Recent years hve seen the increasing use of expert systems and expert system shells in industry and commerce. However, there has been little use of the technology in schools and colleges, partly due to dissatisfaction with the available software. This thesis describes the design, implementation and evaluation of an expert system shell for use in education. An initial evaluation was carried out to establish some criteria for the design of the new shell. This comprised an evaluation of expert system shells by experienced teachers; an evaluation of several expert system shells by the author; and a case-study visit to a school Computer Studies department. The teachers were also asked for their views on the possible uses of shells in the classroom. An expert system shell was designed and implemented as a result of this evalation. This has an environment for building knowledge bases; an environment for consulting knowledge bases; a knowledge representation language allowing separation of domain and problem solving knowledge; and three supplementarty tools. The most important of these is an explanation tool, which allows exploration of the domain knowledge in a knowledge base. In response to posed questions, the tool can produce domain-based explanations using a domain-independent generation technique, and each explanation can be explored further through relevant follow-up questions. This tool provides a means whereby users can learn about the domain of the knowledge base being explored. An evaluation of this shell by teachers with previous experience of expert system shells suggests that it has the potential for use throughout the school curriculum as an educational medium.

This thesis presents the design of a Graphical User Interface (GUI) for computer-based sound synthesis to support users in the externalisation of their musical ideas when interacting with the System in order to create and manipulate sound. The approach taken consisted of three research stages. The first stage was the formulation of a novel visualisation framework to display perceptual dimensions of sound in Visual terms. This framework was based on the findings of existing related studies and a series of empirical investigations of the associations between auditory and visual precepts that we performed for the first time in the area of computer-based sound synthesis. The results of our empirical investigations suggested associations between the colour dimensions of brightness and saturation with the auditory dimensions of pitch and loudness respectively, as well as associations between the multidimensional precepts of visual texture and timbre. The second stage of the research involved the design and implementation of Sound Mosaics, a prototype GUI for sound synthesis based on direct manipulation of visual representations that make use of the visualisation framework developed in the first stage. We followed an iterative design approach that involved the design and evaluation of an initial Sound Mosaics prototype. The insights gained during this first iteration assisted us in revising various aspects of the original design and visualisation framework that led to a revised implementation of Sound Mosaics. The final stage of this research involved an evaluation study of the revised Sound Mosaics prototype that comprised two controlled experiments. First, a comparison experiment with the widely used frequency-domain representations of sound indicated that visual representations created with Sound Mosaics were more comprehensible and intuitive. Comprehensibility was measured as the level of accuracy in a series of sound image association tasks, while intuitiveness was related to subjects' response times and perceived levels of confidence. Second, we conducted a formative evaluation of Sound Mosaics, in which it was exposed to a number of users with and without musical background. Three usability factors were measured: effectiveness, efficiency, and subjective satisfaction. Sound Mosaics was demonstrated to perform satisfactorily in ail three factors for music subjects, although non-music subjects yielded less satisfactory results that can be primarily attributed to the subjects' unfamiliarity with the task of sound synthesis. Overall, our research has set the necessary groundwork for empirically derived and validated associations between auditory and visual dimensions that can be used in the design of cognitively useful GUIs for computer-based sound synthesis and related area.

The work presented in this thesis is built on three assertions: (1) uncertainties should be perceived as integral components of GIS spatial databases; (2) as such and given the importance of uncertainties at all stages of processing spatial data by digital methods, an integrating strategy is needed to provide more direct access to the uncertainties of spatial data during data collection, update, spatial analysis and during the creation of output products; (3) surface-based models and methods are capable of such an integral strategy, by which many kinds of uncertainties of spatial data can be well represented and handled. A cumulative description is given of various uncertainties occurring in geographical abstraction and spatial data acquisitions with special reference to one common area of geographical studies, that is, land cover mapping. Two alternative forms of geographical abstraction or spatial data modelling are introduced: discrete objects and continuous fields. The uncertainties are then discussed with respect to their description, estimation and representation under object and field perspectives. For categorical data, in particular, uncertainties are represented as fuzzy surfaces, whose derivation and analysis are described in detail. To provide an evaluation of the integrated approach and to show how such an integrated strategy can be used to advantage, a case study is developed in the context of suburban land cover mapping, based on a local Edinburgh area. The case study begins with the construction of a co-registered hierarchy of test data with a corresponding hierarchy of accuracies, and continues to the generation and analysis of fuzzy surfaces using the suite of methods introduced previously. The various graphical maps and quantitative data produced show that surface based approaches are well suited to the representation and handling of uncertainties of spatial data, because they are effective and flexible.

