• Sep 2006 - Present: Senior Lecturer in Secure Systems & Software Development, School of Computing, Information Technology and Engineering; Programme Leader for MSc Internet Systems Engineering
• Sep 2005 - Sep 2006: Part - time lecturer and Research Fellow in University of East London, with responsibility in developing Diagnostic Feedback for Virtual Learning Environment using Neural Networks.
• 2005: PhD from Leeds Metropolitan University
• Sep 2003 - Sep 2005: Part-time lecturer and Research Assistant at Leeds Metropolitan University, working on the improvement and development of connectionist language parser
• 1999: Graduated with first class honours in Electronics Engineering and Computing from The Nottingham Trent University

My main area of interest is in the field of Artificial Neural Networks (ANNs) and their applications in pattern recognition, natural langauge processing and intelligence data analysis.

There are many learning methods in the field of Artificial Neural Networks (ANNs). Although the elusive goal of ANNs is to emulate biological, including human learning, depending on the application, one learning rule may be more suitable than another. But the choice is not always clear-cut, despite some fundamental constraints, such as whether the learning is supervised or unsupervised. My research interest centres on addressing the ANN learning styles selection problem by proposing a new adaptive self-optimising reinforcement learning style. In broader sense, my interest lies in the field of artificial neural network (ANN) techniques and applications for pattern recognition and machine learning.

During my PhD, I have developed a new self-optimising reinforcement learning algorithm, known as snap-drift, when incorporated into a modular neural network system, is capable of rapidly adapting to discover provisional solutions that meet criteria imposed by a changing environment. This is analogous to humans optimising selection according to the options available in the surrounding environment.

• MSc Information Technology
• MSc Internet Systems Engineering
• MSc Data Mining & Knowledge Management

• SDM033 - Data Mining
• SD1042 - Introduction to Software Development
• SD0002 - Mathematics

Keynotes

• Lee, S. W. (2008). "Snap-Drift Algorithm". Doctoral Symposium on Research in Computer Science, University of Central Punjab, August 9 -10.

Book Chapters: 

• Lee, S. W.; D. Palmer-Brown; J. Tepper and C. M. Roadknight. (2002). “Performance-guided Neural Network for Rapidly Self-Organising Active Network Management.” In Soft Computing Systems: Design, Management and Applications, A. Abraham J. Ruiz-del-Solar and M. Köppen (Eds.). IOS Press, Amsterdam, pp. 21 – 31.

Journal Papers:

• Palmer-Brown, D.; S. W. Lee.; Draganova, C. and Kang, M.(2009). "Modal Learning Neural Networks." WSAES Transactions on Computers, vol. 8, no. 2, pp. 222 - 236.
• Palmer-Brown, D.; S. W. Lee. (2005). “Continuous Reinforced Snap-Drift Learning in a Neural Architecture for Proxylet Selection in Active Computer Networks.” (Invited paper). International Journal on Simulation: Systems, Science and Technology, vol. 6, no. 9, pp. 11 – 21.  
• Donelan, H.; C. Pattinson; D. Palmer-Brown and S. W. Lee. (2005). “The Analysis of Network Manager’s Behaviour using a Self-Organising Neural Networks.” (Invited paper). International Journal on Simulation: Systems, Science and Technology, vol. 6, no. 9, pp. 22 – 32.
• Lee, S. W.; D. Palmer-Brown and C. M. Roadknight. (2004). “Performance-guided Neural Network for Rapidly Self-Organising Active Network Management.” Neurocomputing, Elsevier Science, Netherlands, Vol. 61, pp. 5 – 20.

Conference Papers:

