An incremental meta-cognitive-based scaffolding fuzzy neural network

The idea of meta-cognitive learning has enriched the landscape of evolving systems, because it emulates three fundamental aspects of human learning: what-to-learn; how-to-learn; and when-to-learn. However, existing meta-cognitive algorithms still exclude Scaffolding theory, which can realize a plug-and-play classifier. Consequently, these algorithms require laborious pre- and/or post-training processes to be carried out in addition to the main training process. This paper introduces a novel meta-cognitive algorithm termed GENERIC-Classifier (gClass), where the how-to-learn part constitutes a synergy of Scaffolding Theory - a tutoring theory that fosters the ability to sort out complex learning tasks, and Schema Theory - a learning theory of knowledge acquisition by humans. The what-to-learn aspect adopts an online active learning concept by virtue of an extended conflict and ignorance method, making gClass an incremental semi-supervised classifier, whereas the when-to-learn component makes use of the standard sample reserved strategy. A generalized version of the Takagi-Sugeno Kang (TSK) fuzzy system is devised to serve as the cognitive constituent. That is, the rule premise is underpinned by multivariate Gaussian functions, while the rule consequent employs a subset of the non-linear Chebyshev polynomial. Thorough empirical studies, confirmed by their corresponding statistical tests, have numerically validated the efficacy of gClass, which delivers better classification rates than state-of-the-art classifiers while having less complexity.

Instructional strategies integrating cognitive style construct : a meta-knowledge processing model

The overarching goal of this dissertation was to evaluate the contextual components of instructional strategies for the acquisition of complex programming concepts. A meta-knowledge processing model is proposed, on the basis of the research findings, thereby facilitating the selection of media treatment for electronic courseware. When implemented, this model extends the work of Smith (1998), as a front-end methodology, for his glass-box interpreter called Bradman, for teaching novice programmers. Technology now provides the means to produce individualized instructional packages with relative ease. Multimedia and Web courseware development accentuate a highly graphical (or visual) approach to instructional formats. Typically, little consideration is given to the effectiveness of screen-based visual stimuli, and curiously, students are expected to be visually literate, despite the complexity of human-computer interaction. Visual literacy is much harder for some people to acquire than for others! (see Chapter Four: Conditions-of-the-Learner) An innovative research programme was devised to investigate the interactive effect of instructional strategies, enhanced with text-plus-textual metaphors or text-plus-graphical metaphors, and cognitive style, on the acquisition of a special category of abstract (process) programming concept. This type of concept was chosen to focus on the role of analogic knowledge involved in computer programming. The results are discussed within the context of the internal/external exchange process, drawing on Ritchey's (1980) concepts of within-item and between-item encoding elaborations. The methodology developed for the doctoral project integrates earlier research knowledge in a novel, interdisciplinary, conceptual framework, including: from instructional science in the USA, for the concept learning models; British cognitive psychology and human memory research, for defining the cognitive style construct; and Australian educational research, to provide the measurement tools for instructional outcomes. The experimental design consisted of a screening test to determine cognitive style, a pretest to determine prior domain knowledge in abstract programming knowledge elements, the instruction period, and a post-test to measure improved performance. This research design provides a three-level discovery process to articulate: 1) the fusion of strategic knowledge required by the novice learner for dealing with contexts within instructional strategies 2) acquisition of knowledge using measurable instructional outcome and learner characteristics 3) knowledge of the innate environmental factors which influence the instructional outcomes This research has successfully identified the interactive effect of instructional strategy, within an individual's cognitive style construct, in their acquisition of complex programming concepts. However, the significance of the three-level discovery process lies in the scope of the methodology to inform the design of a meta-knowledge processing model for instructional science. Firstly, the British cognitive style testing procedure, is a low cost, user friendly, computer application that effectively measures an individual's position on the two cognitive style continua (Riding & Cheema,1991). Secondly, the QUEST Interactive Test Analysis System (Izard,1995), allows for a probabilistic determination of an individual's knowledge level, relative to other participants, and relative to test-item difficulties. Test-items can be related to skill levels, and consequently, can be used by instructional scientists to measure knowledge acquisition. Finally, an Effect Size Analysis (Cohen,1977) allows for a direct comparison between treatment groups, giving a statistical measurement of how large an effect the independent variables have on the dependent outcomes. Combined with QUEST's hierarchical positioning of participants, this tool can assist in identifying preferred learning conditions for the evaluation of treatment groups. By combining these three assessment analysis tools into instructional research, a computerized learning shell, customised for individuals' cognitive constructs can be created (McKay & Garner,1999). While this approach has widespread application, individual researchers/trainers would nonetheless, need to validate with an extensive pilot study programme (McKay,1999a; McKay,1999b), the interactive effects within their specific learning domain. Furthermore, the instructional material does not need to be limited to a textual/graphical comparison, but could be applied to any two or more instructional treatments of any kind. For instance: a structured versus exploratory strategy. The possibilities and combinations are believed to be endless, provided the focus is maintained on linking of the front-end identification of cognitive style with an improved performance outcome. My in-depth analysis provides a better understanding of the interactive effects of the cognitive style construct and instructional format on the acquisition of abstract concepts, involving spatial relations and logical reasoning. In providing the basis for a meta-knowledge processing model, this research is expected to be of interest to educators, cognitive psychologists, communications engineers and computer scientists specialising in computer-human interactions.