Theorem proving in undecidable theories has to resort to <I>semi-decision procedures</I>, i.e. partial computable functions that halt when the input formula is a theorem, but may not terminate otherwise. Their performance can be improved once heuristic mechanisms are introduced to guide search. Heuristic functions, however, sometimes assume the properties of a particular subclass of formulae to be valid in a larger domain, and may as a result fail to recognise a theorem due to the loss of completeness. Efficiency may also be improved when the decidability of subclasses of formulae is explored. <I>Decision procedures </I>establish, after a finite amount of time and computation, whether a formula of a class is a theorem or not. Their main advantage is their total computability, i.e. computation terminates for every formula of the domain, whether it is a theorem or not. Their reduced scope of application and, in some cases, their complexity are, nonetheless, the main limitations. Evidence suggests that the development of efficient mechanical theorem provers requires the integration of both heuristic modules and decision procedures inside hybrid systems. Integration is achieved through at least three strategies. (i) <I>juxtaposition</I>, where each component of the system operates independently from the others, and there is no communication between them, (ii) <I>cooperation</I>, where the behaviour of a component influences others, and communication is direct, and (iii) <I>interfacing</I>, where transformation or simplification steps take place before a formula is delivered to a component of the system, in which case communication amongst modules is mediated. From the decision procedure viewpoint, the main effect of its integration into a prover is the enlargement of the decidable domain, whereas, for the heuristic component, there is a reduction in the number of subproblems it has to address.

No description available.

Computer-generated musical performances are often criticised for being unable to match the expressivity found in performances by humans. Much research has been conducted in the past two decades in order to create computer technology able to perform a given piece music as expressively as humans, largely without success. Two approaches have been often adopted to research into modelling expressive music performance on computers. The first focuses on sound; that is, on modelling patterns of deviations between a recorded human performance and the music score. The second focuses on modelling the cognitive processes involved in a musical performance. Both approaches are valid and can complement each other. In this thesis we propose a third complementary approach, focusing on the guitar, which concerns the physical manipulation of the instrument by the performer: a biomechanical approach. The essence of this thesis is a study on capturing, analyzing and modelling information about motor and biomechanical processes of guitar performance. The focus is on speed, precision, and force of a guitarist's left-hand. The overarching questions behind our study are: 1) Do unintentional actions originating from motor and biomechanical functions during musical performance contribute a material &quot;human feel&quot; to the performance? 2) Would it be possible determine and quantify such unintentional actions? 3) Would it be possible to model and embed such information in a computer system? The contributionst o knowledgep ursued in this thesis include: a) An unprecedented study of guitar mechanics, ergonomics, and playability; b) A detailed study of how the human body performs actions when playing the guitar; c) A methodologyt o formally record quantifiable data about such actionsin performance; d) An approach to model such information, and e) A demonstration of how the above knowledge can be embeddedin a system for music performance.