• Xavier, M.; Dodds, S. and Lee, S. W. (2010). "Closed-Loop Control using a Backpropagation Algorithm: A Practicable Approach for Energy Loss Minimisation in Electrical Drives". 5th Conference in Avances in Computing and Technology (London, United Kingdom, 27th Jan), pp. 72 - 78.
• Beqiri, E.; Lee. S. W.; Draganova, C. and Palmer-Brown, D. (2010). "A Neural Network Approach for Intrusion Detection Systems", 5th Conference in Avances in Computing and Technology (London, United Kingdom, 27th Jan), pp. 209 -217.
• Draganova, C.; Palmer-Brown, D. and Lee, S. W. (2009). " Guided Learning via Diagnostic Feedback to Question Responses", The 14th ACM–SIGCSE Annual Conference on Innovation and Technology in Computer Science Education (ITiCSE'09) (Paris, France, 3rd - 8th July).
• Palmer-Brown, D.; Draganova, C. and Lee, S.W. (2009). " Snap-Drift Neural Network for Selecting Student Feedback", The International Joint Conference on Neural Networks (IJCNN’2009) (Atlanta, Georgia, 14th - 19 June).
• Lee, S. W.; Palmer-Brown, D; Draganova, C.; Kretsis, M. and Preston, P. (2009). "Question Response Grouping for on-line Diagnostic Feedback", 4th Conference in Advances in Computing and Technology (London, United Kingdom, 27th January).
• Walcott, T.H; Palmer-Brown,D and Lee, S.W. (2008). "Creating Intelligent Markets for SMEs using the Snap-Drift Algorithm: A Higher Education College Perspective", 11th Multiconference on Information Society.
• Walcott, T.H; Palmer-Brown,D and Lee, S.W. (2008). "Early SME Market Prediction using USDNN",In Proceedings of the International Conference of Computational Intelligence and Intelligent Systems (ICCIIS'2008)(London,United Kingdom, July 2-4,2008).
• Lee, S. W.; D. Palmer-Brown and C. Draganova (2008). “Diagnostic Feedback by Snap-drift Question Response Grouping”, In proceedings of  The 9th WSEAS International Conference on Neural Networks (NN'08) (Sofia, Bulgaria, May 2-4, 2008).
• Palmer-Brown, D.; M. Kang and S. W. Lee (2008). “Meta-Adaptation: Neurons that Change their Mode”, In proceedings of  The 9th WSEAS International Conference on Neural Networks (NN'08) (Sofia, Bulgaria, May 2-4, 2008).
• Lee, S. W. and D. Palmer-Brown (2007). "Feature Discovery in Speech using Snap-Drift Neural Networks", 2nd Conference in Advances in Computing and Technology (London, United Kingdom, 23th January).
• Ekpenyong, F.; A. Brimicombe; D. Palmer-Brown and S. W. Lee (2007). " Automated Updating Of Road Network Databases: Road Segment Grouping Using Snap-Drift Neural Network", 2nd Conference in Advances in Computing and Technology (London, United Kingdom, 23th January).
• Walcott T.;D.Palmer-Brown; G. Williams; H. Mouratidis and S. W. Lee (2007). " An Assessment Of Neural Network Algorithms That Could Aid Sme Survival", 2nd Conference in Advances in Computing and Technology (London, United Kingdom, 23th January).
• Lee, S. W. and D. Palmer-Brown (2006). "" International Conference on Artificial Neural Networks (ICANN'2006) (Athen, Greece, 10th - 14th September 2006), S. Kollias et al. (Eds.): ICANN 2006, Part II, LNCS 4132, pp. 952 – 962.
• Lee, S. W. and D. Palmer-Brown (2006). “.” 1st Conference in Advances in Computing and Technology (London, United Kingdom, 24th January), pp. 42 - 47.
• Lee, S. W. and D. Palmer-Brown (2005). “Phrase Recognition using Snap-Drift Learning Algorithm.” The International Joint Conference on Neural Networks (IJCNN’2005) (Montreal, Canada, 31st July – 4th August), Vol. 1, pp. 588-592.
• Lee, S. W.; D. Palmer-Brown and C. M. Roadknight. (2004). “Snap-drift: Performance-guided Neural Network for Continuous Learning.” Special House of Commons’ Reception for Younger Researchers, House of Commons (London, UK, 15th September).
• Lee, S. W.; D. Palmer-Brown and C. M. Roadknight. (2004). “Continuous Reinforced Snap-Drift Learning in a Neural Architecture for Proxylet Selection in Active Computer Networks.” (Invited paper). In Proceedings of The 18th European Simulations Multiconference (ESM’2004) (Magdeburg, Germany, 13th – 16th June), pp. 136 – 142.
• Donelan, H.; C. Pattinson; D. Palmer-Brown and S. W. Lee. (2004). “The Analysis of Network Manager’s Behaviour using a Self-Organising Neural Networks.” In Proceedings of The 18th European Simulations Multiconference (ESM’2004) (Magdeburg, Germany, 13th – 16th June), pp. 111 – 116.
• Lee, S. W.; D. Palmer-Brown and C. M. Roadknight. (2004). “Reinforced Snap-Drift Learning for Proxylet Selection in Active Computer Networks.” In Proceedings of the International Joint Conference on Neural Networks (IJCNN’2004) (Budapest, Hungary, 25th – 29th July), Vol. 2, pp. 1545 – 1550.
• Palmer-Brown, D.; S. W. Lee; J. Tepper and C. M. Roadknight. (2003). “Fast Learning Neural Nets with Adaptive Learning Styles.” (Invited Paper). In Proceedings of the 17th European Simulation Multiconference (ESM'2003) (The Nottingham Trent University, Nottingham, UK, 9th - 11th June), pp. 118 – 123.
• Lee, S. W.; D. Palmer-Brown; J. A. Tepper and C. M. Roadknight. (2003). “Snap-Drift: Real-time, Performance-guided Learning.” In Proceedings of the International Joint Conference on Neural Networks (IJCNN’2003) (Portland, Oregon, 20th - 24th July), Vol. 2, pp. 1412 – 1416.
• Lee, S. W.; D. Palmer-Brown; J. Tepper and C.M. Roadknight. (2002). “Performance-guided Neural Network for Self-Organising Network Management.” In Proceedings of London Communication Symposium (LCS'2002) (University College London, London, UK, 9th – 10th September), pp. 269 – 272.
• Lee, S. W. (2003). “P-ART: Performance-guided Neural Network.” End of PhD Research Seminar (BTexact Technologies, Ipswich, UK, 18th December).
• Lee, S. W. (2003). “PART: Performance-guided ART for Active Network Management.” 2nd Annual Researcher Conference (Leeds Metropolitan University, Leeds, UK, 14th – 15th July)
• Lee, S. W. (2002). “Performance-guided Neural Networks for Self-Organising Network Management.” The Annual Research Student Conference (The Nottingham Trent University, Nottingham, UK
• Lee, S. W. (2002). “Performance-guided Adaptive Resonance for Self-Organising Network Management.” 1st Annual Researcher Conference (Leeds Metropolitan University, Leeds, UK, 25th – 26th March).

