Globally optimal on-line learning rules

We present a method for determining the globally optimal on-line learning rule for a soft committee machine under a statistical mechanics framework. This work complements previous results on locally optimal rules, where only the rate of change in generalization error was considered. We maximize the total reduction in generalization error over the whole learning process and show how the resulting rule can significantly outperform the locally optimal rule.

On-line learning of unrealizable tasks

The dynamics of on-line learning is investigated for structurally unrealizable tasks in the context of two-layer neural networks with an arbitrary number of hidden neurons. Within a statistical mechanics framework, a closed set of differential equations describing the learning dynamics can be derived, for the general case of unrealizable isotropic tasks. In the asymptotic regime one can solve the dynamics analytically in the limit of large number of hidden neurons, providing an analytical expression for the residual generalization error, the optimal and critical asymptotic training parameters, and the corresponding prefactor of the generalization error decay.

The theory of on-line learning: a statistical physics approach

In this paper we review recent theoretical approaches for analysing the dynamics of on-line learning in multilayer neural networks using methods adopted from statistical physics. The analysis is based on monitoring a set of macroscopic variables from which the generalisation error can be calculated. A closed set of dynamical equations for the macroscopic variables is derived analytically and solved numerically. The theoretical framework is then employed for defining optimal learning parameters and for analysing the incorporation of second order information into the learning process using natural gradient descent and matrix-momentum based methods. We will also briefly explain an extension of the original framework for analysing the case where training examples are sampled with repetition.

Innovation management using logical framework in hospital-based healthcare units

Innovation is part and parcel of any service in today's environment, so as to remain competitive. Quality improvement in healthcare services is a complex, multi-dimensional task. This study proposes innovation management in healthcare services using a logical framework. A problem tree and an objective tree are developed to identify and mitigate issues and concerns. A logical framework is formulated to develop a plan for implementation and monitoring strategies, potentially creating an environment for continuous quality improvement in a specific unit. We recommend logical framework as a valuable model for innovation management in healthcare services. Copyright © 2006 Inderscience Enterprises Ltd.

Towards an inter-organisational knowledge transfer framework for SMEs

This paper aims to develop a framework for SMEs to help them understand, and thus to improve, the process of knowledge exchange with their customers or suppliers. Through a review of the literature on knowledge transfer, organisational learning, social network theory and electronic networks, the key actors, key factors and their relationships in the process are identified. Finally, a framework containing all above points is proposed.

Cohesion or confusion? Towards a typology for organizational learning research

The study of organizational learning is no longer in its infancy. Since Cyert and March first introduced the notion in the early 1960s, a plethora of books and journal publications have presented their own interpretations of the meaning and significance of the term. Despite such endeavours, there is little common agreement about what organizational learning represents and how future research may build cumulatively upon the many diverse ideas articulated. The intention here is by no means to address these issues, which have been comprehensively examined elsewhere. The purpose is rather to compare and contrast approaches in order to analyse similarities and dissimilarities, together with research challenges, for each approach. This is achieved by presenting a comparative framework to categorize the literature according to (a) its prescriptive/explanatory bias and (b) in line with the level of analysis, examining whether there is a focus on the organization as a whole or upon individuals and their work communities instead. The review concludes by presenting some preliminary suggestions for cross-quadrant research. © Blackwell Publishing Ltd 2006.

Visual data mining: integrating machine learning with information visualization

Today, the data available to tackle many scientific challenges is vast in quantity and diverse in nature. The exploration of heterogeneous information spaces requires suitable mining algorithms as well as effective visual interfaces. Most existing systems concentrate either on mining algorithms or on visualization techniques. Though visual methods developed in information visualization have been helpful, for improved understanding of a complex large high-dimensional dataset, there is a need for an effective projection of such a dataset onto a lower-dimension (2D or 3D) manifold. This paper introduces a flexible visual data mining framework which combines advanced projection algorithms developed in the machine learning domain and visual techniques developed in the information visualization domain. The framework follows Shneiderman’s mantra to provide an effective user interface. The advantage of such an interface is that the user is directly involved in the data mining process. We integrate principled projection methods, such as Generative Topographic Mapping (GTM) and Hierarchical GTM (HGTM), with powerful visual techniques, such as magnification factors, directional curvatures, parallel coordinates, billboarding, and user interaction facilities, to provide an integrated visual data mining framework. Results on a real life high-dimensional dataset from the chemoinformatics domain are also reported and discussed. Projection results of GTM are analytically compared with the projection results from other traditional projection methods, and it is also shown that the HGTM algorithm provides additional value for large datasets. The computational complexity of these algorithms is discussed to demonstrate their suitability for the visual data mining framework.