Greater cognitive effort for better learning : tailoring an instructional design for learners with different levels of knowledge and motivation

The capacity limitation of working memory is a widely recognised determinant of human learning. A cognitive load exceeding the capacity hampers learning. Cognitive load can be controlled by tailoring an instructional design to levels of learner prior knowledge. However, such as design does not necessarily motivate to use the available capacity for better learning. The present review examines literatures on the effects of instructional design, motivation, emotional state, and expertise level on cognitive load and cognitive effort, which ultimately affect working memory performance and learning. This examination suggests further studies on the effects of motivation and negative emotional states on the use of working memory. Prospective findings would help better explain and predict individual differences in the use of working memory for cognitive learning and task performance.

Developing english skills through reflective portfolios in Hong Kong universities

This research concerns the use of portfolios by teachers of English (L2) to assist non-native speakers in Hong Kong universities to complete their studies in English. Portfolios as an English learning tool have yet to win converts from the ranks of language teachers in Hong Kong chiefly because of concerns about reliability and fairness. Two recent initiatives in Hong Kong have, however, prompted a reappraisal of the place of portfolios in English language learning. They include the use of learning portfolios in secondary school and ePortfolios by university students for learning and employment purposes.

As an English (L2) teacher of many years, I initiated my research to investigate the experiences of seven university students in Hong Kong in using reflective portfolios for English learning. Three research questions framed my research: 1) in what ways can reflective portfolios impact on L2 learning strategies? 2) what are the effects of reflective portfolios on progress in L2 acquisition as perceived by students? 3) what are the perceptions of university students towards reflective portfolios as a method of L2 learning?

To gain a holistic understanding of the complex phenomena under scrutiny, a case study methodology and grounded theory were utilised, the former to organise and generate qualitative data, and the latter to analyse data from three sources provided by the seven participating students: semi-structured interviews, portfolio artefacts, and weekly learning diaries.

There were two levels of data analysis. For the first level, analysis focused on coded data from portfolio artefacts, diary entries and interview transcripts as reported by students. The second level involved analysis from the Confucian and sociocultural perspectives. I pursued interpretation and continuous refinement of the data by using techniques drawn from grounded theory. The findings revealed that students generally employed a wide spread of L2 learning strategies in the cognitive, meta-cognitive, and socio-affective domain, reported increased awareness of effective language strategies, and considered portfolios a means of supporting time management and record-keeping, and a site for extended writing practice through reflection.

The findings suggest that students display a cyclical, context-specific shift in learning conception from quantitative to qualitative. Connected to this is students’ apparent ability to formulate strategic responses to externally imposed demands. It is found that such responses are culturally triggered, underpinned by Confucian beliefs. Although the Confucian tradition emphasises respect for established authority, the findings point to students’ creative re-configuration of mental schemata to engender change in role enactment and power relations, with the portfolio as a mediating tool of their experiences.

Based on the findings, I argue that my research has addressed the three research questions and contributed to two crucial aspects of L2 learning. The first pertains to the need for a balanced view of individual effort and social context in second language acquisition, corroborating the significant link between context and learner engagement. Another contribution centres on an enhanced understanding of the relationship between portfolios, reflection and L2, where students’ diaries in English and portfolio artefacts enable them to engage in critical reflection and to identify strategies for L2 improvement.

Haptically enable interactive virtual assembly training system development and evaluation

Virtual training systems are attracting paramount attention from the manufacturing industries due to their potential advantages over the conventional training practices. Significant cost savings can be realized due to the shorter times for the development of different training-scenarios as well as reuse of existing designed engineering (math) models. In addition, use of computer based virtual reality (VR) training systems can shorten the time span from computer aided product design to commercial production due to non-reliance on the hardware parts for training. Within the aforementioned conceptual framework, a haptically enabled interactive and immersive virtual reality (HIVEx) system is presented. Unlike existing VR systems, the presented idea tries to imitate real physical training scenarios by providing comprehensive user interaction, constrained within the physical limitations of the real world. These physical constrains are imposed by the haptics devices in the virtual environment. As a result, in contrast to the existing VR systems that are capable of providing knowledge generally about assembly sequences only, the proposed system helps in cognitive learning and procedural skill development as well, due to its high physically interactive nature.

Haptics and virtual reality in modern manufacturing

Product assembly is one of the most studied processes in modern manufacturing. In recent years a number of computer-based virtual reality systems have been proposed, developed and adopted by the manufacturing industries. Such systems have major advantages over conventional training practices for product assembly. Significant cost savings can be realized due to shorter training-scenario development times and reuse of existing engineering math models. In addition, the time span from the product design to full production can be shortened due to non-reliance on actual components and subsystems for training. Such training systems are effective if the knowledge required to be transferred is just process sequence such as assembly sequence. However, knowledge transfer for procedural and cognitive learning as well as skills development is very limited, due to the lack of user interactivity and immersion.