Collaborations and Fundings:

• 2000 - 2003, CASE Studentship, British Telecom Research Labs, £19,000, Neural Nets for Active Network Adaptation. D Palmer-Brown, Sin Wee Lee
• 2007 - 2008, Higher Education Funding Council for England (HEFCE) E-Learning Fund, £50,000, Diagnostic Feedback tool for Virtual Learning Environment, Dominic Palmer-Brown, Sin Wee Lee
• 2009 - 2010, LEO fund, UEL Learning, Teaching and Assessment, £28,700, Intelligent Feedback to enhanced Student's learning experience. Sin Wee Lee, Chrisina Draganova & Aaron Kans.

Applications of Snap-drift Algorithm:

• Reinforced Snap-Drift Learning for Proxylet Selection in Active Computer Networks
• Data Analysis of Network Manager’s Behaviour
• Feature discovery and clustering of speech waveforms from non-stammering and stammering speakers
• Phrase recognition in a Connectionist Language Parser
• Cluster Analysis on Iris Data using Snap-drift Algorithm
• Diagnostic Feedback tool for Virtual Learning Environment
• Feature discovery and clustering of Geographical Information
• Neural Networks for SME Business Intelligence

• Reviewer, Software Quality Journal (2006 - )
• Reviewer, Journal of Neurocomputing (2007 - )
• Reviewer, Journal of Neural Computing and Applications (2010 - )
• Organising Committee for AC&T (2009 - )
• Organsing Committee for EANN 2009
• Programme Committe for International Conference on Neural Computation (ICNC 2009)
• PhD examining
• PhD supervisions
• Member of the School's Research Degree Sub-Committee
• Member of the School's Quality Sub-Committee
• Member of the School's Collaborative Sub-Committee
• Academic Link Tutor for International Collaborations (Malaysia and Singapore)

(A) Reinforced Snap-Drift Learning for Proxylet Selection in Active Computer Networks

Sin Wee Lee, Dominic Palmer-Brown and Christopher Roadknight (Collaborative research project with BTexact Research Laboratories)

In this project, a new continuous learning method is applied to the problem of optimizing the selection of services in response to user requests in an active computer network simulation environment. The learning is an enhanced version of the ‘snap-drift’ algorithm, which employs the complementary concepts of fast, minimalist learning (snap) and slower drift (towards the input patterns) learning, in a non-stationary environment where new patterns are continually introduced. Snap is based on Adaptive Resonance Theory (ART), and drift on Learning Vector Quantisation (LVQ). The new algorithm swaps its learning style between the two modes of self-organisation when declining performance levels are received, but maintains the same learning style during episodes of improved performance. Performance updates occur at the end of each epoch. Reinforcement also occurs by maintaining successful adaptations, since learning is enabled with a probability that increases with declining performance. The method is capable of rapidly re-learning and is used in the design of a modular neural network system, Performance-guided Adaptive Resonance Theory (P-ART).

