. Unlabeled document collections are becoming increasingly common and available; mining such data sets represents a major contemporary challenge. Using words as features, text documents are often represented as high-dimensional and sparse vectors--a few thousand dimensions and a sparsity of 95 to 99% is typical. In this paper, we study a certain spherical k-means algorithm for clustering such document vectors. The algorithm outputs k disjoint clusters each with a concept vector that is the centroid of the cluster normalized to have unit Euclidean norm. As our first contribution, we empirically demonstrate that, owing to the high-dimensionality and sparsity of the text data, the clusters produced by the algorithm have a certain "fractal-like" and "self-similar" behavior. As our second contribution, we introduce concept decompositions to approximate the matrix of document vectors; these decompositions are obtained by taking the least-squares approximation onto the linear subspace spanned...

Abstract not found

In a peer-to-peer (P2P) system, nodes typically connect to a small set of random nodes (their neighbors), and queries are propagated along these connections. Such query flooding tends to be very expensive. We propose that node connections be influenced by content, so that for example, nodes having many "Jazz" files will connect to other similar nodes. Thus, semantically related nodes form a Semantic Overlay Network (SON). Queries are routed to the appropriate SONs, increasing the chances that matching files will be found quickly, and reducing the search load on nodes that have unrelated content. We have evaluated SONs by using an actual snapshot of music-sharing clients. Our results show that SONs can significantly improve query performance while at the same time allowing users to decide what content to put in their computers and to whom to connect.

To cluster increasingly massive data sets that are common today in data and text mining, we propose a parallel implementation of the k-means clustering algorithm based on the message passing model. The proposed algorithm exploits the inherent data-parallelism in the k-means algorithm. We analytically show that the speedup and the scaleup of our algorithm approach the optimal as the number of data points increases. We implemented our algorithm on an IBM POWERparallel SP2 with a maximum of 16 nodes. On typical test data sets, we observe nearly linear relative speedups, for example, 15.62 on 16 nodes, and essentially linear scaleup in the size of the data set and in the number of clusters desired. For a 2 gigabyte test data set, our implementation drives the 16 node SP2 at more than 1.8 gigaflops. Keywords: k-means, data mining, massive data sets, message-passing, text mining. 1 Introduction Data sets measuring in gigabytes and even terabytes are now quite common in data and text minin...

We address the problem of integrating documents from different sources into a master catalog. This problem is pervasive in web marketplaces and portals. Current technology for automating this process consists of building a classifier that uses the categorization of documents in the master catalog to construct a model for predicting the category of unknown documents. Our key insight is that many of the data sources have their own categorization, and classification accuracy can be improved by factoring in the implicit information in these source categorizations. We show how a Naive Bayes classification can be enhanced to incorporate the similarity information present in source catalogs. Our analysis and empirical evaluation show substantial improvement in the accuracy of catalog integration. Keywords: Classification, Categorization, Data Mining, Catalog Integration, Web Portals, Web Marketplaces 1.

Browsing accounts for much of people's interaction with digital libraries, but it is poorly supported by standard search engines. Conventional systems often operate at the wrong level, indexing words when people think in terms of topics, and returning documents when people want a broader view. As a result, users cannot easily determine what is in a collection, how well a particular topic is covered, or what kinds of queries will provide useful results. We have built

When people use computer-based tools to find answers to general questions, they often are faced with a daunting list of search results or "hits" returned by the search engine. Many search tools address this problem by helping users to make their searches more specific. However, when dozens or hundreds of documents are relevant to their question, users need tools that help them to explore and to understand their search results, rather than ones that eliminate a portion of those results. In this paper, we present DynaCat, a tool that dynamically categorizes search results into a hierarchical organization by using knowledge of important kinds of queries and a model of the domain terminology. Results from our evaluation show that DynaCat helps users find answers to those important types of questions more quickly and easily than when they use a relevance -ranking system or a clustering system. Introduction Current information-retrieval tools usually return results that cons...

Abstract. We describe Athena: a system for creating, exploiting, and maintaining a hierarchy of textual documents through interactive miningbased operations. Requirements of any such system include speed and minimal end-user e ort. Athena satis es these requirements through linear-time classi cation and clustering engines which are applied interactively to speed the development of accurate models. Naive Bayes classi ers are recognized to be among the best for classifying text. We show that our specialization of the Naive Bayes classi er is considerably more accurate (7 to 29 % absolute increase in accuracy) than a standard implementation. Our enhancements include using Lidstone's law of succession instead of Laplace's law, under-weighting long documents, and over-weighting author and subject. We also present a new interactive clustering algorithm, C-Evolve, for topic discovery. C-Evolve rst nds highly accurate cluster digests (partial clusters), gets user feedback to merge and correct these digests, and then uses the classi cation algorithm to complete the partitioning of the data. By allowing this interactivity in the clustering process, C-Evolve achieves considerably higher clustering accuracy (10 to 20 % absolute increase in our experiments) than the popular K-Means and agglomerative clustering methods. 1

In this article, we report our implementation and comparison of two text clustering techniques. One is based on Ward's clustering and the other on Kohonen's Self-organizing Maps. We have evaluated how closely clusters produced by a computer resemble those created by human experts. We have also measured the time that it takes for an expert to "clean up" the automatically produced clusters. The technique based on Ward's clustering was found to be more precise. Both techniques have worked equally well in detecting associations between text documents. We used text messages obtained from group brainstorming meetings.

This paper describes keyword-based Web page categorization. Our goal is to embed our categorization technique into information retrieval (IR) systems to facilitate the end-users' search task. In such systems, search results must be categorized faster, while keeping accuracy high. Our categorization system uses a knowledge base (KB) to assign categories to Web pages. The KB contains a set of characteristic keywords with weights by category, and is automatically generated from training texts. With the keyword-based approach, the algorithms to extract keywords and assign weights to them should be considered, because the algorithms affect strongly both categorization accuracy and processing speed. Furthermore, we must take two characteristics of Web pages into account: (1) the text length is very variable, which makes it harder to use statistics such as word frequency to calculate keyword weights, and (2) a huge number of distinct words are used, which makes the KB bigger and therefore pro...