Representing clinical documents to support automatic retrieval of evidence from the Cochrane Library

The overall aim of our research is to develop a clinical information retrieval system that retrieves systematic reviews and underlying clinical studies from the Cochrane Library to support physician decision making. We believe that in order to accomplish this goal we need to develop a mechanism for effectively representing documents that will be retrieved by the application. Therefore, as a first step in developing the retrieval application we have developed a methodology that semi-automatically generates high quality indices and applies them as descriptors to documents from The Cochrane Library. In this paper we present a description and implementation of the automatic indexing methodology and an evaluation that demonstrates that enhanced document representation results in the retrieval of relevant documents for clinical queries. We argue that the evaluation of information retrieval applications should also include an evaluation of the quality of the representation of documents that may be retrieved. ©2010 IEEE.

A comparative evaluation of term recognition algorithms

Automatic Term Recognition (ATR) is a fundamental processing step preceding more complex tasks such as semantic search and ontology learning. From a large number of methodologies available in the literature only a few are able to handle both single and multi-word terms. In this paper we present a comparison of five such algorithms and propose a combined approach using a voting mechanism. We evaluated the six approaches using two different corpora and show how the voting algorithm performs best on one corpus (a collection of texts from Wikipedia) and less well using the Genia corpus (a standard life science corpus). This indicates that choice and design of corpus has a major impact on the evaluation of term recognition algorithms. Our experiments also showed that single-word terms can be equally important and occupy a fairly large proportion in certain domains. As a result, algorithms that ignore single-word terms may cause problems to tasks built on top of ATR. Effective ATR systems also need to take into account both the unstructured text and the structured aspects and this means information extraction techniques need to be integrated into the term recognition process.

Studies on the generalisation of Gaussian processes and Bayesian neural networks

The assessment of the reliability of systems which learn from data is a key issue to investigate thoroughly before the actual application of information processing techniques to real-world problems. Over the recent years Gaussian processes and Bayesian neural networks have come to the fore and in this thesis their generalisation capabilities are analysed from theoretical and empirical perspectives. Upper and lower bounds on the learning curve of Gaussian processes are investigated in order to estimate the amount of data required to guarantee a certain level of generalisation performance. In this thesis we analyse the effects on the bounds and the learning curve induced by the smoothness of stochastic processes described by four different covariance functions. We also explain the early, linearly-decreasing behaviour of the curves and we investigate the asymptotic behaviour of the upper bounds. The effect of the noise and the characteristic lengthscale of the stochastic process on the tightness of the bounds are also discussed. The analysis is supported by several numerical simulations. The generalisation error of a Gaussian process is affected by the dimension of the input vector and may be decreased by input-variable reduction techniques. In conventional approaches to Gaussian process regression, the positive definite matrix estimating the distance between input points is often taken diagonal. In this thesis we show that a general distance matrix is able to estimate the effective dimensionality of the regression problem as well as to discover the linear transformation from the manifest variables to the hidden-feature space, with a significant reduction of the input dimension. Numerical simulations confirm the significant superiority of the general distance matrix with respect to the diagonal one.In the thesis we also present an empirical investigation of the generalisation errors of neural networks trained by two Bayesian algorithms, the Markov Chain Monte Carlo method and the evidence framework; the neural networks have been trained on the task of labelling segmented outdoor images.

Individual differences in human-computer interaction

This thesis initially presents an 'assay' of the literature pertaining to individual differences in human-computer interaction. A series of experiments is then reported, designed to investigate the association between a variety of individual characteristics and various computer task and interface factors. Predictor variables included age, computer expertise, and psychometric tests of spatial visualisation, spatial memory, logical reasoning, associative memory, and verbal ability. These were studied in relation to a variety of computer-based tacks, including: (1) word processing and its component elements; (ii) the location of target words within passages of text; (iii) the navigation of networks and menus; (iv) command generation using menus and command line interfaces; (v) the search and selection of icons and text labels; (vi) information retrieval. A measure of self-report workload was also included in several of these experiments. The main experimental findings included: (i) an interaction between spatial ability and the manipulation of semantic but not spatial interface content; (ii) verbal ability being only predictive of certain task components of word processing; (iii) age differences in word processing and information retrieval speed but not accuracy; (iv) evidence of compensatory strategies being employed by older subjects; (v) evidence of performance strategy differences which disadvantaged high spatial subjects in conditions of low spatial information content; (vi) interactive effects of associative memory, expertise and command strategy; (vii) an association between logical reasoning and word processing but not information retrieval; (viii) an interaction between expertise and cognitive demand; and (ix) a stronger association between cognitive ability and novice performance than expert performance.

