Anban Pillay

2022

Price, C. S. ., Moodley, D. ., Pillay, A. ., & Rens, G. . (2022). An adaptive probabilistic agent architecture for modelling sugarcane growers’ decision-making. South African Computer Journal, 34(1). http://doi.org/https://doi.org/10.18489/sacj.v34i1.857

Abstract BibTex Entry BibTeX Download PDF

An adaptive probabilistic agent architecture for modelling sugarcane growers’ decision-making

Building computational models of agents in dynamic, partially observable and stochastic environments is challenging. We propose a cognitive computational model of sugarcane growers’ daily decision-making to examine sugarcane supply chain complexities. Growers make decisions based on uncertain weather forecasts; cane dryness; unforeseen emergencies; and the mill’s unexpected call for delivery of a different amount of cane. The Belief-Desire-Intention (BDI) architecture has been used to model cognitive agents in many domains, including agriculture. However, typical implementations of this architecture have represented beliefs symbolically, so uncertain beliefs are usually not catered for. Here we show that a BDI architecture, enhanced with a dynamic decision network (DDN), suitably models sugarcane grower agents’ repeated daily decisions. Using two complex scenarios, we demonstrate that the agent selects the appropriate intention, and suggests how the grower should act adaptively and proactively to achieve his goals. In addition, we provide a mapping for using a DDN in a BDI architecture. This architecture can be used for modelling sugarcane grower agents in an agent-based simulation. The mapping of the DDN’s use in the BDI architecture enables this work to be applied to other domains for modelling agents’ repeated decisions in partially observable, stochastic and dynamic environments.

@article{488,
  author = {C. Sue Price and Deshen Moodley and Anban Pillay and Gavin Rens},
  title = {An adaptive probabilistic agent architecture for modelling sugarcane growers’ decision-making},
  abstract = {Building computational models of agents in dynamic, partially observable and stochastic environments is challenging.  We propose a cognitive computational model of sugarcane growers’ daily decision-making to examine sugarcane supply chain complexities.  Growers make decisions based on uncertain weather forecasts; cane dryness; unforeseen emergencies; and the mill’s unexpected call for delivery of a different amount of cane.  The Belief-Desire-Intention (BDI) architecture has been used to model cognitive agents in many domains, including agriculture.  However, typical implementations of this architecture have represented beliefs symbolically, so uncertain beliefs are usually not catered for.  Here we show that a BDI architecture, enhanced with a dynamic decision network (DDN), suitably models sugarcane grower agents’ repeated daily decisions.  Using two complex scenarios, we demonstrate that the agent selects the appropriate intention, and suggests how the grower should act adaptively and proactively to achieve his goals.  In addition, we provide a mapping for using a DDN in a BDI architecture.  This architecture can be used for modelling sugarcane grower agents in an agent-based simulation.  The mapping of the DDN’s use in the BDI architecture enables this work to be applied to other domains for modelling agents’ repeated decisions in partially observable, stochastic and dynamic environments.},
  year = {2022},
  journal = {South African Computer Journal},
  volume = {34},
  pages = {152-191},
  issue = {1},
  url = {https://sacj.cs.uct.ac.za/index.php/sacj/article/view/857},
  doi = {https://doi.org/10.18489/sacj.v34i1.857},
}

2020

Clark, A. ., Pillay, A. ., & Moodley, D. . (2020). A system for pose analysis and selection in virtual reality environments. In SAICSIT ’20: Conference of the South African Institute of Computer Scientists and Information Technologists 2020. Virtual: ACM Digital Library. Retrieved from https://dl.acm.org/doi/proceedings/10.1145/3410886 (Original work published 2020)

Abstract BibTex Entry BibTeX Download PDF

Clark, Pillay & Moodley.pdf

Depth cameras provide a natural and intuitive user interaction mechanism in virtual reality environments by using hand gestures as the primary user input. However, building robust VR systems that use depth cameras are challenging. Gesture recognition accuracy is affected by occlusion, variation in hand orientation and misclassification of similar hand gestures. This research explores the limits of the Leap Motion depth camera for static hand pose recognition in virtual reality applications. We propose a system for analysing static hand poses and for systematically identifying a pose set that can achieve a near-perfect recognition accuracy. The system consists of a hand pose taxonomy, a pose notation, a machine learning classifier and an algorithm to identify a reliable pose set that can achieve near perfect accuracy levels. We used this system to construct a benchmark hand pose data set containing 2550 static hand pose instances, and show how the algorithm can be used to systematically derive a set of poses that can produce an accuracy of 99% using a Support Vector Machine classifier.

