Effectiveness of document representation for classification

Conventionally, document classification researches focus on improving the learning capabilities of classifiers. Nevertheless, according to our observation, the effectiveness of classification is limited by the suitability of document representation. Intuitively, the more features that are used in representation, the more comprehensive that documents are represented. However, if a representation contains too many irrelevant features, the classifier would suffer from not only the curse of high dimensionality, but also overfitting. To address this problem of suitableness of document representations, we present a classifier-independent approach to measure the effectiveness of document representations. Our approach utilises a labelled document corpus to estimate the distribution of documents in the feature space. By looking through documents in this way, we can clearly identify the contributions made by different features toward the document classification. Some experiments have been performed to show how the effectiveness is evaluated. Our approach can be used as a tool to assist feature selection, dimensionality reduction and document classification.

Spoken document representations for probabilistic retrieval

This paper presents some developments in query expansion and document representation of our spoken document retrieval system and shows how various retrieval techniques affect performance for different sets of transcriptions derived from a common speech source. Modifications of the document representation are used, which combine several techniques for query expansion, knowledge-based on one hand and statistics-based on the other. Taken together, these techniques can improve Average Precision by over 19% relative to a system similar to that which we presented at TREC-7. These new experiments have also confirmed that the degradation of Average Precision due to a word error rate (WER) of 25% is quite small (3.7% relative) and can be reduced to almost zero (0.2% relative). The overall improvement of the retrieval system can also be observed for seven different sets of transcriptions from different recognition engines with a WER ranging from 24.8% to 61.5%. We hope to repeat these experiments when larger document collections become available, in order to evaluate the scalability of these techniques.

Enhancing an Incremental Clustering Algorithm for Web Page Collections

With the size and state of the Internet today, a good quality approach to organizing this mass of information is of great importance. Clustering web pages into groups of similar documents is one approach, but relies heavily on good feature extraction and document representation as well as a good clustering approach and algorithm. Due to the changing nature of the Internet, resulting in a dynamic dataset, an incremental approach is preferred. In this work we propose an enhanced incremental clustering approach to develop a better clustering algorithm that can help to better organize the information available on the Internet in an incremental fashion. Experiments show that the enhanced algorithm outperforms the original histogram based algorithm by up to 7.5%.

XML documents clustering using tensor space model - A preliminary study

A hierarchical structure is used to represent the content of the semi-structured documents such as XML and XHTML. The traditional Vector Space Model (VSM) is not sufficient to represent both the structure and the content of such web documents. Hence in this paper, we introduce a novel method of representing the XML documents in Tensor Space Model (TSM) and then utilize it for clustering. Empirical analysis shows that the proposed method is scalable for a real-life dataset as well as the factorized matrices produced from the proposed method helps to improve the quality of clusters due to the enriched document representation with both the structure and the content information.

XML documents clustering using Tensor Space Model

The traditional Vector Space Model (VSM) is not able to represent both the structure and the content of XML documents. This paper introduces a novel method of representing XML documents in a Tensor Space Model (TSM) and then utilizing it for clustering. Empirical analysis shows that the proposed method is scalable for large-sized datasets; as well, the factorized matrices produced from the proposed method help to improve the quality of clusters through the enriched document representation of both structure and content information.

The heterogeneous cluster ensemble method using hubness for clustering text documents

We propose a cluster ensemble method to map the corpus documents into the semantic space embedded in Wikipedia and group them using multiple types of feature space. A heterogeneous cluster ensemble is constructed with multiple types of relations i.e. document-term, document-concept and document-category. A final clustering solution is obtained by exploiting associations between document pairs and hubness of the documents. Empirical analysis with various real data sets reveals that the proposed meth-od outperforms state-of-the-art text clustering approaches.