A composite framework for the strategic alignment of information systems development

Information systems are corporate resources, therefore information systems development must be aligned with corporate strategy. This thesis proposes that effective strategic alignment of information systems requires information systems development, information systems planning and strategic management to be united. Literature in these areas is examined, breaching the academic boundaries which separate these areas, to contribute a synthesised approach to the strategic alignment of information systems development. Previous work in information systems planning has extended information systems development techniques, such as data modelling, into strategic planning activities, neglecting techniques of strategic management. Examination of strategic management in this thesis, identifies parallel trends in strategic management and information systems development; the premises of the learning school of strategic management are similar to those of soft systems approaches to information systems development. It is therefore proposed that strategic management can be supported by a soft systems approach. Strategic management tools and techniques frame individual views of a strategic situation; soft systems approaches can integrate these diverse views to explore the internal and external environments of an organisation. The information derived from strategic analysis justifies the need for an information system and provides a starting point for information systems development. This is demonstrated by a composite framework which enables each information system to be justified according to its direct contribution to corporate strategy. The proposed framework was developed through action research conducted in a number of organisations of varying types. This suggests that the framework can be widely used to support the strategic alignment of information systems development, thereby contributing to organisational success.

Moving from HRM to knowledge productivity:a study of whether and how organizations become learning organizations

Practitioners and academics are in broad agreement that, above all, organizations need to be able to learn, to innovate and to question existing ways of working. This thesis develops a model to take into account, firstly, what determines whether or not organizations endorse practices designed to facilitate learning. Secondly, the model evaluates the impact of such practices upon organizational outcomes, measured in terms of products and technological innovation. Researchers have noted that organizations that are committed to producing innovation show great resilience in dealing with adverse business conditions (e.g. Pavitt, 1991; Leonard Barton, 1998). In effect, such organizations bear many of the characteristics associated with the achievement of ‘learning organization’ status (Garvin, 1993; Pedler, Burgoyne & Boydell, 1999; Senge, 1990). Seven studies are presented to support this theoretical framework. The first empirical study explores the antecedents to effective learning. The three following studies present data to suggest that people management practices are highly significant in determining whether or not organizations are able to produce sustained innovation. The thesis goes on to explore the relationship between organizational-level job satisfaction, learning and innovation, and provides evidence to suggest that there is a strong, positive relationship between these variables. The final two chapters analyze learning and innovation within two similar manufacturing organizations. One manifests relatively low levels of innovation whilst the other is generally considered to be outstandingly innovative. I present the comparative framework for exploring the different approaches to learning manifested by the two organizations. The thesis concludes by assessing the extent to which the theoretical model presented in the second chapter is borne out by the findings of the study. Whilst this is a relatively new field of inquiry, findings reveal that organizations have a much stronger chance of producing sustained innovation where they manage people proactively where people process themselves to be satisfied at work. Few studies to date have presented empirical evidence to substantiate theoretical endorsements to engage in higher order learning, so this research makes an important contribution to existing literature in this field.

A prediction of outcomes: a psychological investigation into cognitive and affective development within the framework of adult literacy provision