@{379,
  author = {Andrew Clark and Anban Pillay and Deshen Moodley},
  title = {A system for pose analysis and selection in virtual reality environments},
  abstract = {Depth cameras provide a natural and intuitive user interaction mechanism in virtual reality environments by using hand gestures as the primary user input. However, building robust VR systems that use depth cameras are challenging. Gesture recognition accuracy is affected by occlusion, variation in hand orientation and misclassification of similar hand gestures. This research explores the limits of the Leap Motion depth camera for static hand pose recognition in virtual reality applications. We propose a system for analysing static hand poses and for systematically identifying a pose set that can achieve a near-perfect recognition accuracy. The system consists of a hand pose taxonomy, a pose notation, a machine learning classifier and an algorithm to identify a reliable pose set that can achieve near perfect accuracy levels. We used this system to construct a benchmark hand pose data set containing 2550 static hand pose instances, and show how the algorithm can be used to systematically derive a set of poses that can produce an accuracy of 99% using a Support Vector Machine classifier.},
  year = {2020},
  journal = {SAICSIT &#039;20: Conference of the South African Institute of Computer Scientists and Information Technologists 2020},
  pages = {210-216},
  month = {14/09/2020},
  publisher = {ACM Digital Library},
  address = {Virtual},
  isbn = {978-1-4503-8847-4},
  url = {https://dl.acm.org/doi/proceedings/10.1145/3410886},
}

2019

Price, C. S. ., Moodley, D. ., & Pillay, A. . (2019). Modelling uncertain adaptive decisions: Application to KwaZulu-Natal sugarcane growers. In Forum for Artificial Intelligence Research (FAIR2019). Cape Town: CEUR. Retrieved from http://ceur-ws.org/Vol-2540/FAIR2019_paper_53.pdf

Abstract BibTex Entry BibTeX Download PDF

FAIR2019_paper_53.pdf

A dynamic Bayesian decision network was developed to model the preharvest burning decision-making processes of sugarcane growers in a KwaZulu-Natal sugarcane supply chain and extends previous work by Price et al. (2018). This model was created using an iterative development approach. This paper recounts the development and validation process of the third version of the model. The model was validated using Pitchforth and Mengersen (2013)’s framework for validating expert elicited Bayesian networks. During this process, growers and cane supply members assessed the model in a focus group by executing the model, and reviewing the results of a prerun scenario. The participants were generally positive about how the model represented their decision-making processes. However, they identified some issues that could be addressed in the next iteration. Dynamic Bayesian decision networks offer a promising approach to modelling adaptive decisions in uncertain conditions. This model can be used to simulate the cognitive mechanism for a grower agent in a simulation of a sugarcane supply chain.

@{244,
  author = {C. Sue Price and Deshen Moodley and Anban Pillay},
  title = {Modelling uncertain adaptive decisions: Application to KwaZulu-Natal sugarcane growers},
  abstract = {A dynamic Bayesian decision network was developed to model the preharvest burning decision-making processes of sugarcane growers in a KwaZulu-Natal sugarcane supply chain and extends previous work by Price et al. (2018). This model was created using an iterative development approach. This paper recounts the development and validation process of the third version of the model. The model was validated using Pitchforth and Mengersen (2013)’s framework for validating expert elicited Bayesian networks. During this process, growers and cane supply members assessed the model in a focus group by executing the model, and reviewing the results of a prerun scenario. The participants were generally positive about how the model represented their decision-making processes. However, they identified some issues that could be addressed in the next iteration. Dynamic Bayesian decision networks offer a promising approach to modelling adaptive decisions in uncertain conditions. This model can be used to simulate the cognitive mechanism for a grower agent in a simulation of a sugarcane supply chain.},
  year = {2019},
  journal = {Forum for Artificial Intelligence Research (FAIR2019)},
  pages = {145-160},
  month = {4/12-6/12},
  publisher = {CEUR},
  address = {Cape Town},
  url = {http://ceur-ws.org/Vol-2540/FAIR2019_paper_53.pdf},
}

Ikram, M. S., Pillay, A. ., & Jembere, E. . (2019). Using social networks to enhance a deep learning approach to solve the cold-start problem in recommender systems. In Forum for Artificial Intelligence Research (FAIR2019). Cape Town: CEUR. Retrieved from http://ceur-ws.org/Vol-2540/FAIR2019_paper_51.pdf