(B) Data Analysis of Network Manager’s Behaviour

Helen Donelan, Colin Pattinson, Dominic Palmer-Brown and Sin Wee Lee (Collaborative research project with Computer Communications Research Group, Leeds Metropolitan University)

The aim of this research is to provide a tool for exposing trainee network managers, through a life like situation, where both normal network operation and fault scenario could be simulated in order to train the network manager. P-ART is adapted and implemented to analyse the interaction data to enable an assessment of the capabilities of the network manager as well as the understanding of how the network management tasks are being approached. The results show that P-ART can be used to uncover hidden patterns in user behaviour and provide novel insights into that behaviour. The output formed by P-ART is used to compare instances of good and bad practice and reveal patterns embedded within the data that are difficult to recognise.

(C) Feature discovery and clustering of speech waveforms from non-stammering and stammering speakers

Sin Wee Lee, Dominic Palmer-Brown, Nicole Whitworth and Monica Bray (Collaborative research project with Phonetics Research Group, Leeds Metropolitan University)

P-ART is proposed to act as a tool for feature discovery and clustering of speech waveforms from non-stammering and stammering speakers. The learning algorithm is an unsupervised version of snap-drift which employs the complementary concepts of fast, minimalist learning (snap) & slow drift (towards the input pattern) learning. The Snap-Drift Neural Network (SDNN) is toggled between snap and drift modes on successive epochs. The speech waveforms are drawn from a phonetically annotated corpus, which facilitates phonetic interpretation of the classes of patterns discovered by the SDNN.

(D) Phrase recognition in a Connectionist Language Parser

Sin Wee Lee and Dominic Palmer-Brown

The application involves phrase recognition using a set of phrases from the Lancaster Parsed Corpus (LPC) (Garside, 1987). The learning algorithm is the classifier version of snap-drift. The twin modes of minimalist learning (snap) and slow drift towards the input pattern are applied alternately. Each neuron of the Snap-Drift Neural Network (SDNN) swaps between snap and drift modes when declining performance is indicated on that particular node, so that each node has its learning mode toggled independently of the other nodes. Learning on each node is also reinforced by enabling learning with a probability that decreases with increasing performance. The simulations demonstrate that learning is stable, and the results have consistently shown similar classification performance and advantages in terms of speed in comparison with a Multilayer Perceptron (MLP) and backpropagation neural networks applied to the same problem.

(E) Cluster Analysis on Iris Data using Snap-drift Algorithm

Sin Wee Lee and Dominic Palmer-Brown

The unsupervised snap-drift algorithm will be used for clustering the iris data make famous by Fisher (1936) that has been widely used in cluster analysis. It consists of three species, namely, Iris Setosa, Iris Virginia and Iris Versicolor. Each species contains fifty data points, and each data point has 4 dimensions, which represents sepal length, sepal width, petal length and petal width. Unsupervised snap-drift, used in the feature discovery of phonetics speech is used to do cluster analysis of these real-value iris data, preliminary results shows that after a small number of epochs, the system gets 98% reliably correct classification, using about 10 output categorie.

(F) Diagnostic Feedback tool for Virtual Learning Environment

Sin Wee Lee, Dominic Palmer-Brown, Chrisina Draganova

Snap-drift Neural Networks (SDNNs), which employ the complementary concepts of fast, minimalist (snap) learning and slow (drift towards the input pattern) learning is used for the automatic generation of diagnostic feedback for students in an e-learning environment. By discovering the hidden features in the students' responses, such as those to multiple choice questions, typical error patterns, and by taking into account levels of question difficulty, and linguistic (phrasal and lexical) analysis/recognition in short answers, the SDNN will select the most appropriate feedback mix from a corpus. This system will be tested on 'live' students to facilitate automatic diagnosis of deficits in areas of knowledge and to provide specific guidance.

• CELEST Technology Website - Articles focus on analysis and applications of neural network systems, particularly on Adaptive Resonance Theory (ART).
• The Adaptive Resonance Theory (ART) clearinghouse with an extensive bibliography on ART.