Optimization Vol.56, no. 2 (2012), p. 233–249
- Full Text: false
- Reviewed:
- Description: This paper contributes to classification and identification in modern finance through advanced optimization. In the last few decades, financial misalignments and, thereby, financial crises have been increasing in numbers due to the rearrangement of the financial world. In this study, as one of the most remarkable of these, countries' debt crises, which result from illiquidity, are tried to predict with some macroeconomic variables. The methodology consists of a combination of two predictive regression models, logistic regression and robust conic multivariate adaptive regression splines (RCMARS), as linear and nonlinear parts of a generalized partial linear model. RCMARS has an advantage of coping with the noise in both input and output data and of obtaining more consistent optimization results than CMARS. An advanced version of conic generalized partial linear model which includes robustification of the data set is introduced: robust conic generalized partial linear model (RCGPLM). This new model is applied on a data set that belongs to 45 emerging markets with 1,019 observations between the years 1980 and 2005. © 2012 Springer Science+Business Media, LLC.
Deltamides and croconamides: Expanding the range of dual h‐bond donors for selective anion recognition
- Authors: Zwicker, Vincent , Yuen, Karen , Smith, David , Ho, Junming , Qin, Lei , Turner, Peter , Jolliffe, Katrina
- Date: 2018
- Type: Text , Journal article
- Relation: Chemistry : a European journal Vol. 24, no. 5 (2018), p. 1140-1150
- Full Text: false
- Reviewed:
- Description: Dual H‐bond donors are widely used as recognition motifs in anion receptors. We report the synthesis of a library of dual H‐bond receptors, incorporating the deltic and croconic acid derivatives, termed deltamides and croconamides, respectively, and a comparison of their anion binding affinities (for monovalent species) and Brønsted acidities to those of the well‐established urea and squaramide dual H‐bond donor motifs. For dual H‐bonding cores with identical substituents, the trend in Brønsted acidity is croconamides>squaramides>deltamides>ureas, with the croconamides found to be 10–15 pKa units more acidic than the corresponding ureas. In contrast to the trends displayed by ureas, deltamides and squaramides, N,N′‐dialkyl croconamides displayed higher binding affinity to chloride than the N,N′‐diaryl derivatives, which was attributed to partial deprotonation of the N,N′‐diaryl derivatives at neutral pH. A number of differences in anion binding selectivity were observed upon comparison of the dual H‐bond cores. Whereas the squaramides display similar affinity for both chloride and acetate ions, the ureas have significantly higher affinity for acetate than chloride ions and the deltamides display higher affinity for dihydrogenphosphate ions than other oxoanions or halides. These inherent differences in binding affinity could be exploited in the design of anion receptors with improved ability to discriminate between monovalent anions. Croconamides spotted in the delta! Bite‐angle and Brønsted acidity control anion binding affinity and selectivity by amides from the oxo‐carbon family.
An application of near infra-red fibre bragg grating as dynamic sensor in SHM of thin composite laminates
- Authors: Zohari, Mohd , Kahandawa, Gayan , Epaarachchi, Jayantha , Lau, Alan , Cook, Kevin , Canning, John
- Date: 2013
- Type: Text , Conference paper
- Relation: Structural health monitoring 2013 : a roadmap to intelligent structures : proceedings of the 9th International Workshop on Structural Health Monitoring p. 267-275
- Full Text: false
- Reviewed:
- Description: Vibration testing is an essential component in Structural Health Monitoring (SHM). It can provide vital information regarding the integrity of critical structure; for instance, it can provide information on progressive failure monitoring of composites structure in the aerospace industry. Over the past decade, there have been many successful researches showing extraordinary ability of Fiber Bragg grating (FBG) sensors as a dynamic sensor. Ability of acquiring both static and dynamic strain measurements, make FBG sensor as a good alternative to replace the conventional vibration sensors. In addition the physical size of FBG sensor provides greater access to embed them in composite structures without affecting to any properties of the composite. However, in most applications to date, people have used only the FBG with wavelength 1550 nm. Moreover, FBG sensors with this wavelength are commonly use in industries such as telecommunications and medical industries. However, there is an option of using near infra-red (NIR) FBG range which comparably cheap in term of total system design. This paper details the use of near infra-red (NIR) FBGs as dynamic sensors; a part of SHM system for the monitoring of the damages in a thin glass fiber composite plates. Results reveal that the NIR FBG range gives good response to an impact and; also to applied high frequency vibrations.