Abstract BibTex Entry BibTeX Download PDF

paper

The Cold-Start problem refers to the initial sparsity of data available to Recommender Systems that leads to poor recommendations to users. This research compares a Deep Learning Approach, a Deep Learning Approach that makes use of social information and Matrix Factorization. The social information was used to form communities of users. The intuition behind this approach is that users within a given community are likely to have similar interests. A community detection algorithm was used to group users. Thereafter a deep learning model was trained on each community. The comparative models were evaluated on the Yelp Round 9 Academic Dataset. The dataset was pruned to consist only of users with at least 1 social link. The evaluation metrics used were Mean Squared Error (MSE) and Mean Absolute Error (MAE). The evaluation was carried out using 5-fold cross-validation. The results showed that the use of social information improved on the results achieved from the Deep Learning Approach, and grouping users into communities was advantageous. However, the Deep Learning Approach that made use of social information did not outperform SVD++, a state of the art approach for recommender systems. However, the new approach shows promise for improving Deep Learning models.

@{243,
  author = {Muhammad Ikram and Anban Pillay and Edgar Jembere},
  title = {Using social networks to enhance a deep learning approach to solve the cold-start problem in recommender systems},
  abstract = {The Cold-Start problem refers to the initial sparsity of data available to Recommender Systems that leads to poor recommendations to users. This research compares a Deep Learning Approach, a Deep Learning Approach that makes use of social information and Matrix Factorization. The social information was used to form communities of users. The intuition behind this approach is that users within a given community are likely to have similar interests. A community detection algorithm was used to group users. Thereafter a deep learning model was trained on each community. The comparative models were evaluated on the Yelp Round 9 Academic Dataset. The dataset was pruned to consist only of users with at least 1 social link. The evaluation metrics used were Mean Squared Error (MSE) and Mean Absolute Error (MAE). The evaluation was carried out using 5-fold cross-validation. The results showed that the use of social information improved on the results achieved from the Deep Learning Approach, and grouping users into communities was advantageous. However, the Deep Learning Approach that made use of social information did not outperform SVD++, a state of the art approach for recommender systems. However, the new approach shows promise for improving Deep Learning models.},
  year = {2019},
  journal = {Forum for Artificial Intelligence Research (FAIR2019)},
  pages = {173-184},
  month = {4/12-6/12},
  publisher = {CEUR},
  address = {Cape Town},
  url = {http://ceur-ws.org/Vol-2540/FAIR2019_paper_51.pdf},
}

Jeewa, A. ., Pillay, A. ., & Jembere, E. . (2019). Directed curiosity-driven exploration in hard exploration, sparse reward environments. In Forum for Artificial Intelligence Research (FAIR). Cape Town: CEUR. Retrieved from http://ceur-ws.org/Vol-2540/FAIR2019_paper_42.pdf

Abstract BibTex Entry BibTeX Download PDF

Paper

Training agents in hard exploration, sparse reward environments is a diﬃcult task since the reward feedback is insuﬃcient for meaningful learning. In this work, we propose a new technique, called Directed Curiosity, that is a hybrid of Curiosity-Driven Exploration and distancebased reward shaping. The technique is evaluated in a custom navigation task where an agent tries to learn the shortest path to a distant target, in environments of varying diﬃculty. The technique is compared to agents trained with only a shaped reward signal, a curiosity signal as well as a sparse reward signal. It is shown that directed curiosity is the most successful in hard exploration environments, with the beneﬁts of the approach being highlighted in environments with numerous obstacles and decision points. The limitations of the shaped reward function are also discussed.

@{242,
  author = {Asad Jeewa and Anban Pillay and Edgar Jembere},
  title = {Directed curiosity-driven exploration in hard exploration, sparse reward environments},
  abstract = {Training agents in hard exploration, sparse reward environments is a diﬃcult task since the reward feedback is insuﬃcient for meaningful learning. In this work, we propose a new technique, called Directed Curiosity, that is a hybrid of Curiosity-Driven Exploration and distancebased reward shaping. The technique is evaluated in a custom navigation task where an agent tries to learn the shortest path to a distant target, in environments of varying diﬃculty. The technique is compared to agents trained with only a shaped reward signal, a curiosity signal as well as a sparse reward signal. It is shown that directed curiosity is the most successful in hard exploration environments, with the beneﬁts of the approach being highlighted in environments with numerous obstacles and decision points. The limitations of the shaped reward function are also discussed.},
  year = {2019},
  journal = {Forum for Artificial Intelligence Research (FAIR)},
  pages = {12 -24},
  month = {4/12-6/12},
  publisher = {CEUR},
  address = {Cape Town},
  url = {http://ceur-ws.org/Vol-2540/FAIR2019_paper_42.pdf},
}