This thesis presents an unusual application of user modelling, the domain of interest being the physical abilities of a user. Specifically, a model which identifies aspects of keyboard use with which a user has difficulty is presented. An empirical study of keyboard and mouse use by people with and without motor disabilities is described. The study focuses on common input errors due to physical innacuracies in using these devices. For the majority of these errors, there exist keyboard or mouse configuration facilities intended to reduce or eliminate them. While such facilities are now integrated into the majority of modern operating systems, there is little published data describing their effect on keyboard or mouse usability. This thesis offers evidence that they can be extremely useful, even essential, but that further research and interface development are required. The user model focuses on four of the most commonly observed keyboard difficulties, and makes recommendations for settings for three keyboard configuration facilities. As a user modelling task, this application presents a number of interesting challenges for which traditional user modelling techniques are inadequate . The users to be modelled may tire easily, and may have cognitive disabilities in addition to physical ones. This makes it difficult to gain information by questioning users, or requiring them to perform given tasks. On shared machines many different users with very different requirements will be encountered, while individuals may also have varying requirements. This calls for a dynamic technique. This thesis attempts to show that it is feasable to draw accurate conclusions about a user's keyboard configuration requirements without questioning them, or requiring them to perform controlled tasks. The model presented records for a user's input unintrusively, and examines this input for evidence of physical errors or difficulties. Conclusions are based on the assumption that the user is typing English text. Any other textual language could be used.

Using a computer to answer questions has been a human dream since the beginning of the digital era. A first step towards the achievement of such an ambitious goal is to deal with natural language to enable the computer to understand what its user asks. The discipline that studies the connection between natural language and the representation of its meaning via computational models is computational linguistics. According to such discipline, Question Answering can be defined as the task that, given a question formulated in natural language, aims at finding one or more concise answers in the form of sentences or phrases. Question Answering can be interpreted as a sub-discipline of information retrieval with the added challenge of applying sophisticated techniques to identify the complex syntactic and semantic relationships present in text. Although it is widely accepted that Question Answering represents a step beyond standard information retrieval, allowing a more sophisticated and satisfactory response to the user's information needs, it still shares a series of unsolved issues with the latter. First, in most state-of-the-art Question Answering systems, the results are created independently of the questioner's characteristics, goals and needs. This is a serious limitation in several cases: for instance, a primary school child and a History student may need different answers to the questlon: When did the Middle Ages begin? Moreover, users often issue queries not as standalone but in the context of a wider information need, for instance when researching a specific topic. Although it has recently been proposed that providing Question Answering systems with dialogue interfaces would encourage and accommodate the submission of multiple related questions and handle the user's requests for clarification, interactive Question Answering is still at its early stages: Furthermore, an issue which still remains open in current Question Answering is that of efficiently answering complex questions, such as those invoking definitions and descriptions (e.g. What is a metaphor?). Indeed, it is difficult to design criteria to assess the correctness of answers to such complex questions. These are the central research problems addressed by this thesis, and are solved as follows. An in-depth study on complex Question Answering led to the development of classifiers for complex answers. These exploit a variety of lexical, syntactic and shallow semantic features to perform textual classification using tree-kernel functions for Support Vector Machines. The issue of personalization is solved by the integration of a User Modelling component within the Question Answering model. The User Model is able to filter and re-rank results based on the user's reading level and interests. The issue of interactivity is approached by the development of a dialogue model and a dialogue manager suitable for open-domain interactive Question Answering. The utility of such model is corroborated by the integration of an interactive interface to allow reference resolution and follow-up conversation into the core Question Answering system and by its evaluation. Finally, the models of personalized and interactive Question Answering are integrated in a comprehensive framework forming a unified model for future Question Answering research.

Much work has been done on developing algorithms for automated surface defect detection. However, comparisons between these models and human perception are rarely carried out. This thesis aims to investigate how well human observers can nd defects in textured surfaces, over a wide range of task di culties. Stimuli for experiments will be generated using texture synthesis methods and human search strategies will be captured by use of an eye tracker. Two di erent modelling approaches will be explored. A computational LNL-based model will be developed and compared to human performance in terms of the number of xations required to find the target. Secondly, a stochastic simulation, based on empirical distributions of saccades, will be compared to human search strategies.

No description available.