NIR fibre bragg grating as dynamic sensor : An application of 1D digital wavelet analysis for signal denoising
- Authors: Zohari, Mohd , Kahandawa, Gayan , Epaarachchi, Jayantha , Lau, Alan , Canning, John , Cook, Kevin
- Date: 2013
- Type: Text , Conference paper
- Relation: Fourth International Conference on Smart Materials and Nanotechnology in Engineering
- Full Text: false
- Reviewed:
- Description: During the past decade, many successful studies have evidently shown remarkable capability of Fiber Bragg Gratings (FBG) sensor for dynamic sensing. Most of the research works utilized the 1550 nm wavelength range of FBG sensors. However near infra-red (NIR) FBG sensors can offer the lower cost of Structural health Monitoring (SHM) systems which uses cheaper silicon sources and detectors. Unfortunately, the excessive noise levels that experienced in NIR wavelengths have caused the rejection of sensor that operating in this range of wavelengths for SHM systems. However, with the appropriate use of signal processing tools, these noisy signals can be easily ‘cleaned’. Wavelet analysis is one of the powerful signal processing tools nowadays, not only for time-frequency analysis but also for signal denoising. This present study revealed that the NIR FBG range gave good response to impact signals. Furthermore, these ‘noisy’ signals’ response were successfully filtered using one dimensional wavelet analysis.
EMG-triggered stimulation post spinal cord injury: A case report
- Authors: Zoghi, Maryam , Galea, Mary
- Date: 2018
- Type: Text , Journal article
- Relation: Physiotherapy theory and practice :a n international journal of physical therapy Vol. 34, no. 4 (2018), p. 309-315
- Full Text: false
- Reviewed:
- Description: Background: Mechanical injury in patients with spinal cord injury (SCI) rarely transects the cord completely, even when the injury is classified as complete. These patients can show sub-clinical evidence of spared motor connections, which might be amenable to targeted rehabilitation. Neurophysiological evaluations can complement the clinical evaluation by providing objective data about conduction across the SCI site. Case Description: A twenty-four year old patient with SCI was admitted to a rehabilitation centre 49 days post traumatic SCI. His injury was categorized as motor and sensory complete (AIS A) with a neurological level of C4. The strength of his triceps bilaterally was recorded 0/5 repeatedly by his therapists during the five-month period post-injury. As a result, no training was provided for these muscles during the rehabilitation program. Neurophysiological Assessment: Motor evoked potentials (MEPs) were recorded from his left triceps with transcranial magnetic stimulation (TMS) which confirmed the existence of spared corticospinal connections to this muscle post-injury. Intervention: He completed a series of active-assisted exercises with an EMG-triggered neuromuscular stimulation (NMS) device for his left triceps comprising 20-minutes elbow extension (15 trials), three times per day for 4 weeks. Outcome: The strength of his left triceps gradually improved to 2/5. Discussion: Neurophysiological evaluation can be useful in identifying residual function below the level of injury, which can, in turn, be enhanced through appropriate rehabilitation strategies.
Brain motor control assessment post early intensive hand rehabilitation after spinal cord Injury
- Authors: Zoghi, Maryam , Galea, Mary
- Date: 2018
- Type: Text , Journal article
- Relation: Topics in Spinal cord injury rehabilitation Vol. 24, no. 2 (2018), p. 157-166
- Full Text: false
- Reviewed:
- Description: The Brain Motor Control Assessment (BMCA) is a surface electromyography (sEMG)-based measure of motor output from the central nervous system during a variety of reflex and voluntary motor tasks. The aim of this study was to assess the pattern of voluntary movements in patients with spinal cord injury (SCI) to investigate whether BMCA could add more resolution to clinical assessments and the recovery path of these patients. Ten participants were recruited from the Royal Talbot Rehabilitation Centre as part of a multicenter randomized controlled trial. Four participants received usual care while the other 3 participants received usual care plus an intensive task-specific hand training program in conjunction with functional electrical stimulation for 8 weeks. BMCA assessments were completed for 7 participants at this center 4 times over a period of 1 year. Generalized linear model analysis showed a significant main effect of task ( < .001) and assessment time ( = .003) on the Similarity Index. However, there were no significant interactions among the factors ( > .05). Based on ARAT or summed upper limb strength scores, some participants showed significant improvement after 8 weeks of rehabilitation, however this improvement was not reflected in the pattern of muscle activation that was captured by BMCA. The quantifiable features of BMCA through surface EMG may increase the resolution of SCI characterization by adding subclinical details to the clinical picture of lesion severity and progression during rehabilitation.