2018

Moodley, D. ., Pillay, A. ., & Seebregts, C. . (2018). Establishing a Health Informatics Research Laboratory in South Africa . In Digital Re-imagination Colloquium 2018. NEMISA. Retrieved from http://uir.unisa.ac.za/bitstream/handle/10500/25615/Digital%20Skills%20Proceedings%202018.pdf?sequence=1&isAllowed=y

Abstract BibTex Entry BibTeX Download PDF

HeAL-NEMISA-2018.pdf

Aim/Purpose The aim of this project was to explore models for stimulating health informatics innovation and capacity development in South Africa. Background There is generally a critical lack of health informatics innovation and capacity in South Africa and sub-Saharan Africa. This is despite the wide anticipation that digital health systems will play a fundamental role in strengthening health systems and improving service delivery Methodology We established a program over four years to train Masters and Doctoral students and conducted research projects across a wide range of biomedical and health informatics technologies at a leading South African university. We also developed a Health Architecture Laboratory Innovation and Development Ecosystem (HeAL-IDE) designed to be a long-lasting and potentially reproducible output of the project. Contribution We were able to demonstrate a successful model for building innovation and capacity in a sustainable way. Key outputs included: (i)a successful partnership model; (ii) a sustainable HeAL-IDE; (iii) research papers; (iv) a world-class software product and several demonstrators; and (iv) highly trained staff. Findings Our main findings are that: (i) it is possible to create a local ecosystem for innovation and capacity building that creates value for the partners (a university and a private non-profit company); (ii) the ecosystem is able to create valuable outputs that would be much less likely to have been developed singly by each partner, and; (iii) the ecosystem could serve as a powerful model for adoption in other settings. Recommendations for Practitioners Non-profit companies and non-governmental organizations implementing health information systems in South Africa and other low resource settings have an opportunity to partner with local universities for purposes of internal capacity development and assisting with the research, reflection and innovation aspects of their projects and programmes. Recommendation for Researchers Applied health informatics researchers working in low resource settings could productively partner with local implementing organizations in order to gain a better understanding of the challenges and requirements at field sites and to accelerate the testing and deployment of health information technology solutions. Impact on Society This research demonstrates a model that can deliver valuable software products for public health. Future Research It would be useful to implement the model in other settings and research whether the model is more generally useful

@{252,
  author = {Deshen Moodley and Anban Pillay and Chris Seebregts},
  title = {Establishing a Health Informatics Research Laboratory in South Africa},
  abstract = {Aim/Purpose 
The aim of this project was to explore models for stimulating health
informatics innovation and capacity development in South Africa.
Background 
There is generally a critical lack of health informatics innovation and capacity in South Africa and sub-Saharan Africa. This is despite the wide anticipation that digital health systems will play a fundamental role in strengthening health systems and improving service delivery
Methodology 
We established a program over four years to train Masters and Doctoral students and conducted research projects across a wide range of biomedical and health informatics technologies at a leading South African university. We also developed a Health Architecture Laboratory Innovation and Development Ecosystem (HeAL-IDE) designed to be a long-lasting and potentially reproducible output of the project.
Contribution 
We were able to demonstrate a successful model for building innovation and capacity in a sustainable way. Key outputs included: (i)a successful partnership model; (ii) a sustainable HeAL-IDE; (iii) research papers; (iv) a world-class software product and several
demonstrators; and (iv) highly trained staff.
Findings 
Our main findings are that: (i) it is possible to create a local ecosystem for innovation and capacity building that creates value for the partners (a university and a private non-profit company); (ii) the ecosystem is able to create valuable outputs that would be much less likely to have been developed singly by each partner, and; (iii) the ecosystem could serve as a powerful model for adoption in other settings.
Recommendations for Practitioners
Non-profit companies and non-governmental organizations implementing health information systems in South Africa and other low resource settings have an opportunity to partner with local universities for purposes of internal capacity development and assisting with the research, reflection and innovation aspects of their projects and programmes.
Recommendation for Researchers
Applied health informatics researchers working in low resource settings could productively partner with local implementing organizations in order to gain a better understanding of the challenges and requirements at field sites and to accelerate the testing and deployment of health information technology solutions.
Impact on Society 
This research demonstrates a model that can deliver valuable software products for public health.
Future Research 
It would be useful to implement the model in other settings and research whether the model is more generally useful},
  year = {2018},
  journal = {Digital Re-imagination Colloquium 2018},
  pages = {16 - 24},
  month = {13/03 - 15/03},
  publisher = {NEMISA},
  isbn = {978-0-6399275-0-3},
  url = {http://uir.unisa.ac.za/bitstream/handle/10500/25615/Digital%20Skills%20Proceedings%202018.pdf?sequence=1&amp;isAllowed=y},
}

Ndaba, M. ., Pillay, A. ., & Ezugwu, A. . (2018). An Improved Generalized Regression Neural Network for Type II Diabetes Classiﬁcation. In ICCSA 2018, LNCS 10963 (10963rd ed.). Springer International Publishing AG.

