Malcolm Clark

Automatically structuring domain knowledge from text: An overview of current research

This paper presents an overview of automatic methods for building domain knowledge structures (domain models) from text collections. Applications of domain models have a long history within knowledge engineering and artificial intelligence. In the last couple of decades they have surfaced noticeably as a useful tool within natural language processing, information retrieval and semantic web technology. Inspired by the ubiquitous propagation of domain model structures that are emerging in several research disciplines, we give an overview of the current research landscape and some techniques and approaches. We will also discuss trade-offs between different approaches and point to some recent trends.

Classifying XML Documents by Using Genre Features

Customers generally like to see alternative products and compare their characteristics and prices before deciding on one of them. Therefore, proposing alternative products is one of the crucial issues for e-commerce applications to increase customer satisfaction. This paper proposes a fuzzy similarity-based approach to determine similar products recorded in a database and submit them intelligently to the customer in a ranked way as alternative products.The categorization of documents is traditionally topic-based. This paper presents a complementary analysis of research and experiments on genre to show that encouraging results can be obtained by using genre structure (form) features. We conducted an experiment to assess the effectiveness of using eXtensible Mark-Up Language (XML) tag information, and part-of-speech (P-O-S) features, for the classification of genres, testing the hypothesis that if a focus on genre can lead to high precision on normal textual documents, then good results can be achieved using XML tag information in addition to P-O-S information. An experiment was carried out on a subsection of the initiative for the evaluation of XML (INEX) 1.4 collection. The features were extracted and documents were classified using machine learning algorithms, which yielded encouraging results for logistic regression and neural networks. We propose that utilizing these features and training a classifier may benefit retrieval for most World Wide Web (WWW) technologies such as XML and eXtensible Hypertext Markup Language) XHTML.

Compact Representations in XML Retrieval

This paper describes the participation of the Information Retrieval and Interaction group of Robert Gordon University in the INEX 2006 ad hoc track. We focused on two questions: “What potential evidence do human assessors use to identify relevant XML elements?” and “How can this evidence be used by computers for the same task?”. Our main strategy was to investigate evidence taken not only from the content, but also from the shallow features of how texts were displayed. We employed the vector space model and the language model combining estimates based on element full-text and the compact representation of the element. We analyzed a range of non-content priors to boost retrieval effectiveness.

You have e-mail, what happens next? Tracking the eyes for genre

This paper reports on an approach to the analysis of form (layout and formatting) during genre recognition recorded using eye tracking. The researchers focused on eight different types of e-mail, such as calls for papers, newsletters and spam, which were chosen to represent different genres. The study involved the collection of oculographic behavior data based on the scanpath duration and scanpath length based metric, to highlight the ways in which people view the features of genres. We found that genre analysis based on purpose and form (layout features, etc.) was an effective means of identifying the characteristics of these e-mails. The research, carried out on a group of 24 participants, highlighted their interaction and interpretation of the e-mail texts and the visual cues or features perceived. In addition, the ocular strategies of scanning and skimming, they employed for the processing of the texts by block, genre and representation were evaluated.

Perceiving and Using Genre by Form - An Eye-Tracking Study

Abstract This paper reports on an approach to the analysis of genre recognition using eye-tracking. The researchers focused on eight different types of e-mail, such as calls for papers, newsletters and spam, which were chosen to represent different genres. The study involved the collection of oculographic behaviour data metrics, such as fixations and saccades to highlight the ways in which people view the features of genres. We found that genre analysis based on purpose and form (layout features, etc) was an effective means of identifying the characteristics of these e-mails. The research, carried out on a group of 24 participants, highlighted their interaction with the e-mail texts and the visual cues or features perceived as well as the strategies they employed for the processing of the texts. The results showed that readers can determine the purpose and form of genres, that form and content can occasionally be separable, that some features cause fixations and that some readers are prompted to respond by using saccadic behaviour (e.g. regressive saccades) over the shape of the e-mails (form).