GYAN: A methodology for rule extraction from artificial neural networks

Artificial neural network (ANN) learning methods provide a robust and non-linear approach to approximating the target function for many classification, regression and clustering problems. ANNs have demonstrated good predictive performance in a wide variety of practical problems. However, there are strong arguments as to why ANNs are not sufficient for the general representation of knowledge. The arguments are the poor comprehensibility of the learned ANN, and the inability to represent explanation structures. The overall objective of this thesis is to address these issues by: (1) explanation of the decision process in ANNs in the form of symbolic rules (predicate rules with variables); and (2) provision of explanatory capability by mapping the general conceptual knowledge that is learned by the neural networks into a knowledge base to be used in a rule-based reasoning system. A multi-stage methodology GYAN is developed and evaluated for the task of extracting knowledge from the trained ANNs. The extracted knowledge is represented in the form of restricted first-order logic rules, and subsequently allows user interaction by interfacing with a knowledge based reasoner. The performance of GYAN is demonstrated using a number of real world and artificial data sets. The empirical results demonstrate that: (1) an equivalent symbolic interpretation is derived describing the overall behaviour of the ANN with high accuracy and fidelity, and (2) a concise explanation is given (in terms of rules, facts and predicates activated in a reasoning episode) as to why a particular instance is being classified into a certain category.

Hierarchical models for 2D presence/absence data having ambiguous zeroes: With a biogeographical case study on dingo behaviour

This dissertation is primarily an applied statistical modelling investigation, motivated by a case study comprising real data and real questions. Theoretical questions on modelling and computation of normalization constants arose from pursuit of these data analytic questions. The essence of the thesis can be described as follows. Consider binary data observed on a two-dimensional lattice. A common problem with such data is the ambiguity of zeroes recorded. These may represent zero response given some threshold (presence) or that the threshold has not been triggered (absence). Suppose that the researcher wishes to estimate the effects of covariates on the binary responses, whilst taking into account underlying spatial variation, which is itself of some interest. This situation arises in many contexts and the dingo, cypress and toad case studies described in the motivation chapter are examples of this. Two main approaches to modelling and inference are investigated in this thesis. The first is frequentist and based on generalized linear models, with spatial variation modelled by using a block structure or by smoothing the residuals spatially. The EM algorithm can be used to obtain point estimates, coupled with bootstrapping or asymptotic MLE estimates for standard errors. The second approach is Bayesian and based on a three- or four-tier hierarchical model, comprising a logistic regression with covariates for the data layer, a binary Markov Random field (MRF) for the underlying spatial process, and suitable priors for parameters in these main models. The three-parameter autologistic model is a particular MRF of interest. Markov chain Monte Carlo (MCMC) methods comprising hybrid Metropolis/Gibbs samplers is suitable for computation in this situation. Model performance can be gauged by MCMC diagnostics. Model choice can be assessed by incorporating another tier in the modelling hierarchy. This requires evaluation of a normalization constant, a notoriously difficult problem. Difficulty with estimating the normalization constant for the MRF can be overcome by using a path integral approach, although this is a highly computationally intensive method. Different methods of estimating ratios of normalization constants (N Cs) are investigated, including importance sampling Monte Carlo (ISMC), dependent Monte Carlo based on MCMC simulations (MCMC), and reverse logistic regression (RLR). I develop an idea present though not fully developed in the literature, and propose the Integrated mean canonical statistic (IMCS) method for estimating log NC ratios for binary MRFs. The IMCS method falls within the framework of the newly identified path sampling methods of Gelman & Meng (1998) and outperforms ISMC, MCMC and RLR. It also does not rely on simplifying assumptions, such as ignoring spatio-temporal dependence in the process. A thorough investigation is made of the application of IMCS to the three-parameter Autologistic model. This work introduces background computations required for the full implementation of the four-tier model in Chapter 7. Two different extensions of the three-tier model to a four-tier version are investigated. The first extension incorporates temporal dependence in the underlying spatio-temporal process. The second extensions allows the successes and failures in the data layer to depend on time. The MCMC computational method is extended to incorporate the extra layer. A major contribution of the thesis is the development of a fully Bayesian approach to inference for these hierarchical models for the first time. Note: The author of this thesis has agreed to make it open access but invites people downloading the thesis to send her an email via the 'Contact Author' function.

Towards research-based innovation and reform: Singapore education in transition

The challenges facing the Singapore education system in the new millennium are unique and unprecedented in Asia. Demands for new skills, knowledges, and flexible competencies for globalised economies and cosmopolitan cultures will require system-wide innovation and reform. But there is a dearth of international benchmarks and prototypes for such reforms. This paper describes the current Core Research Program underway at the National Institute of Education in Singapore, a multilevel analysis of Singaporean schooling, pedagogy, youth and educational outcomes. It describes student background, performance, classroom practices, student artefacts and outcomes, and student longitudinal life pathways. The case is made that a systematic focus on teachers' and students' work in everyday classroom contexts is the necessary starting point for pedagogical innovation and change. This, it is argued, can constitute a rich multidisciplinary evidence base for educational policy. (Contains 1 figure, 1 table and 3 notes.)

Valuing the impact of the teacher librarian from an evidence base

Evidence-based practice as it applies to the Library and Information (LIS) sector and in particular teacher librarians is the focus of this research investigation. The context for this research is Australian school libraries and teacher librarians. This is a research in progress and the report here will include some very early findings and lessons learned from the initial pilot study. The contributions of this research will be in developing a framework for the library and information sector with a particular application for teacher librarians. Providing meaningful evidence of work practices that demonstrate contributions to the schools goals and mission statements in conjunction with contributions to student academic, social and cultural achievements are crucial for the future of the teacher librarian.

An empirical test of causal relationships of factors affecting ICT adoption for building project management : an Indian SME case study

Purpose - Building project management (BPM) requires effective coordination and collaboration between multiple project team organisations which can be achieved by real time information flow between all participants. In the present scenario, this can be achieved by the use of information communication technologies (ICT). The purpose of this paper is to present part of a research project conducted to study the causal relationships between factors affecting ICT adoption for BPM by small and medium enterprises. Design/methodology/approach - This paper discusses structural equation modelling (SEM) analysis conducted to test the causal relationships between quantitative factors. Data for quantitative analysis were gathered through a questionnaire survey conducted in the Indian construction industry. Findings - SEM analysis results help in demonstrating that an increased and matured use of ICT for general administration within the organisation would lead to: an improved ICT infrastructure within the organisation; development of electronic databases; and a staff that is confident of using information technology (IT) tools. In such a scenario, staff would use advanced software and IT technologies for project management (PM) processes and that would lead to an increased adoption of ICT for PM processes. But, for general administration also, ICT adoption would be enhanced if the organisation is interacting more with geographically separated agencies and senior management perceives that significant benefits would accrue by adoption of ICT. All the factors are inter-related and their effect cannot be maximized in isolation. Originality/value - The results provide direction to building project managements for strategically adopting the effective use of ICT within their organisations and for BPM general.