Abstract BibTex Entry BibTeX Download PDF

10.1007@978-3-319-95171-3_52.pdf

This paper proposes an improved Generalized Regression Neural Network (KGRNN) for the diagnosis of type II diabetes. Dia- betes, a widespread chronic disease, is a metabolic disorder that develops when the body does not make enough insulin or is unable to use insulin eﬀectively. Type II diabetes is the most common type and accounts for an estimated 90% of cases. The novel KGRNN technique reported in this study uses an enhanced K-Means clustering technique (CVE-K-Means) to produce cluster centers (centroids) that are used to train the network. The technique was applied to the Pima Indian diabetes dataset, a widely used benchmark dataset for Diabetes diagnosis. The technique outper- forms the best known GRNN techniques for Type II diabetes diagnosis in terms of classiﬁcation accuracy and computational time and obtained a classiﬁcation accuracy of 86% with 83% sensitivity and 87% speciﬁcity. The Area Under the Receiver Operating Characteristic Curve (ROC) of 87% was obtained.

@inbook{195,
  author = {Moeketsi Ndaba and Anban Pillay and Absalom Ezugwu},
  title = {An Improved Generalized Regression Neural Network for Type II Diabetes Classiﬁcation},
  abstract = {This paper proposes an improved Generalized Regression Neural Network (KGRNN) for the diagnosis of type II diabetes. Dia- betes, a widespread chronic disease, is a metabolic disorder that develops when the body does not make enough insulin or is unable to use insulin eﬀectively. Type II diabetes is the most common type and accounts for an estimated 90% of cases. The novel KGRNN technique reported in this study uses an enhanced K-Means clustering technique (CVE-K-Means) to produce cluster centers (centroids) that are used to train the network. The technique was applied to the Pima Indian diabetes dataset, a widely used benchmark dataset for Diabetes diagnosis. The technique outper- forms the best known GRNN techniques for Type II diabetes diagnosis in terms of classiﬁcation accuracy and computational time and obtained a classiﬁcation accuracy of 86% with 83% sensitivity and 87% speciﬁcity. The Area Under the Receiver Operating Characteristic Curve (ROC) of 87% was obtained.},
  year = {2018},
  journal = {ICCSA 2018, LNCS 10963},
  edition = {10963},
  pages = {659-671},
  publisher = {Springer International Publishing AG},
  isbn = {3319951718},
}

Jembere, E. ., Rawatlal, R. ., & Pillay, A. . (2018). Matrix Factorisation for Predicting Student Performance. In 2017 7th World Engineering Education Forum (WEEF). IEEE.

Abstract BibTex Entry BibTeX Download PDF

WeeF-Paper-2017-Final.pdf

Predicting student performance in tertiary institutions has potential to improve curriculum advice given to students, the planning of interventions for academic support and monitoring and curriculum design. The student performance prediction problem, as defined in this study, is the prediction of a student’s mark for a module, given the student’s performance in previously attempted modules. The prediction problem is amenable to machine learning techniques, provided that sufficient data is available for analysis. This work reports on a study undertaken at the College of Agriculture, Engineering and Science at University of KwaZulu-Natal that investigates the efficacy of Matrix Factorization as a technique for solving the prediction problem. The study uses Singular Value Decomposition (SVD), a Matrix Factorization technique that has been successfully used in recommender systems. The performance of the technique was benchmarked against the use of student and course average marks as predictors of performance. The results obtained suggests that Matrix Factorization performs better than both benchmarks.