Cathodal transcranial direct‐current stimulation for treatment of drug‐resistant temporal lobe epilepsy: A pilot randomized controlled trial
- Authors: Zoghi, Maryam , O'Brien, Terence , Kwan, Patrick , Cook, Mark , Galea, Mary , Jaberzadeh, Shapour
- Date: 2016
- Type: Text , Journal article
- Relation: Epilepsia Open Vol. 1, no. 3-4 (2016), p. 130-135
- Full Text: false
- Reviewed:
- Description: Summary Objective To investigate the effect of cathodal transcranial direct‐current stimulation (c‐tDCS) on seizure frequency in patients with drug‐resistant temporal lobe epilepsy (TLE). Method Twenty‐nine patients with drug‐resistant TLE participated in this study. They were randomized to experimental or sham group. Twenty participants (experimental group) received within‐session repeated c‐tDCS intervention over the affected temporal lobe, and nine (sham group) received sham tDCS. Paired‐pulse transcranial magnetic stimulation was used to assess short interval intracortical inhibition (SICI) in primary motor cortex ipsilateral to the affected temporal lobe. SICI was measured from motor evoked potentials recorded from the contralateral first dorsal interosseous muscle. Adverse effects were monitored during and after each intervention in both groups. A seizure diary was given to each participant to complete for 4 weeks following the tDCS intervention. The mean response ratio was calculated from their seizure rates before and after the tDCS intervention. Results The experimental group showed a significant increase in SICI compared to the sham group (F = 10.3, p = 0.005). None of the participants reported side effects of moderate or severe degree. The mean response ratio in seizure frequency was −42.14% (standard deviation [SD] 35.93) for the experimental group and −16.98% (SD 52.41) for the sham group. Significance Results from this pilot study suggest that tDCS may be a safe and efficacious nonpharmacologic intervention for patients with drug‐resistant TLE. Further evaluation in larger double‐blind randomized controlled trials is warranted.
Brain motor control assessment of upper limb function in patients with spinal cord injury
- Authors: Zoghi, Maryam , Galea, Mary , Morgan, David
- Date: 2016
- Type: Text , Journal article
- Relation: Journal of Spinal Cord Medicine Vol. 39, no. 2 (2016), p. 162-174
- Full Text: false
- Reviewed:
- Description: Background: The brain motor control assessment (BMCA) for the upper limb has been developed to add resolution to the clinical evaluation in patients with spinal cord injury (SCI). BMCA is a surface electromyography (sEMG)-based measure of motor output from the central nervous system during a variety of reflex and voluntary motor tasks performed under strictly controlled conditions. Method: Nine participants were recruited and assessed four times over a period of 1 year in a prospective cohort study design. The sEMG of 15 muscles (7 muscles from each upper limb and rectus abdominis) were recorded throughout the following stages of the BMCA protocol: (i) relaxation, (ii) reinforcement maneuvers, (iii) voluntary tasks, (iv) tendon-tap reflex responses, (v) vibration responses. Results: Similarity index (SI) values were significantly lower in the SCI group for unilateral shoulder abduction (P = 0.006) and adduction (P = 0.021), elbow extension (P = 0.038), wrist flexion/extension with palm up (P < 0.001 P < 0.001) and wrist flexion with palm down (P = 0.016). sEMG magnitudes were also significantly lower in the SCI group for wrist flexion/extension with palm up (P < 0.001 P = 0.042). SI changes over time were significant for tasks related to wrist joint (P = 0.002). Conclusion: Clinicians who are involved in rehabilitation of patients with SCI can use the BMCA to assess their patients' motor control abilities and monitor their progression throughout their rehabilitation process. The results of this type of neurophysiological assessment might be useful to tailor therapeutic strategies for each patient.
What does it mean when they don't seem to learn from experience?