Clustering web documents using hierarchical representation with multi-granularity

Web document cluster analysis plays an important role in information retrieval by organizing large amounts of documents into a small number of meaningful clusters. Traditional web document clustering is based on the Vector Space Model (VSM), which takes into account only two-level (document and term) knowledge granularity but ignores the bridging paragraph granularity. However, this two-level granularity may lead to unsatisfactory clustering results with “false correlation”. In order to deal with the problem, a Hierarchical Representation Model with Multi-granularity (HRMM), which consists of five-layer representation of data and a twophase clustering process is proposed based on granular computing and article structure theory. To deal with the zero-valued similarity problemresulted from the sparse term-paragraphmatrix, an ontology based strategy and a tolerance-rough-set based strategy are introduced into HRMM. By using granular computing, structural knowledge hidden in documents can be more efficiently and effectively captured in HRMM and thus web document clusters with higher quality can be generated. Extensive experiments show that HRMM, HRMM with tolerancerough-set strategy, and HRMM with ontology all outperform VSM and a representative non VSM-based algorithm, WFP, significantly in terms of the F-Score.

Discovering the preferences of physicians with regards to rank-ordered medical documents

The practice of evidence-based medicine involves consulting documents from repositories such as Scopus, PubMed, or the Cochrane Library. The most common approach for presenting retrieved documents is in the form of a list, with the assumption that the higher a document is on a list, the more relevant it is. Despite this list-based presentation, it is seldom studied how physicians perceive the importance of the order of documents presented in a list. This paper describes an empirical study that elicited and modeled physicians' preferences with regard to list-based results. Preferences were analyzed using a GRIP method that relies on pairwise comparisons of selected subsets of possible rank-ordered lists composed of 3 documents. The results allow us to draw conclusions regarding physicians' attitudes towards the importance of having documents ranked correctly on a result list, versus the importance of retrieving relevant but misplaced documents. Our findings should help developers of clinical information retrieval applications when deciding how retrieved documents should be presented and how performance of the application should be assessed. © 2012 Springer-Verlag Berlin Heidelberg.

Building and applying a concept hierarchy representation of a user profile

Term dependence is a natural consequence of language use. Its successful representation has been a long standing goal for Information Retrieval research. We present a methodology for the construction of a concept hierarchy that takes into account the three basic dimensions of term dependence. We also introduce a document evaluation function that allows the use of the concept hierarchy as a user profile for Information Filtering. Initial experimental results indicate that this is a promising approach for incorporating term dependence in the way documents are filtered.

Timeline generation with social attention

Timeline generation is an important research task which can help users to have a quick understanding of the overall evolution of any given topic. It thus attracts much attention from research communities in recent years. Nevertheless, existing work on timeline generation often ignores an important factor, the attention attracted to topics of interest (hereafter termed "social attention"). Without taking into consideration social attention, the generated timelines may not reflect users' collective interests. In this paper, we study how to incorporate social attention in the generation of timeline summaries. In particular, for a given topic, we capture social attention by learning users' collective interests in the form of word distributions from Twitter, which are subsequently incorporated into a unified framework for timeline summary generation. We construct four evaluation sets over six diverse topics. We demonstrate that our proposed approach is able to generate both informative and interesting timelines. Our work sheds light on the feasibility of incorporating social attention into traditional text mining tasks. Copyright © 2013 ACM.