@{194,
  author = {Edgar Jembere and Randhir Rawatlal and Anban Pillay},
  title = {Matrix Factorisation for Predicting Student Performance},
  abstract = {Predicting student performance in tertiary institutions has potential to improve curriculum advice given to students, the planning of interventions for academic support and monitoring and curriculum design. The student performance prediction problem, as defined in this study, is the prediction of a student’s mark for a module, given the student’s performance in previously attempted modules. The prediction problem is amenable to machine learning techniques, provided that sufficient data is available for analysis. This work reports on a study undertaken at the College of Agriculture, Engineering and Science at University of KwaZulu-Natal that investigates the efficacy of Matrix Factorization as a technique for solving the prediction problem. The study uses Singular Value Decomposition (SVD), a Matrix Factorization technique that has been successfully used in recommender systems. The performance of the technique was benchmarked against the use of student and course average marks as predictors of performance. The results obtained suggests that Matrix Factorization performs better than both benchmarks.},
  year = {2018},
  journal = {2017 7th World Engineering Education Forum (WEEF)},
  pages = {513-518},
  month = {13/11-16/11},
  publisher = {IEEE},
  isbn = {978-1-5386-1523-2},
}

Ndaba, M. ., Pillay, A. ., & Ezugwu, A. . (2018). A Comparative Study of Machine Learning Techniques for Classifying Type II Diabetes Mellitus.

Abstract BibTex Entry BibTeX Download PDF

Thesis_Ndaba_v1.4.pdf

Diabetes is a metabolic disorder that develops when the body does not make enough insulin or is not able to use insulin effectively. Accurate and early detection of diabetes can aid in effective management of the disease. Several machine learning techniques have shown promise as cost ef- fective ways for early diagnosis of the disease to reduce the occurrence of health complications arising due to delayed diagnosis. This study compares the efﬁcacy of three broad machine learning approaches; viz. Artiﬁcial Neural Networks (ANNs), Instance-based classiﬁcation technique, and Statistical Regression to diagnose type II diabetes. For each approach, this study proposes novel techniques that extend the state of the art. The new techniques include Artiﬁcial Neural Networks hybridized with an improved K-Means clustering and a boosting technique; improved variants of Logistic Regression (LR), K-Nearest Neighbours algorithm (KNN), and K-Means clustering. The techniques were evaluated on the Pima Indian diabetes dataset and the results were compared to recent results reported in the literature. The highest classiﬁcation accuracy of 100% with 100% sensitivity and 100% speciﬁcity were achieved using an ensemble of the Boosting technique, the enhanced K-Means clustering algorithm (CVE-K-Means) and the Generalized Regression Neu- ral Network (GRNN): B-KGRNN. A hybrid of CVE-K-Means algorithm and GRNN (KGRNN) achieved the best accuracy of 86% with 83% sensitivity. The improved LR model (LR-n) achieved the highest classiﬁcation accuracy of 84% with 72% sensitivity. The new multi-layer percep- tron (MLP-BPX) achieved the best accuracy of 82% and 72% sensitivity. A hybrid of KNN and CVE-K-Means (CKNN) technique achieved the best accuracy of 81% and 89% sensitivity. CVE- K-Means technique achieved the best accuracy of 80% and 61% sensitivity. The B-KGRNN, KGRNN, LR-n, and CVE-K-Means technique outperformed similar techniques in literature in terms of classiﬁcation accuracy by 15%, 1%, 2%, and 3% respectively. CKNN and KGRNN tech- nique proved to have less computational complexity compared to the standard KNN and GRNN algorithm. Employing data pre-processing techniques such as feature extraction and missing value removal improved the classiﬁcation accuracy of machine learning techniques by more than 11% in most instances.

@phdthesis{192,
author = {Moeketsi Ndaba and Anban Pillay and Absalom Ezugwu},
title = {A Comparative Study of Machine Learning Techniques for Classifying Type II Diabetes Mellitus},
abstract = {Diabetes is a metabolic disorder that develops when the body does not make enough insulin or is not able to use insulin effectively. Accurate and early detection of diabetes can aid in effective management of the disease. Several machine learning techniques have shown promise as cost ef- fective ways for early diagnosis of the disease to reduce the occurrence of health complications arising due to delayed diagnosis. This study compares the efﬁcacy of three broad machine learning approaches; viz. Artiﬁcial Neural Networks (ANNs), Instance-based classiﬁcation technique, and Statistical Regression to diagnose type II diabetes. For each approach, this study proposes novel techniques that extend the state of the art. The new techniques include Artiﬁcial Neural Networks hybridized with an improved K-Means clustering and a boosting technique; improved variants of Logistic Regression (LR), K-Nearest Neighbours algorithm (KNN), and K-Means clustering. The techniques were evaluated on the Pima Indian diabetes dataset and the results were compared to recent results reported in the literature. The highest classiﬁcation accuracy of 100% with 100% sensitivity and 100% speciﬁcity were achieved using an ensemble of the Boosting technique, the enhanced K-Means clustering algorithm (CVE-K-Means) and the Generalized Regression Neu- ral Network (GRNN): B-KGRNN. A hybrid of CVE-K-Means algorithm and GRNN (KGRNN) achieved the best accuracy of 86% with 83% sensitivity. The improved LR model (LR-n) achieved the highest classiﬁcation accuracy of 84% with 72% sensitivity. The new multi-layer percep- tron (MLP-BPX) achieved the best accuracy of 82% and 72% sensitivity. A hybrid of KNN and CVE-K-Means (CKNN) technique achieved the best accuracy of 81% and 89% sensitivity. CVE- K-Means technique achieved the best accuracy of 80% and 61% sensitivity. The B-KGRNN, KGRNN, LR-n, and CVE-K-Means technique outperformed similar techniques in literature in terms of classiﬁcation accuracy by 15%, 1%, 2%, and 3% respectively. CKNN and KGRNN tech- nique proved to have less computational complexity compared to the standard KNN and GRNN algorithm. Employing data pre-processing techniques such as feature extraction and missing value removal improved the classiﬁcation accuracy of machine learning techniques by more than 11% in most instances.},
year = {2018},
volume = {MSc},
}

