It is common practice in the U.S. for courts to require that the parties to a legal case make available to one another all material relevant to the case, including electronically held data and documents. For large corporations, such relevant information may encompass terabytes of e-mail and other files spanning many years. The challenge of E-Discovery in response to a court order is (in a relatively short amount of time) to identify, assemble, individuate, access, categorize, and analyze an organization's electronically held material, segregate all "privileged" material (which can be legally withheld), and present to the court all (and only) the required documents. The techniques needed to accomplish such a task necessarily include search, clustering, classification, filtering, social network analysis, extraction, and more - and no one of these is sufficient. The panel will describe this problem in detail, providing additional background and context on E-Discovery generally, and will explore specific techniques and cases that amply demonstrate why E-Discovery is quintessentially a "CIKM" problem -- multi-disciplinary, multi-technology, and "multi-difficult".

Social networking promises individuals new dimensions of freedom to interact, associate, and give expression to their talents. Recently, systems such as Mechanical Turk have started to facilitate self-organizing collaboration on work-related tasks. Such developments raise interesting questions. Is it possible to create (and sustain) businesses that do not have traditional, formal structure - without traditional "employees"? Can we find and organize (and optimize) talent on the web for task-oriented work - spontaneously and efficiently? How do people relate to one another in possibly evanescent workgroups? One aspect of the challenge in the Social Workspace is understanding and modeling the user behavior and the economic basis for creating, preserving, and exchanging value in the marketplace when workgroup identity, orientations to property, recruiting and managing appropriate talent are not organized under traditional company structures. Another aspect is the technology needed to support virtual organizations and work. The panel will discuss trends in social work and the evolving (scientific) basis of our understanding of new models of workers and organizations.

At the NTCIR-4 workshop, Justsystem Corporation (JSC) and Clairvoyance Corporation (CC) collaborated in the cross-language retrieval task (CLIR). Our goal was to evaluate the performance and robustness of our recently developed commercial-grade CLIR systems for English and Asian languages. The main contribution of this article is the investigation of different strategies, their interactions in both monolingual and bilingual retrieval tasks, and their respective contributions to operational retrieval systems in the context of NTCIR-4. We report results of Japanese and English monolingual retrieval and results of Japanese-to-English bilingual retrieval. In monolingual retrieval analysis, we examine two special properties of the NTCIR experimental design (two levels of relevance and identical queries in multiple languages) and explore how they interact with strategies of our retrieval system, including pseudo-relevance feedback, multi-word term down-weighting, and term weight merging strategies. Our analysis shows that the choice of language (English or Japanese) does not have a significant impact on retrieval performance. Query expansion is slightly more effective with relaxed judgments than with rigid judgments. For better retrieval performance, weights of multi-word terms should be lowered. In the bilingual retrieval analysis, we aim to identify robust strategies that are effective when used alone and when used in combination with other strategies. We examine cross-lingual specific strategies such as translation disambiguation and translation structuring, as well as general strategies such as pseudo-relevance feedback and multi-word term down-weighting. For shorter title topics, pseudo-relevance feedback is a major performance enhancer, but translation structuring affects retrieval performance negatively when used alone or in combination with other strategies. All experimented strategies improve retrieval performance for the longer description topics, with pseudo-relevance feedback and translation structuring as the major contributors.

We conducted a survey of thirty of the approximately 1,700 customers of Justsystem Corporation's knowledge-management applications. Our goal was to discover the kinds of functions that customers hoped to address in their next-generation use of knowledge management technology and to assess the core processes that we will need to deploy in our products to address their desired solutions. In particular, we sought to analyze our customers' requirements along dimensions that take account of both the context of use of the application and its stage in the cycle of knowledge creation and use. As part of our analysis, we were able to classify all customer cases as focused by one or more of three Goals, supported by one or more of eleven technology Means. To establish appropriate categories of use, we exploited the stages of the SECI Model, several other transactional categories of knowledge use, and whether activities were targeted at internal or external users. Through the analysis, we found the typical technology components (Means) for each stage of knowledge creation and use associated with each set of goals. We consider such analysis essential to the task of designing next-generation knowledge-management applications and critical to overcoming the unfortunate tendency of developers to devise solutions that bear little relation to the true needs of users.

