Abstract ⎯ This paper describes the development and application of visualisation techniques for users to access and explore information in a digital library effectively and intuitively. Salient semantic structures and citation patterns are extracted from several collections of documents, including the ACM SIGCHI conference proceedings (1995 ⎯ 1997) and ACM Hypertext conference proceedings (1987 ⎯ 1998), using Latent Semantic Indexing and Pathfinder Network Scaling. The unique spatial metaphor leads to a natural combination of search and navigation within the same semantic space in a 3-dimensional virtual world. Author co-citation patterns are visualised through a number of author co-citation maps in attempts to reveal the structure of the field of hypertext, including an overall co-citation map of 367 authors and three periodical maps. These maps highlight predominant research areas in the field. This approach provides a means of transcending the boundaries of collections of documents and visualising more profound patterns in terms of semantic structures and co-citation networks. © 1999 Elsevier Science Ltd. All rights reserved. 1.

An association rule in data mining is an implication of the form X Y where X is a set of antecedent items and Y is the consequent item. For years researchers have developed many tools to visualize association rules. However, few of these tools can handle more than dozens of rules, and none of them can effectively manage rules with multiple antecedents. Thus, it is extremely difficult to visualize and understand the association information of a large data set even when all the rules are available. This paper presents a novel visualization technique to tackle many of these problems. We apply the technology to a text mining study on large corpora. The results indicate that our design can easily handle hundreds of multiple antecedent association rules in a three-dimensional display with minimum human interaction, low occlusion percentage, and no screen swapping. Keywords: Text Visualization, Information Visualization, Text Mining, Data Mining, Association Rule 1 INTRODUCTION Association...

TopCat (Topic Categories) is a technique for identifying topics that recur in articles in a text corpus. Natural language processing techniques are used to identify key entities in individual articles, allowing us to represent an article as a set of items. This allows us to view the problem in a database/data mining context: Identifying related groups of items. This paper presents a novel method for identifying related items based on "traditional" data mining techniques. Frequent itemsets are generated from the groups of items, followed by clusters formed with a hypergraph partitioning scheme. We present an evaluation against a manually-categorized "ground truth" news corpus showing this technique is effective in identifying "topics" in collections of news articles.

Effective visualization is an important aspect of active data mining. In the context of association rules, this need has been driven by the large amount of rules produced from a run of the algorithm. To be able to address real user needs, the rules need to be summarized and organized so that it can be interpreted and applied in a timely manner. In this paper, we propose two visualization techniques that is an improvement over those used by existing data mining packages. In particular, we address the visualization of "differences" in the set of rules due to incremental changes in the data source. We show that visualization in this aspect is important to active data mining as it uncovers new insights not possible from inspecting individual data mining results.

Abstract: Virtual Environments present opportunities for novel interaction with and visualization of abstract data. However, the inherent difficulties of rendering three dimensional environments, as well as bottlenecks in traditional virtual reality systems make real-time manipulation of realistic volumes of data difficult. We present an architectural solution that addresses many of these problems, and Q-Space, an exemplar virtual environment in which large amounts of data can be manipulated in real-time in a three dimensional world 1.

The design of consistent information visualisation models is a key component in the development of visual data mining methods. However, it is a challenging activity to find the methods, techniques and corresponding tools that are suitable for specific visual mining task, or a particular type of data. The comparison of visualisation techniques across different designs is not a trivial problem. This paper discusses the issues connected with the development of consistent approach to formal development, evaluation and comparison of visualisation methods. Proposed formal approach is illustrated with examples of development of a visualisation model for data from the area of team collaboration in virtual environments and evaluation of visualisation schemes for the results of text analysis. The papers concludes with the discussion of the limitations of the proposed approach and future directions of the research and development of proposed approach.

Abstract: Self-Organising Maps are utilised in many data mining and knowledge management applications. Although various visualisations have been proposed for SOM, these techniques lack in distinguishing between the items mapped to the same unit. Here we present a novel technique for the visualisation of Self-Organising Maps that displays inputs not in the centre of the map units, but shifts them towards the closest neighbours, the degree of the movement depending on the similarity to the neighbours. The night-sky visualisation facilitates better understanding of the underlying data. We report results from applying our method on two synthetic and a real-life data set.

With the rapid expansion of scientific research, the ability to effectively find or integrate new domain knowledge in the sciences is proving increasingly difficult. The development of methods and tools for assisting researchers to effectively extract problem-oriented knowledge from heterogeneous and massive information sources, and for using this knowledge in problem-solving is one of the most fundamental research directions for the information and computer sciences today. There is a need for new tools to support more precise identification of relevant research articles and provide visual clues regarding relationships among the document sets. We present the Telemakus system in which aggregated citation information and extracted research findings are displayed in a schema-based document surrogate and an interactive mapping tool provides graphical displays of research inter-relationships from documents across a domain. This system is an innovative approach to creating useful and precise document surrogates and may re-conceptualize the way we currently represent, retrieve, and assimilate research findings from the published literature. Keywords:

As more and more data are collected and stored in multiple databases, data mining over different relational DBMSs is becoming increasingly important. Usually the data sets are stored in some different DBMSs, such as DB2, Oracle, Sybase, etc. Data mining software should be able to use the data from all of these DBMSs. In this thesis we use JDBC (Java Database Connectivity) to achieve this goal. The test DBMSs we used are Oracle and DB2. We implement the association rule algorithm in the form of SQL queries. Three algorithms in SQL-92 (K-Way Join, 2-Groupby, and Subquery) and three algorithms in SQL-OR (Vertical, GatherJoin and GatherJoin Variant) are implemented and compared to each other based on their performance on different kinds of synthetic generated data sets. Scale-Up experiments have also been conducted for these six approaches. We develop an association rule visualization system, which includes tabular form and three-dimensional graphics. By providing the user sorting and filtering abilities, this rule visualization system makes it flexible and efficient for the user to manage and understand the association rules. As a result, this visualization system becomes an essential part of our association rule software. Finally, we compare our association rule software with one of the commercial data mining tools (Intelligent Miner from IBM) in various aspects, such as data accessibility, user interface, input/output, and rule visualization.

Abstract. Identification of significant differences in sets of data is a common task of data mining. This paper describes a novel visualization technique that allows the user to interactively explore and analyze differences in mean values of analyzed attributes. Statistical tests of hypotheses are used to identify the significant differences and the results are then presented using Hasse diagrams. The presented technique has been tested on real data coming from pedagogical tests focused on evaluation of mathematical skills of secondary school students in Czech Republic. The results show that the proposed tool provides comprehensible representation of the data. 1