Waltham, M. ., Moodley, D. ., & Pillay, A. . (2018). Q-Cog: A Q-Learning Based Cognitive Agent Architecture for Complex 3D Virtual Worlds. Durban University.

Abstract BibTex Entry BibTeX Download PDF

Q-Cog copy.pdf

Intelligent cognitive agents requiring a high level of adaptability should contain min- imal initial data and be able to autonomously gather new knowledge from their own experiences. 3D virtual worlds provide complex environments in which autonomous software agents may learn and interact. In many applications within this domain, such as video games and virtual reality, the environment is partially observable and agents must make decisions and react in real-time. Due to the dynamic nature of virtual worlds, adaptability is of great importance for virtual agents. The Reinforce- ment Learning paradigm provides a mechanism for unsupervised learning that allows agents to learn from their own experiences in the environment. In particular, the Q- Learning algorithm allows agents to develop an optimal action-selection policy based on their environment experiences. This research explores the potential of cognitive architectures utilizing Reinforcement Learning whereby agents may contain a library of action-selection policies within virtual environments. The proposed cognitive archi- tecture, Q-Cog, utilizes a policy selection mechanism to develop adaptable 3D virtual agents. Results from experimentation indicates that Q-Cog provides an eﬀective basis for developing adaptive self-learning agents for 3D virtual worlds.

@phdthesis{190,
  author = {Michael Waltham and Deshen Moodley and Anban Pillay},
  title = {Q-Cog: A Q-Learning Based Cognitive Agent  Architecture for Complex 3D Virtual Worlds},
  abstract = {Intelligent cognitive agents requiring a high level of adaptability should contain min- imal initial data and be able to autonomously gather new knowledge from their own experiences. 3D virtual worlds provide complex environments in which autonomous software agents may learn and interact. In many applications within this domain, such as video games and virtual reality, the environment is partially observable and agents must make decisions and react in real-time. Due to the dynamic nature of virtual worlds, adaptability is of great importance for virtual agents. The Reinforce- ment Learning paradigm provides a mechanism for unsupervised learning that allows agents to learn from their own experiences in the environment. In particular, the Q- Learning algorithm allows agents to develop an optimal action-selection policy based on their environment experiences. This research explores the potential of cognitive architectures utilizing Reinforcement Learning whereby agents may contain a library of action-selection policies within virtual environments. The proposed cognitive archi- tecture, Q-Cog, utilizes a policy selection mechanism to develop adaptable 3D virtual agents. Results from experimentation indicates that Q-Cog provides an eﬀective basis for developing adaptive self-learning agents for 3D virtual worlds.},
  year = {2018},
  volume = {MSc},
  publisher = {Durban University},
}

Dzitiro, J. ., Jembere, E. ., & Pillay, A. . (2018). A DeepQA Based Real-Time Document Recommender System. In Southern Africa Telecommunication Networks and Applications Conference (SATNAC) 2018. South Africa: SATNAC.