- Authors: Zink, Robyn , Dyson, Michael
- Date: 2009
- Type: Text , Journal article
- Relation: Cambridge Journal of Education Vol. 39, no. 2 (2009), p. 163-174
- Full Text: false
- Reviewed:
- Description: In this paper we use an example of when students appear not to have learnt from their experience to examine some of the 'orthodoxies' of experiential education. This frames a discussion exploring how it is possible for a teacher to declare that students have got it 'wrong' in relation to learning from experience. It is argued that both learning through experience and who has experiences is viewed through specific forms of reason within contemporary experiential education. The paper concludes with a challenge to open reason up to greater scrutiny in experiential education and consider the possibilities that emerge through the indeterminacy of relationships inherent in experience and education. [ABSTRACT FROM AUTHOR]
Factors influencing hand hygiene practice of nursing students : a descriptive, mixed-methods study
- Authors: Zimmerman, Peta-Anne , Sladdin, Ishtar , Shaban, Ramon , Gilbert, Julia , Brown, Lynne
- Date: 2020
- Type: Text , Journal article
- Relation: Nurse Education in Practice Vol. 44, no. (2020), p.
- Full Text: false
- Reviewed:
- Description: Developing nursing students' knowledge and practice of infection prevention and control (IPC) is fundamental to safe healthcare. A two-phase descriptive, mixed-method study conducted within a Bachelor of Nursing program at an Australian university aimed to explore: (i) theoretical knowledge of IPC, highlighting hand hygiene, of nursing students and; (ii) nursing students' and clinical facilitators' perceptions of factors influencing these practices during clinical placement. Phase One utilised an anonymous validated questionnaire assessing students' knowledge; identifying variables influencing students' IPC practices, subjected to descriptive and inferential analysis. Phase Two were semi-structured interviews exploring clinical facilitators' experiences/perceptions of students during clinical placement, analysed thematically. Students' demonstrated satisfactory knowledge of IPC in their second and third year, but clinical facilitators perceived that. students lacked awareness of the importance of these practices. Five themes arose from the interviews: (i) understanding workplace culture; (ii) students' modelling local behaviour; (iii) enhancing and consolidating knowledge for practice; (iv) adjusting to practice reality and; (v) accessing additional hand hygiene resources. Factors specific to workplace setting and culture were perceived to influence nursing students' socialisation. Future practice/education strategies could address these factors by ensuring students receive adequate supervision during clinical placement, and having strong advocates/role models present in the workplace. © 2020 Elsevier Ltd
Rainfall and grazing : Not the only barriers to arid-zone conifer recruitment
- Authors: Zimmer, Heidi , Florentine, Singarayer , Enke, Rita , Westbrooke, Martin
- Date: 2017
- Type: Text , Journal article
- Relation: Australian Journal of Botany Vol. 65, no. 2 (2017), p. 109-119
- Full Text: false
- Reviewed:
- Description: An understanding of the drivers of infrequent recruitment is fundamental in managing for species persistence. Callitris glaucophylla Joy Thomps. & L.A.S. Johnson (white cypress-pine) is a slow-growing, long-lived conifer, with a distribution that extends across arid Australia. Arid populations of C. glaucophylla are endangered in New South Wales, and are characterised by infrequent recruitment. We examined recruitment patterns of C. glaucophylla in differential grazing exclosures (excluding rabbits, excluding large herbivores or excluding both) and in unfenced areas. More recruitment occurred in rabbit-proof exclosures, compared with nearby large herbivore and control exclosures, although some rabbit-proof exclosures recorded no recruitment. Increases in recruitment at several long-term exclosures were associated with wet periods, as was recruitment at some unfenced sites. Apart from grazing and rainfall, recruitment was related to mature tree size and stand density (probably because of their influence on seed availability). These endangered arid C. glaucophylla woodlands are all that remains of a once extensive distribution, and are habitat for a suite of threatened species. Reduction in grazing pressure, particularly from rabbits, is clearly critical to maintain recruitment in these remnant populations. Chances of recruitment appear to be enhanced in low-density stands, around large trees, and in dune blow outs - fencing should focus on these areas. © 2017 CSIRO.