Pattern-based topic models for information filtering

Topic modelling, such as Latent Dirichlet Allocation (LDA), was proposed to generate statistical models to represent multiple topics in a collection of documents, which has been widely utilized in the fields of machine learning and information retrieval, etc. But its effectiveness in information filtering is rarely known. Patterns are always thought to be more representative than single terms for representing documents. In this paper, a novel information filtering model, Pattern-based Topic Model(PBTM) , is proposed to represent the text documents not only using the topic distributions at general level but also using semantic pattern representations at detailed specific level, both of which contribute to the accurate document representation and document relevance ranking. Extensive experiments are conducted to evaluate the effectiveness of PBTM by using the TREC data collection Reuters Corpus Volume 1. The results show that the proposed model achieves outstanding performance.

LRD : Latent Relation Discovery for vector space expansion and information retrieval

In this paper, we propose a text mining method called LRD (latent relation discovery), which extends the traditional vector space model of document representation in order to improve information retrieval (IR) on documents and document clustering. Our LRD method extracts terms and entities, such as person, organization, or project names, and discovers relationships between them by taking into account their co-occurrence in textual corpora. Given a target entity, LRD discovers other entities closely related to the target effectively and efficiently. With respect to such relatedness, a measure of relation strength between entities is defined. LRD uses relation strength to enhance the vector space model, and uses the enhanced vector space model for query based IR on documents and clustering documents in order to discover complex relationships among terms and entities. Our experiments on a standard dataset for query based IR shows that our LRD method performed significantly better than traditional vector space model and other five standard statistical methods for vector expansion.

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.

Mapping and Displaying Structural Transformations between XML and PDF

Documents are often marked up in XML-based tagsets to delineate major structural components such as headings, paragraphs, figure captions and so on, without much regard to their eventual displayed appearance. And yet these same abstract documents, after many transformations and 'typesetting' processes, often emerge in the popular format of Adobe PDF, either for dissemination or archiving. Until recently PDF has been a totally display-based document representation, relying on the underlying PostScript semantics of PDF. Early versions of PDF had no mechanism for retaining any form of abstract document structure but recent releases have now introduced an internal structure tree to create the so called 'Tagged PDF'. This paper describes the development of a plugin for Adobe Acrobat which creates a two-window display. In one window is shown an XML document original and in the other its Tagged PDF counterpart is seen, with an internal structure tree that, in some sense, matches the one seen in XML. If a component is highlighted in either window then the corresponding structured item, with any attendant text, is also highlighted in the other window. Important applications of correctly Tagged PDF include making PDF documents reflow intelligently on small screen devices and enabling them to be read out in correct reading order, via speech synthesiser software, for the visually impaired. By tracing structure transformation from source document to destination one can implement the repair of damaged PDF structure or the adaptation of an existing structure tree to an incrementally updated document.

Journal publishing with Acrobat: the CAJUN project

The publication of material in electronic form should ideally preserve, in a unified document representation, all of the richness of the printed document while maintaining enough of its underlying structure to enable searching and other forms of semantic processing. Until recently it has been hard to find a document representation which combined these attributes and which also stood some chance of becoming a de facto multi-platform standard. This paper sets out experience gained within the Electronic Publishing Research Group at the University of Nottingham in using Adobe Acrobat software and its underlying PDF (Portable Document Format) notation. The CAJUN project1 (CD-ROM Acrobat Journals Using Networks) began in 1993 and has used Acrobat software to produce electronic versions of journal papers for network and CD-ROM dissemination. The paper describes the project's progress so far and also gives a brief assessment of PDF's suitability as a universal document interchange standard.

Enhancing composite Digital Documents Using XML-based Standoff Markup

Document representations can rapidly become unwieldy if they try to encapsulate all possible document properties, ranging from abstract structure to detailed rendering and layout. We present a composite document approach wherein an XMLbased document representation is linked via a shadow tree of bi-directional pointers to a PDF representation of the same document. Using a two-window viewer any material selected in the PDF can be related back to the corresponding material in the XML, and vice versa. In this way the treatment of specialist material such as mathematics, music or chemistry (e.g. via read aloud or play aloud ) can be activated via standard tools working within the XML representation, rather than requiring that application-specific structures be embedded in the PDF itself. The problems of textual recognition and tree pattern matching between the two representations are discussed in detail. Comparisons are drawn between our use of a shadow tree of pointers to map between document representations and the use of a code-replacement shadow tree in technologies such as XBL.