What do people think about when they answer theory of planned behaviour questionnaires? A 'think aloud' study

Two studies aiming to identify the nature and extent of problems that people have when completing theory of planned behaviour (TPB) questionnaires, using a cognitive interviewing approach are reported. Both studies required participants to 'think aloud' as they completed TPB questionnaires about: (a) increasing physical activity (six general public participants); and (b) binge drinking (13 students). Most people had no identifiable problems with the majority of questions. However, there were problems common to both studies, relating to information retrieval and to participants answering different questions from those intended by researchers. Questions about normative influence were particularly problematic. The standard procedure for developing TPB questionnaires may systematically produce problematic questions. Suggestions are made for improving this procedure. Copyright © 2007 SAGE Publications.

Learning higher-level features with convolutional restricted Boltzmann machines for sentiment analysis

In recent years, learning word vector representations has attracted much interest in Natural Language Processing. Word representations or embeddings learned using unsupervised methods help addressing the problem of traditional bag-of-word approaches which fail to capture contextual semantics. In this paper we go beyond the vector representations at the word level and propose a novel framework that learns higher-level feature representations of n-grams, phrases and sentences using a deep neural network built from stacked Convolutional Restricted Boltzmann Machines (CRBMs). These representations have been shown to map syntactically and semantically related n-grams to closeby locations in the hidden feature space. We have experimented to additionally incorporate these higher-level features into supervised classifier training for two sentiment analysis tasks: subjectivity classification and sentiment classification. Our results have demonstrated the success of our proposed framework with 4% improvement in accuracy observed for subjectivity classification and improved the results achieved for sentiment classification over models trained without our higher level features.

Task-based annotation and retrieval for image information management

Continuing advances in digital image capture and storage are resulting in a proliferation of imagery and associated problems of information overload in image domains. In this work we present a framework that supports image management using an interactive approach that captures and reuses task-based contextual information. Our framework models the relationship between images and domain tasks they support by monitoring the interactive manipulation and annotation of task-relevant imagery. During image analysis, interactions are captured and a task context is dynamically constructed so that human expertise, proficiency and knowledge can be leveraged to support other users in carrying out similar domain tasks using case-based reasoning techniques. In this article we present our framework for capturing task context and describe how we have implemented the framework as two image retrieval applications in the geo-spatial and medical domains. We present an evaluation that tests the efficiency of our algorithms for retrieving image context information and the effectiveness of the framework for carrying out goal-directed image tasks. © 2010 Springer Science+Business Media, LLC.

An information foraging theory based user study of an adaptive user interaction framework for content-based image retrieval

This paper presents the design and results of a task-based user study, based on Information Foraging Theory, on a novel user interaction framework - uInteract - for content-based image retrieval (CBIR). The framework includes a four-factor user interaction model and an interactive interface. The user study involves three focused evaluations, 12 simulated real life search tasks with different complexity levels, 12 comparative systems and 50 subjects. Information Foraging Theory is applied to the user study design and the quantitative data analysis. The systematic findings have not only shown how effective and easy to use the uInteract framework is, but also illustrate the value of Information Foraging Theory for interpreting user interaction with CBIR. © 2011 Springer-Verlag Berlin Heidelberg.

Applying Information Foraging Theory to understand user interaction with content-based image retrieval

The paper proposes an ISE (Information goal, Search strategy, Evaluation threshold) user classification model based on Information Foraging Theory for understanding user interaction with content-based image retrieval (CBIR). The proposed model is verified by a multiple linear regression analysis based on 50 users' interaction features collected from a task-based user study of interactive CBIR systems. To our best knowledge, this is the first principled user classification model in CBIR verified by a formal and systematic qualitative analysis of extensive user interaction data. Copyright 2010 ACM.

Task-based image annotation and retrieval

In order to address problems of information overload in digital imagery task domains we have developed an interactive approach to the capture and reuse of image context information. Our framework models different aspects of the relationship between images and domain tasks they support by monitoring the interactive manipulation and annotation of task-relevant imagery. The approach allows us to gauge a measure of a user's intentions as they complete goal-directed image tasks. As users analyze retrieved imagery their interactions are captured and an expert task context is dynamically constructed. This human expertise, proficiency, and knowledge can then be leveraged to support other users in carrying out similar domain tasks. We have applied our techniques to two multimedia retrieval applications for two different image domains, namely the geo-spatial and medical imagery domains. © Springer-Verlag Berlin Heidelberg 2007.