A study of hydrodynamic pressure in the A-pillar regions of idealised and production cars
- Authors: Zimmer, Gary , Alam, Firoz , Watkins, Simon
- Date: 2004
- Type: Text , Conference paper
- Relation: Paper presented at the 2nd BSME - ASME International Conference on Thermal Engineering, Dhaka, Bangladesh : 2nd January, 2004
- Full Text: false
- Reviewed:
- Description: E1
- Description: 2003000704
A study of the A-pillar vortex of a passenger car
- Authors: Zimmer, Gary , Alam, Firoz , Watkins, Simon
- Date: 2003
- Type: Text , Conference paper
- Relation: Paper presented at the International Conference on Mechanical Engineering 2003, Dhaka, Bangladesh : 26th December, 2003
- Full Text: false
- Reviewed:
- Description: E1
- Description: 2003000641
Acute cardiovascular responses during post-exercise recovery to short-term exhaustive exercise
- Authors: Zichy-Woinarski, Christopher
- Date: 2000
- Type: Text , Thesis , Masters
- Full Text: false
- Description: "The aim of this study was to examine the effect of consecutive bouts of acute exhaustive exercise in a hot ambient environmemt on baroreflex gain, plasma volume and heart variability one week post-exercise in trained individuals." Problem with pages xviii-xx Pages 9-37 missing.
- Description: Masters of Applied Science
Discover multiple novel labels in multi-instance multi-label learning
- Authors: Zhu, Yue , Ting, Kaiming , Zhou, Zhi-Hua
- Date: 2017
- Type: Text , Conference paper
- Relation: Thirty-First AAAI Conference on Artificial Intelligence p. 2977-2984
- Full Text: false
- Reviewed:
- Description: Multi-instance multi-label learning (MIML) is a learning paradigm where an object is represented by a bag of instances and each bag is associated with multiple labels. Ordinary MIML setting assumes a fixed target label set. In real applications, multiple novel labels may exist outside this set, but hidden in the training data and unknown to the MIML learner. Existing MIML approaches are unable to discover the hidden novel labels, let alone predicting these labels in the previously unseen test data. In this paper, we propose the first approach to discover multiple novel labels in MIML problem using an efficient augmented lagrangian optimization, which has a bag-dependent loss term and a bag-independent clustering regularization term, enabling the known labels and multiple novel labels to be modeled simultaneously. The effectiveness of the proposed approach is validated in experiments.
New class adaptation via instance generation in one-pass class incremental learning
- Authors: Zhu, Yue , Ting, Kaiming , Zhou, Zhi-Hua
- Date: 2017
- Type: Text , Conference paper
- Relation: 2017 IEEE International Conference on Data Mining (ICDM)
- Full Text: false
- Reviewed:
- Description: One pass learning updates a model with only a single scan of the dataset, without storing historical data. Previous studies focus on classification tasks with a fixed class set, and will perform poorly in an open dynamic environment when new classes emerge in a data stream. The performance degrades because the classifier needs to receive a sufficient number of instances from new classes to establish a good model. This can take a long period of time. In order to reduce this period to deal with any-time prediction task, we introduce a framework to handle emerging new classes called One-Pass Class Incremental Learning (OPCIL). The central issue in OPCIL is: how to effectively adapt a classifier of existing classes to incorporate emerging new classes. We call it the new class adaptation issue, and propose a new approach to address it, which requires only one new class instance. The key is to generate pseudo instances which are optimized to satisfy properties that produce a good discriminative classifier. We provide the necessary propertiesand optimization procedures required to address this issue. Experiments validate the effectiveness of this approach.
Density-ratio based clustering for discovering clusters with varying densities
- Authors: Zhu, Ye , Ting, Kaiming , Carman, Mark
- Date: 2016
- Type: Text , Journal article
- Relation: Pattern Recognition Vol. 60, no. (2016), p. 983-997
- Full Text: false
- Reviewed:
- Description: Density-based clustering algorithms are able to identify clusters of arbitrary shapes and sizes in a dataset which contains noise. It is well-known that most of these algorithms, which use a global density threshold, have difficulty identifying all clusters in a dataset having clusters of greatly varying densities. This paper identifies and analyses the condition under which density-based clustering algorithms fail in this scenario. It proposes a density-ratio based method to overcome this weakness, and reveals that it can be implemented in two approaches. One approach is to modify a density-based clustering algorithm to do density-ratio based clustering by using its density estimator to compute density-ratio. The other approach involves rescaling the given dataset only. An existing density-based clustering algorithm, which is applied to the rescaled dataset, can find all clusters with varying densities that would otherwise impossible had the same algorithm been applied to the unscaled dataset. We provide an empirical evaluation using DBSCAN, OPTICS and SNN to show the effectiveness of these two approaches. © 2016 Elsevier Ltd