This talk will focus on a research technology platform where haptics and virtual reality are integrated to create an effective environment for production assembly operators’ training. In this system virtual reality provides the grounds for realistic visualization, as well as immersion, whereas haptics enforces physical constraints within the virtual world generating the feelings of realistic interaction, making it accessible for formal learning and better understanding during task performance.

The developed research technology platform imitates real physical training scenarios by providing comprehensive user interaction, constrained within the physical limitations of the real world. Through the utilization of a haptics device, providing realistic force feedback, users are able to engage in product assembly training with a stronger sense of ‘reality’.

The influence of confidence in memory on checking behaviours

Recent theories of Obsessive Compulsive Disorder have suggested that meta-cognitive beliefs, particular confidence in memory, may relate to the maintenance of symptoms. For example, individuals who worry that their memory is deficient may check in order to ensure that the doors are locked, regardless of their actual memory ability. While studies have related meta-memory beliefs to OCD symptoms, and have found that memory is affected by checking behaviours; to date, no experimental literature has attempted to ascertain whether lowered confidence in memory directly leads to greater checking behaviours. In the current study 46 non-clinical participants completed a battery of questionnaires measuring these constructs, before undertaking a “memory task” designed to manipulate their level of confidence in their memory. The effect of the manipulation on the level of checking on a visuomemory task was subsequently assessed. The questionnaires indicated that cognitive confidence predicted variation in obsessive-compulsive symptoms over-and-above the influence of depressive symptoms and other OCD-related beliefs. However, while confidence in memory was successfully manipulated, the group with increased memory confidence was not found to have lower checking behaviours. Limitations of the study are discussed and directions for further research are suggested.

A representation construction approach

In this chapter we lay out the principles of an approach to teaching and learning science based on student generation, negotiation and refinement of representations in a guided inquiry process. We first tell the story of how we developed this perspective, building on Chapters 1 and 2, and the research approach that led to these principles. The principles of the representation construction approach are described, then exemplified using detailed analysis of parts of classroom learning sequences on force, and substances. We then give examples of teacher responses and beliefs, and finally provide evidence of student conceptual, and meta-representational learning, from this approach.

A meta-learning approach to automatic kernel selection for support vector machines

Appropriate choice of a kernel is the most important ingredient of the kernel-based learning methods such as support vector machine (SVM). Automatic kernel selection is a key issue given the number of kernels available, and the current trial-and-error nature of selecting the best kernel for a given problem. This paper introduces a new method for automatic kernel selection, with empirical results based on classification. The empirical study has been conducted among five kernels with 112 different classification problems, using the popular kernel based statistical learning algorithm SVM. We evaluate the kernels’ performance in terms of accuracy measures. We then focus on answering the question: which kernel is best suited to which type of classification problem? Our meta-learning methodology involves measuring the problem characteristics using classical, distance and distribution-based statistical information. We then combine these measures with the empirical results to present a rule-based method to select the most appropriate kernel for a classification problem. The rules are generated by the decision tree algorithm C5.0 and are evaluated with 10 fold cross validation. All generated rules offer high accuracy ratings.

Towards insightful algorithm selection for optimisation using meta-learning concepts

In this paper we propose a meta-learning inspired framework for analysing the performance of meta-heuristics for optimization problems, and developing insights into the relationships between search space characteristics of the problem instances and algorithm performance. Preliminary results based on several meta-heuristics for well-known instances of the Quadratic Assignment Problem are presented to illustrate the approach using both supervised and unsupervised learning methods.

A framework for re-thinking learning in science from recent cognitive science perspectives

Recent accounts by cognitive scientists of factors affecting cognition imply the need to reconsider current dominant conceptual theories about science learning. These new accounts emphasize the role of context, embodied practices and narrative-based representation rather than learners’ cognitive constructs. In this paper we analyze data from a longitudinal study of primary school children’s learning to outline a framework based on these contemporary accounts, and to delineate key points of difference from conceptual change perspectives. The findings suggest this framework provides strong theoretical and practical insights into how children learn and the key role of representational negotiation in this learning. We argue that the nature and process of conceptual change can be re-interpreted in terms of the development of students’ representational resources.

Meta-learning for data summarization based on instance selection method

The purpose of instance selection is to identify which instances (examples, patterns) in a large dataset should be selected as representatives of the entire dataset, without significant loss of information. When a machine learning method is applied to the reduced dataset, the accuracy of the model should not be significantly worse than if the same method were applied to the entire dataset. The reducibility of any dataset, and hence the success of instance selection methods, surely depends on the characteristics of the dataset, as well as the machine learning method. This paper adopts a meta-learning approach, via an empirical study of 112 classification datasets from the UCI Repository [1], to explore the relationship between data characteristics, machine learning methods, and the success of instance selection method.