A Context Vector-Based Self Organizing Map for Information Visualization

A CONTEXT VECTOR-BASED SELF ORGANIZING MAP FOR INFORMATION VISUALIZATION David A. Rushall, Marc R. Ilgen HNC Software, Inc., 5930 Cornerstone Court West, San Diego, CA 92121 USA email: dar@hnc.com, mri@hne.eom ABSTRACT HNC Software, Inc. has developed a system called DOCUVERSE for visualizing the information content of large textual corpora. The system is built around two separate neural network methodologies: context vectors and self organizing maps. Context vectors (CVs) are high dimensional information representations that encode the semantic content of the textual entities they represent. Self organizing maps (SOMs) are capable of transforming an input, high dimensional signal space into a much lower (usually two or three) dimensional output space useful for visualization. Related information themes contained in the corpus, depicted graphically, are presented in spatial proximity to one another. Neither process requires human intervention, nor an external knowledge base. Together, these neural network techniques can be utilized to automatically identi~ the relevant information themes present in a corpus, and present those themes to the user in a intuitive visual form. 1. INTRODUCTION In recent years there has been an explosion in the amount of information available on-line. Much of this explosion has been fueled by the spectacular growth of the Internet and especially the World Wide Web. Along with this spectacular growth has come new challenges for effectively locating on-line information, especially when browsing rather than performing a directed search for a specific piece of information. Key word and linguistically based directed search engines offer some ability to present relevant information to the user, and have resulted in useful Interact products such as Yahoo [1], Lycos [2], and Alta Vista[3]. However, these engines suffer fi'om the fact that they require the user to specify a query of limited length and they offer no visual interface for browsing. To solve the problem of browsing the information space in order to fred information of interest, new techniques for data retrieval and presentation must be developed. HNC has developed an underlying information representation technology and a concept for information visualization that can solve the problem of effectively browsing large textual corpora. As part of HNC's involvement in the ARPA sponsored TIPSTER program, HNC has developed a neural network technique that can learn word level relationships from free text. This capability is based upon an approach called context vectors which encodes the meaning and context of words and documents in the form of unit vectors in a high dimensional vector space. Furthermore, as part of HNC's involvement in the US intelligence community- sponsored P1000 visualization effort, HNC has applied a secondary neural network process, the Self Organizing Map (SOM) [4], which uses the document context vectors to build a visual representation of the information content of the corpus. The combination of these technologies allows users to effectively browse the information space, to locate related documents, and to discover relationships between different themes in the information space. The remainder of this paper is organized as follows. Section 2 presents an overview of both context vectors …

Content truncated to 500 words