An investigation into inductive parameter learning in complex hierarchical knowledge structures representing clinical expertise

This dissertation investigates the very important and current problem of modelling human expertise. This is an apparent issue in any computer system emulating human decision making. It is prominent in Clinical Decision Support Systems (CDSS) due to the complexity of the induction process and the vast number of parameters in most cases. Other issues such as human error and missing or incomplete data present further challenges. In this thesis, the Galatean Risk Screening Tool (GRiST) is used as an example of modelling clinical expertise and parameter elicitation. The tool is a mental health clinical record management system with a top layer of decision support capabilities. It is currently being deployed by several NHS mental health trusts across the UK. The aim of the research is to investigate the problem of parameter elicitation by inducing them from real clinical data rather than from the human experts who provided the decision model. The induced parameters provide an insight into both the data relationships and how experts make decisions themselves. The outcomes help further understand human decision making and, in particular, help GRiST provide more accurate emulations of risk judgements. Although the algorithms and methods presented in this dissertation are applied to GRiST, they can be adopted for other human knowledge engineering domains.

Cues and knowledge structures used by mental-health professionals when making risk assessments

Background: Research into mental-health risks has tended to focus on epidemiological approaches and to consider pieces of evidence in isolation. Less is known about the particular factors and their patterns of occurrence that influence clinicians’ risk judgements in practice. Aims: To identify the cues used by clinicians to make risk judgements and to explore how these combine within clinicians’ psychological representations of suicide, self-harm, self-neglect, and harm to others. Method: Content analysis was applied to semi-structured interviews conducted with 46 practitioners from various mental-health disciplines, using mind maps to represent the hierarchical relationships of data and concepts. Results: Strong consensus between experts meant their knowledge could be integrated into a single hierarchical structure for each risk. This revealed contrasting emphases between data and concepts underpinning risks, including: reflection and forethought for suicide; motivation for self-harm; situation and context for harm to others; and current presentation for self-neglect. Conclusions: Analysis of experts’ risk-assessment knowledge identified influential cues and their relationships to risks. It can inform development of valid risk-screening decision support systems that combine actuarial evidence with clinical expertise.

The variance in sales performance explained by the knowledge structures of salespeople

The last major study of sales performance variance explained by salespeople attributes was by Churchill et al. (1985). They examined the effect of role, skills, motivation, personal factors, aptitude, and organizational/environmental factors on sales performance—factors that have dominated the sales performance area. About the same time, Weitz, Sujan, and Sujan (1986) introduced the concepts of salespeople's knowledge structures. Considerable work on the relationship of the elements of knowledge structures and performance can be found in the literature. In this research note, we determine the degree to which sales performance can be explained by knowledge structure variables, a heretofore unexplored area. If knowledge structure variables explain more variance than traditional variables, then this paper would be a call to further research in this area. In examining this research question in a retail context, we find that knowledge structure variables explain 50.2 percent of the variance in sales performance. We also find that variance explained by knowledge structures is significantly different based on gender. The impact of knowledge structures on performance was higher for men than for women. The models using education demonstrated smaller differences.

Discovering knowledge structures in mind maps of mental health risks

This thesis addressed the problem of risk analysis in mental healthcare, with respect to the GRiST project at Aston University. That project provides a risk-screening tool based on the knowledge of 46 experts, captured as mind maps that describe relationships between risks and patterns of behavioural cues. Mind mapping, though, fails to impose control over content, and is not considered to formally represent knowledge. In contrast, this thesis treated GRiSTs mind maps as a rich knowledge base in need of refinement; that process drew on existing techniques for designing databases and knowledge bases. Identifying well-defined mind map concepts, though, was hindered by spelling mistakes, and by ambiguity and lack of coverage in the tools used for researching words. A novel use of the Edit Distance overcame those problems, by assessing similarities between mind map texts, and between spelling mistakes and suggested corrections. That algorithm further identified stems, the shortest text string found in related word-forms. As opposed to existing approaches’ reliance on built-in linguistic knowledge, this thesis devised a novel, more flexible text-based technique. An additional tool, Correspondence Analysis, found patterns in word usage that allowed machines to determine likely intended meanings for ambiguous words. Correspondence Analysis further produced clusters of related concepts, which in turn drove the automatic generation of novel mind maps. Such maps underpinned adjuncts to the mind mapping software used by GRiST; one such new facility generated novel mind maps, to reflect the collected expert knowledge on any specified concept. Mind maps from GRiST are stored as XML, which suggested storing them in an XML database. In fact, the entire approach here is ”XML-centric”, in that all stages rely on XML as far as possible. A XML-based query language allows user to retrieve information from the mind map knowledge base. The approach, it was concluded, will prove valuable to mind mapping in general, and to detecting patterns in any type of digital information.