Abstract BibTex Entry BibTeX Download PDF

SATNAC 2018.pdf

Recommending relevant documents to users in real- time as they compose their own documents differs from the traditional task of recommending products to users. Variation in the users’ interests as they work on their documents can undermine the effectiveness of classical recommender system techniques that depend heavily on off-line data. This necessitates the use of real-time data gathered as the user is composing a document to determine which documents the user will most likely be interested in. Classical methodologies for evaluating recommender systems are not appropriate for this problem. This paper proposed a methodology for evaluating real-time document recommender system solutions. The proposed method- ology was then used to show that a solution that anticipates a user’s interest and makes only high conﬁdence recommendations performs better than a classical content-based ﬁltering solution. The results obtained using the proposed methodology conﬁrmed that there is a need for a new breed of recommender systems algorithms for real-time document recommender systems that can anticipate the user’s interest and make only high conﬁdence recommendations.

@{189,
  author = {Joshua Dzitiro and Edgar Jembere and Anban Pillay},
  title = {A DeepQA Based Real-Time Document Recommender System},
  abstract = {Recommending relevant documents to users in real- time as they compose their own documents differs from the traditional task of recommending products to users. Variation in the users’ interests as they work on their documents can undermine the effectiveness of classical recommender system techniques that depend heavily on off-line data. This necessitates the use of real-time data gathered as the user is composing a document to determine which documents the user will most likely be interested in. Classical methodologies for evaluating recommender systems are not appropriate for this problem. This paper proposed a methodology for evaluating real-time document recommender system solutions. The proposed method- ology was then used to show that a solution that anticipates a user’s interest and makes only high conﬁdence recommendations performs better than a classical content-based ﬁltering solution. The results obtained using the proposed methodology conﬁrmed that there is a need for a new breed of recommender systems algorithms for real-time document recommender systems that can anticipate the user’s interest and make only high conﬁdence recommendations.},
  year = {2018},
  journal = {Southern Africa Telecommunication Networks and Applications Conference (SATNAC) 2018},
  pages = {304-309},
  month = {02/09-05/09},
  publisher = {SATNAC},
  address = {South Africa},
}

Price, C. S. ., Moodley, D. ., & Pillay, A. . (2018). Dynamic Bayesian decision network to represent growers’ adaptive pre-harvest burning decisions in a sugarcane supply chain. In Proceedings of the Annual Conference of the South African Institute of Computer Scientists and Information Technologists (SAICSIT ’18). New York NY: ACM. Retrieved from https://dl.acm.org/citation.cfm?id=3278681

Abstract BibTex Entry BibTeX Download

Sugarcane growers usually burn their cane to facilitate its harvesting and transportation. Cane quality tends to deteriorate after burning, so it must be delivered as soon as possible to the mill for processing. This situation is dynamic and many factors, including weather conditions, delivery quotas and previous decisions taken, affect when and how much cane to burn. A dynamic Bayesian decision network (DBDN) was developed, using an iterative knowledge engineering approach, to represent sugarcane growers’ adaptive pre-harvest burning decisions. It was evaluated against five different scenarios which were crafted to represent the range of issues the grower faces when making these decisions. The DBDN was able to adapt reactively to delays in deliveries, although the model did not have enough states representing delayed delivery statuses. The model adapted proactively to rain forecasts, but only adapted reactively to high wind forecasts. The DBDN is a promising way of modelling such dynamic, adaptive operational decisions.

@{181,
  author = {C. Sue Price and Deshen Moodley and Anban Pillay},
  title = {Dynamic Bayesian decision network to represent growers’ adaptive pre-harvest burning decisions in a sugarcane supply chain},
  abstract = {Sugarcane growers usually burn their cane to facilitate its harvesting and transportation.  Cane quality tends to deteriorate after burning, so it must be delivered as soon as possible to the mill for processing.  This situation is dynamic and many factors, including weather conditions, delivery quotas and previous decisions taken, affect when and how much cane to burn.  A dynamic Bayesian decision network (DBDN) was developed, using an iterative knowledge engineering approach, to represent sugarcane growers’ adaptive pre-harvest burning decisions.  It was evaluated against five different scenarios which were crafted to represent the range of issues the grower faces when making these decisions.  The DBDN was able to adapt reactively to delays in deliveries, although the model did not have enough states representing delayed delivery statuses.  The model adapted proactively to rain forecasts, but only adapted reactively to high wind forecasts.   The DBDN is a promising way of modelling such dynamic, adaptive operational decisions.},
  year = {2018},
  journal = {Proceedings of the Annual Conference of the South African Institute of Computer Scientists and Information Technologists (SAICSIT &#039;18)},
  pages = {89-98},
  month = {26/09-28/09},
  publisher = {ACM},
  address = {New York NY},
  isbn = {978-1-4503-6647-2},
  url = {https://dl.acm.org/citation.cfm?id=3278681},
}