No abstract available.

The Web provides the largest, exploitable collection of language use. If we can mine the Web to build abstract models of language use, these models may have many applications. Here we present one example of using the implicit intelligence of language use to solve an important problem for machine translation programs and cross-lingual applications. This problem involves the translation of words written in katakana characters in Japanese. In this paper, we describe techniques of discovering katakana transliteration of English names and of finding English translations of multiword katakana sequences using implicit language models of English and Japanese found on the Web. These techniques were evaluated against human-constructed English-katakana glosses.

We describe an efficient, robust method for selecting and optimizing terms for a classification or filtering task. Terms are extracted from positive examples in training data based on several alternative term-selection algorithms, then combined additively after a simple term-score normalization step to produce a merged and ranked master term vector. The score threshold for the master vector is set via beta-gamma regulation over all the available training data. The process avoids para-meter calibrations and protracted training. It also results in compact profiles for run-time evaluation of test (new) documents. Results on TREC-2002 filtering-task datasets demonstrate substantial improvements over TREC-median results and rival both idealized IR-based results and optimized (and expensive) SVM-based classifiers in general effectiveness.

For cross language information retrieval (CLIR) based on bilingual translation dictionaries, good performance depends upon lexical coverage in the dictionary. This is especially true for languages possessing few inter-language cognates, such as between Japanese and English. In this paper, we describe a method for automatically creating and validating candidate Japanese transliterated terms of English words. A phonetic English dictionary and a set of probabilistic mapping rules are used for automatically generating transliteration candidates. A monolingual Japanese corpus is then used for automatically validating the transliterated terms. We evaluate the usage of the extracted English-Japanese transliteration pairs with Japanese to English retrieval experiments over the CLEF bilingual test collections. The use of our automatically derived extension to a bilingual translation dictionary improves average precision, both before and after pseudo-relevance feedback, with gains ranging from 2.5% to 64.8%.

In this paper, we present a method based on document probes to quantify and diagnose topic structure, distinguishing topics as monolithic, structured, or diffuse. The method also yields a structure analysis that can be used directly to optimize filter (classifier) creation. Preliminary results illustrate the predictive value of the approach on TREC/Reuters-96 topics.

In this paper we examine the learning behavior of a heuristic threshold setting approach to information filtering. In particular, we study how different initial threshold settings and different updating parameter settings affect threshold learning. The results on one of the TREC news databases indicate that (1) learning allows recovery from the inevitable non-optimality of the initial conditions, and (2) a greater “willingness to learn” (expressed by a deliberate lowering of the score threshold in the learning stage) does eventually lead to a higher performance in spite of the expected initial performance penalty.

Information retrieval is an important application area of natural-language processing where one encounters the genuine challenge of processing large quantities of unrestricted natural-language text. This paper reports on the application of a few simple, yet robust and efficient noun-phrase analysis techniques to create better indexing phrases for information retrieval. In particular, we describe an hybrid approach to the extraction of meaningful (continuous or discontinuous) subcompounds from complex noun phrases using both corpus statistics and linguistic heuristics. Results of experiments show that indexing based on such extracted subcompound improves both recall and precision in an information retrieval system. The noun-phrase analysis techniques are also potentially useful for book indexing and automatic thesaurus extraction.

The Canon Group is an informal organization of medical informatics researchers who are working on the problem of developing a "deeper" representation formalism for use in exchanging data and developing applications. Individuals in the group represent experts in such areas as knowledge representation and computational linguistics, as well as in a variety of medical subdisciplines. All share the view that current mechanisms for the characterization of medical phenomena are either inadequate (limited or rigid) or idiosyncratic (useful for a specific application but incapable of being generalized or extended). The Group proposes to focus on the design of a general schema for medical-language representation including the specification of the resources and associated procedures required to map language (including standard terminologies) into representations that make all implicit relations "visible," reveal "hidden attributes," and generally resolve ambiguous or vague references. The Group is proceeding by examining large numbers of texts (records) in medical sub-domains to identify candidate "concepts" and by attempting to develop general rules and representations for elements such as attributes and values so that all concepts may be expressed uniformly.