Ontology ranking based on the analysis of concept structures

In view of the need to provide tools to facilitate the re-use of existing knowledge structures such as ontologies, we present in this paper a system, AKTiveRank, for the ranking of ontologies. AKTiveRank uses as input the search terms provided by a knowledge engineer and, using the output of an ontology search engine, ranks the ontologies. We apply a number of metrics in an attempt to investigate their appropriateness for ranking ontologies, and compare the results with a questionnaire-based human study. Our results show that AKTiveRank will have great utility although there is potential for improvement.

Complex temporal patterns in molecular dynamics:a direct measure of the phase-space exploration by the trajectory at macroscopic time scales

Computer simulated trajectories of bulk water molecules form complex spatiotemporal structures at the picosecond time scale. This intrinsic complexity, which underlies the formation of molecular structures at longer time scales, has been quantified using a measure of statistical complexity. The method estimates the information contained in the molecular trajectory by detecting and quantifying temporal patterns present in the simulated data (velocity time series). Two types of temporal patterns are found. The first, defined by the short-time correlations corresponding to the velocity autocorrelation decay times (â‰0.1â€ps), remains asymptotically stable for time intervals longer than several tens of nanoseconds. The second is caused by previously unknown longer-time correlations (found at longer than the nanoseconds time scales) leading to a value of statistical complexity that slowly increases with time. A direct measure based on the notion of statistical complexity that describes how the trajectory explores the phase space and independent from the particular molecular signal used as the observed time series is introduced. © 2008 The American Physical Society.

Numerical modelling of complex femtosecond laser inscribed fiber gratings - comparison wih experiment

We present experimental studies and numerical modeling based on a combination of the Bidirectional Beam Propagation Method and Finite Element Modeling that completely describes the wavelength spectra of point by point femtosecond laser inscribed fiber Bragg gratings, showing excellent agreement with experiment. We have investigated the dependence of different spectral parameters such as insertion loss, all dominant cladding and ghost modes and their shape relative to the position of the fiber Bragg grating in the core of the fiber. Our model is validated by comparing model predictions with experimental data and allows for predictive modeling of the gratings. We expand our analysis to more complicated structures, where we introduce symmetry breaking; this highlights the importance of centered gratings and how maintaining symmetry contributes to the overall spectral quality of the inscribed Bragg gratings. Finally, the numerical modeling is applied to superstructure gratings and a comparison with experimental results reveals a capability for dealing with complex grating structures that can be designed with particular wavelength characteristics.

A method for automatically eliciting node weights in a hierarchical knowledge-based structure for reasoning with uncertainty

Hierarchical knowledge structures are frequently used within clinical decision support systems as part of the model for generating intelligent advice. The nodes in the hierarchy inevitably have varying influence on the decisionmaking processes, which needs to be reflected by parameters. If the model has been elicited from human experts, it is not feasible to ask them to estimate the parameters because there will be so many in even moderately-sized structures. This paper describes how the parameters could be obtained from data instead, using only a small number of cases. The original method [1] is applied to a particular web-based clinical decision support system called GRiST, which uses its hierarchical knowledge to quantify the risks associated with mental-health problems. The knowledge was elicited from multidisciplinary mental-health practitioners but the tree has several thousand nodes, all requiring an estimation of their relative influence on the assessment process. The method described in the paper shows how they can be obtained from about 200 cases instead. It greatly reduces the experts’ elicitation tasks and has the potential for being generalised to similar knowledge-engineering domains where relative weightings of node siblings are part of the parameter space.

Multiple conduction paths in boron δ-doped diamond structures

Impedance spectroscopy has been used to investigate conductivity within boron-doped diamond in an intrinsic/delta-doped/intrinsic (i-d-i) multilayer structure. For a 5 nm thick delta layer, three conduction pathways are observed, which can be assigned to transport within the delta layer and to two differing conduction paths in the i-layers adjoining the delta layer. For transport in the i-layers, thermal trapping/detrapping processes can be observed, and only at the highest temperature investigated (673 K) can transport due to a single conduction process be seen. Impedance spectroscopy is an ideal nondestructive tool for investigating the electrical characteristics of complex diamond structures.

