This article describes the implementation of a system that is able to organize vast document collections according to textual similarities. It is based on the Self-Organizing Map (SOM) algorithm. As the feature vectors for the documents we use statistical representations of their vocabularies. The main goal in our work has been to scale up the SOM algorithm to be able to deal with large amounts of high-dimensional data. In a practical experiment we mapped 6,840,568 patent abstracts onto a 1,002,240-node SOM. As the feature vectors we used 500-dimensional vectors of stochastic figures obtained as random projections of weighted word histograms. Keywords Data mining, exploratory data analysis, knowledge discovery, large databases, parallel implementation, random projection, Self-Organizing Map (SOM), textual documents. I. Introduction A. From simple searches to browsing of self-organized data collections Locating documents on the basis of keywords and simple search expressions is a c...

Analyzing and monitoring logs that portray system, user, and network activity is essential to meet the requirements of high security and optimal resource availability. While most systems now possess satisfactory logging facilities, the tools to monitor and interpret such event logs are still in their infancy. This paper describes an approach to relieve system and network administrators from manually scanning sequences of log entries. An experimental system based on unsupervised neural networks and spring layouts to automatically classify events contained in logs is explained, and the use of complementary information visualization techniques to visually present and interactively analyze the results is then discussed. The system we present can be used to analyze past activity as well as to monitor real-time events. We illustrate the system’s use for event logs generated by a firewall, however it can be easily coupled to any source of sequential and structured event logs.

We present an empirical study in which we investigated group vs. individual performance with collaborative information visualization environments (CIVEs), the effects of system transparency on users' performance and the effects of different collaborative settings on CIVE usage. Subjects searched for findings with CIVEs, working either alone, in a collocated dyad using a shared electronic whiteboard, or in a remote dyad using application sharing. Groups answered more questions correctly and took less time with the more transparent CIVE than groups using the less transparent CIVE. We interpret our results to mean that groups have better self-corrective abilities when the system is transparent. We present a stage model to explain the collaborative process of using CIVEs, which accounts for task type, collaborative setting, and system transparency.

We describe an attempt to overcome information overload through information visualization --- in a particular domain, group memory. A brief review of information visualization is followed by a brief description of our methodology. We Z. discuss our system, which uses multidimensional scaling MDS to visualize relationships between documents, and which Z. we tested on 60 subjects, mostly students. We found three important and statistically significant differences between task performance on an MDS-generated display and on a randomly generated display. With some qualifications, we conclude that MDS speeds up and improves the quality of manual classification of documents and that the MDS display agrees with subject perceptions of which documents are similar and should be displayed together. q 1999 Elsevier Science B.V. All rights reserved.

This paper is part of a final deliverables of independent study conducted during summer 2005 quarter

This paper describes a study to explore the integration of Group Support Systems (GSS) and Artificial Intelligence (AI) technology to provide solicitation, analytical, visualization and sense-making support for attitudes from large distributed marketing focus groups. The paper describes two experiments and the concomitant evolutionary design and development of an attitude analysis process and the MindMine Comment Analysis Tool.

Web 2.0 has great potential to serve as a public sphere (Habermas, 1974; Habermas, 1989) – a distributed arena of voices where all who want to do so can participate. A well-functioning public sphere is important for pluralistic decision-making at many levels, ranging from small organizations to society at large. In this paper, we analyze the capability of the blogosphere in its current form to support such a role. This analysis leads to the identification of the principal issues that prevent the blogosphere from realizing its full potential as a public sphere. Most significantly, we propose that the sheer volume of content overwhelms blog readers, forcing them to restrict themselves to only a small subset of valuable content. This ultimately reduces their level of informedness. Based on past research on managing discourse, we propose four design artifacts that would alleviate these issues: a communal repository, textual clustering, visual cues, and a participation facility for blog users. We present a prototype system, called FeedWiz, which implements several of these design artifacts. Based on this initial design, we formulate a research agenda for the creation of new tools that effectively harness the potential of the growing body of user-generated content in the blogosphere and beyond.

collaborative writing tools and processes to enhance eWriting in an eGovernment setting