The nearest neighbor problem is the following: Given a set of n points P = fp 1 ; : : : ; png in some metric space X, preprocess P so as to efficiently answer queries which require finding the point in P closest to a query point q 2 X. We focus on the particularly interesting case of the d-dimensional Euclidean space where X = ! d under some l p norm. Despite decades of effort, the current solutions are far from satisfactory; in fact, for large d, in theory or in practice, they provide little improvement over the brute-force algorithm which compares the query point to each data point. Of late, there has been some interest in the approximate nearest neighbors problem, which is: Find a point p 2 P that is an ffl-approximate nearest neighbor of the query q in that for all p 0 2 P , d(p; q) (1 + ffl)d(p 0 ; q). We present two algorithmic results for the approximate version that significantly improve the known bounds: (a) preprocessing cost polynomial in n and d, and a trul...

: The web harbors a large number of communities -- groups of content-creators sharing a common interest -- each of which manifests itself as a set of interlinked web pages. Newgroups and commercial web directories together contain of the order of 20000 such communities; our particular interest here is on emerging communities -- those that have little or no representation in such fora. The subject of this paper is the systematic enumeration of over 100,000 such emerging communities from a web crawl: we call our process trawling. We motivate a graph-theoretic approach to locating such communities, and describe the algorithms, and the algorithmic engineering necessary to find structures that subscribe to this notion, the challenges in handling such a huge data set, and the results of our experiment. Keywords: web mining, communities, trawling, link analysis 1. Overview The web has several thousand well-known, explicitly-defined communities -- groups of individuals who share a common int...

Given two documents A and B we define two mathematical notions: their resemblance r(A, B)andtheircontainment c(A, B) that seem to capture well the informal notions of "roughly the same" and "roughly contained." The basic idea is to reduce these issues to set intersection problems that can be easily evaluated by a process of random sampling that can be done independently for each document. Furthermore, the resemblance can be evaluated using a fixed size sample for each document.

. 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...

Application of data mining techniques to the World Wide Web, referred to as Web mining, has been the focus of several recent research projects and papers. However, there is no established vocabulary, leading to confusion when comparing research e orts. The term Web mining has been used intwo distinct ways. The rst, called Web content mining in this paper, is the process of information discovery from sources across the World Wide Web. The second, called Web usage mining, is the process of mining for user browsing and access patterns. In this paper we de ne Web mining and present an overview of the various research issues, techniques, and development e orts. We brie y describe WEBMINER, a system for Web usage mining, and conclude this paper by listing research issues. 1

In this paper we show several results obtained by combining the use of stable distributions with pseudorandom generators for bounded space. In particular: ffl we show how to maintain (using only O(log n=ffl 2 ) words of storage) a sketch C(p) of a point p 2 l n 1 under dynamic updates of its coordinates, such that given sketches C(p) and C(q) one can estimate jp \Gamma qj 1 up to a factor of (1 + ffl) with large probability. This solves the main open problem of [10]. ffl we obtain another sketch function C 0 which maps l n 1 into a normed space l m 1 (as opposed to C), such that m = m(n) is much smaller than n; to our knowledge this is the first dimensionality reduction lemma for l 1 norm ffl we give an explicit embedding of l n 2 into l n O(log n) 1 with distortion (1 + 1=n \Theta(1) ) and a nonconstructive embedding of l n 2 into l O(n) 1 with distortion (1 + ffl) such that the embedding can be represented using only O(n log 2 n) bits (as opposed to at least...

Users of Web search engines are often forced to sift through the long ordered list of document "snippets" returned by the engines. The IR community has explored document clustering as an alternative method of organizing retrieval results, but clustering has yet to be deployed on most major search engines. The NorthernLight search engine organizes its output into "custom folders" based on pre-computed document labels, but does not reveal how the folders are generated or how well they correspond to users' interests. In this paper, we introduce Grouper -- an interface to the results of the HuskySearch meta-search engine, which dynamically groups the search results into clusters labeled by phrases extracted from the snippets. In addition, we report on the first empirical comparison of user Web search behavior on a standard ranked-list presentation versus a clustered presentation. By analyzing HuskySearch logs, we are able to demonstrate substantial differences in the number of documents f...

Caching in the World Wide Web is based on two critical assumptions: that a significant fraction of requests reaccess resources that have already been retrieved; and that those resources do not change between accesses. We tested the validity of these assumptions, and their dependence on characteristics of Web resources, including access rate, age at time of reference, content type, resource size, and Internet top-level domain. We also measured the rate at which resources change, and the prevalence of duplicate copies in the Web. We quantified the potential benefit of a shared proxycaching server in a large environment by using traces that were collected at the Internet connection points for two large corporations, representing significant numbers of references. Only 22% of the resources referenced in the traces we analyzed were accessed more than once, but about half of the references were to those multiplyreferenced resources. Of this half, 13% were to a resource that had been modifi...

This paper introduces a technique for mining a collection of user transactions with an Internet search engine to discover clusters of similar queries and similar URLs. The information we exploit is "clickthrough data": each record consists of a user's query to a search engine along with the URL which the user selected from among the candidates offered by the search engine. By viewing this dataset as a bipartite graph, with the vertices on one side corresponding to queries and on the other side to URLs, one can apply an agglomerative clustering algorithm to the graph's vertices to identify related queries and URLs. One noteworthy feature of the proposed algorithm is that it is "content-ignorant"---the algorithm makes no use of the actual content of the queries or URLs, but only how they co-occur within the clickthrough data. We describe how to enlist the discovered clusters to assist users in web search, and measure the effectiveness of the discovered clusters in the Lycos search engine...

Often, in the real world, entities have two or more representations in databases. Duplicate records do not share a common key and/or they contain errors that make duplicate matching a dif cult task. Errors are introduced as the result of transcription errors, incomplete information, lack of standard formats or any combination of these factors. In this article, we present a thorough analysis of the literature on duplicate record detection. We cover similarity metrics that are commonly used to detect similar eld entries, and we present an extensive set of duplicate detection algorithms that can detect approximately duplicate records in a database. We also cover multiple techniques for improving the ef ciency and scalability of approximate duplicate detection algorithms. We conclude with a coverage of existing tools and with a brief discussion of the big open problems in the area.