KEMNAD:a knowledge engineering methodology for negotiating agent development

Automated negotiation is widely applied in various domains. However, the development of such systems is a complex knowledge and software engineering task. So, a methodology there will be helpful. Unfortunately, none of existing methodologies can offer sufficient, detailed support for such system development. To remove this limitation, this paper develops a new methodology made up of: (1) a generic framework (architectural pattern) for the main task, and (2) a library of modular and reusable design pattern (templates) of subtasks. Thus, it is much easier to build a negotiating agent by assembling these standardised components rather than reinventing the wheel each time. Moreover, since these patterns are identified from a wide variety of existing negotiating agents (especially high impact ones), they can also improve the quality of the final systems developed. In addition, our methodology reveals what types of domain knowledge need to be input into the negotiating agents. This in turn provides a basis for developing techniques to acquire the domain knowledge from human users. This is important because negotiation agents act faithfully on the behalf of their human users and thus the relevant domain knowledge must be acquired from the human users. Finally, our methodology is validated with one high impact system.

Semantic search components:a blueprint for effective query language interfaces

Formulating complex queries is hard, especially when users cannot understand all the data structures of multiple complex knowledge bases. We see a gap between simplistic but user friendly tools and formal query languages. Building on an example comparison search, we propose an approach in which reusable search components take an intermediary role between the user interface and formal query languages.

Numerical modelling of complex femtosecond laser inscribed fiber gratings - comparison wih experiment

We present experimental studies and numerical modeling based on a combination of the Bidirectional Beam Propagation Method and Finite Element Modeling that completely describes the wavelength spectra of point by point femtosecond laser inscribed fiber Bragg gratings, showing excellent agreement with experiment. We have investigated the dependence of different spectral parameters such as insertion loss, all dominant cladding and ghost modes and their shape relative to the position of the fiber Bragg grating in the core of the fiber. Our model is validated by comparing model predictions with experimental data and allows for predictive modeling of the gratings. We expand our analysis to more complicated structures, where we introduce symmetry breaking; this highlights the importance of centered gratings and how maintaining symmetry contributes to the overall spectral quality of the inscribed Bragg gratings. Finally, the numerical modeling is applied to superstructure gratings and a comparison with experimental results reveals a capability for dealing with complex grating structures that can be designed with particular wavelength characteristics.

Representing human expertise by the OWL web ontology language to support knowledge engineering in decision support systems

Clinical decision support systems (CDSSs) often base their knowledge and advice on human expertise. Knowledge representation needs to be in a format that can be easily understood by human users as well as supporting ongoing knowledge engineering, including evolution and consistency of knowledge. This paper reports on the development of an ontology specification for managing knowledge engineering in a CDSS for assessing and managing risks associated with mental-health problems. The Galatean Risk and Safety Tool, GRiST, represents mental-health expertise in the form of a psychological model of classification. The hierarchical structure was directly represented in the machine using an XML document. Functionality of the model and knowledge management were controlled using attributes in the XML nodes, with an accompanying paper manual for specifying how end-user tools should behave when interfacing with the XML. This paper explains the advantages of using the web-ontology language, OWL, as the specification, details some of the issues and problems encountered in translating the psychological model to OWL, and shows how OWL benefits knowledge engineering. The conclusions are that OWL can have an important role in managing complex knowledge domains for systems based on human expertise without impeding the end-users' understanding of the knowledge base. The generic classification model underpinning GRiST makes it applicable to many decision domains and the accompanying OWL specification facilitates its implementation.

Ranking ontologies with AKTiveRank

Ontology search and reuse is becoming increasingly important as the quest for methods to reduce the cost of constructing such knowledge structures continues. A number of ontology libraries and search engines are coming to existence to facilitate locating and retrieving potentially relevant ontologies. The number of ontologies available for reuse is steadily growing, and so is the need for methods to evaluate and rank existing ontologies in terms of their relevance to the needs of the knowledge engineer. This paper presents AKTiveRank, a prototype system for ranking ontologies based on a number of structural metrics.

Metrics for ranking ontologies

Representing knowledge using domain ontologies has shown to be a useful mechanism and format for managing and exchanging information. Due to the difficulty and cost of building ontologies, a number of ontology libraries and search engines are coming to existence to facilitate reusing such knowledge structures. The need for ontology ranking techniques is becoming crucial as the number of ontologies available for reuse is continuing to grow. In this paper we present AKTiveRank, a prototype system for ranking ontologies based on the analysis of their structures. We describe the metrics used in the ranking system and present an experiment on ranking ontologies returned by a popular search engine for an example query.