This investigation seeks to explore the hypothesis, derived from observation and practice, that there is a strong relationship between the development of literacy skills and the growth of confidence in adult literacy students. Implicit in the developmental approach is the notion of progression towards some cognitive goal. Such a goal necessitates the establishment of a base line of existing attainment, together with subsequent assessment so that progress and development can be measured. The study includes an evaluation of existing formal and informal methods of initial and subsequent assessment and diagnosis available at the time for Adult Literacy Scheme Co-ordinators. Underlying the funding by Cheshire County Council for the project is the assumption that the results will be available for all practitioners and that the tools of measurement may be used by other Adult Literacy Co-ordinators in the County. It is intended, therefore, that this research should result in practical outcomes in which methods of assessment will involve active participation by students as well as by tutors, becoming part of the learning process. It is hypothesised that this kind of co-operation could lead ultimately to self-directed learning and student-independence. For the purposes of this research, a balance is attempted in the use of assessment tools, between standardised tests and informal methods. The study provides facts about students! reading habits; as well as their reading levels, their spelling levels, their handwriting, their writing skills and their writing habits. The study seeks to show the students' feelings towards education, their educational attainments and the type of school which they attended. The study also attempts to come to some measurement of those aspects of student personality which relate to confidence, by means of tests and questionnaires. The study concludes with an examination of the link between cognitive and affective progress.

Learning in ultrametric committee machines

The problem of learning by examples in ultrametric committee machines (UCMs) is studied within the framework of statistical mechanics. Using the replica formalism we calculate the average generalization error in UCMs with L hidden layers and for a large enough number of units. In most of the regimes studied we find that the generalization error, as a function of the number of examples presented, develops a discontinuous drop at a critical value of the load parameter. We also find that when L>1 a number of teacher networks with the same number of hidden layers and different overlaps induce learning processes with the same critical points.

Learning sentiment classification model from labeled features

We propose a novel framework where an initial classifier is learned by incorporating prior information extracted from an existing sentiment lexicon. Preferences on expectations of sentiment labels of those lexicon words are expressed using generalized expectation criteria. Documents classified with high confidence are then used as pseudo-labeled examples for automatical domain-specific feature acquisition. The word-class distributions of such self-learned features are estimated from the pseudo-labeled examples and are used to train another classifier by constraining the model's predictions on unlabeled instances. Experiments on both the movie review data and the multi-domain sentiment dataset show that our approach attains comparable or better performance than exiting weakly-supervised sentiment classification methods despite using no labeled documents.

A novel framework of training hidden Markov support vector machines from lightly-annotated data

Natural language understanding (NLU) aims to map sentences to their semantic mean representations. Statistical approaches to NLU normally require fully-annotated training data where each sentence is paired with its word-level semantic annotations. In this paper, we propose a novel learning framework which trains the Hidden Markov Support Vector Machines (HM-SVMs) without the use of expensive fully-annotated data. In particular, our learning approach takes as input a training set of sentences labeled with abstract semantic annotations encoding underlying embedded structural relations and automatically induces derivation rules that map sentences to their semantic meaning representations. The proposed approach has been tested on the DARPA Communicator Data and achieved 93.18% in F-measure, which outperforms the previously proposed approaches of training the hidden vector state model or conditional random fields from unaligned data, with a relative error reduction rate of 43.3% and 10.6% being achieved.

Small business development and the “learning organisation”

The paper proffers a tentative conceptualisation of the “small business strategic learning process”, demonstrating the complexity of the small firm learning and management task. The framework, built upon personal construct theory and learning theories, is elaborated through the grounding of relevant areas of the strategic management literature in an understanding of the distinctive managerial and behavioural features of the small business. The framework is then utilised to underpin consideration of the concepts of “organisational learning” and the “learning organisation” within a small firm developmental context. It is suggested that whilst organisational learning may be a key and effective small business management approach to underpin sustainable development, the learning organisation, as currently conceived in the mainstream literature, fails to recognise and address the idiosyncrasies, problems and constraints relating to sustainable small business development. There does appear, however, to be great potential for extending understanding of the learning organisation concept into the small business context. An indicative research agenda is suggested.

More than a cognitive experience:unfamiliarity, invalidation, and emotion in organizational learning

Literature on organizational learning (OL) lacks an integrative framework that captures the emotions involved as OL proceeds. Drawing on personal construct theory, we suggest that organizations learn where their members reconstrue meaning around questions of strategic significance for the organization. In this 5-year study of an electronics company, we explore the way in which emotions change as members perceive progress or a lack of progress around strategic themes. Our framework also takes into account whether OL involves experiences that are familiar or unfamiliar and the implications for emotions. We detected similar patterns of emotion arising over time for three different themes in our data, thereby adding to OL perspectives that are predominantly cognitive in orientation. © The Author(s) 2013.