Discussion of evaluation methodology in information retrieval focuses on the average precision over a set of fixed recall values in an effort to evaluate the retrieval effectiveness of a search algorithm. Highlights include a review of traditional evaluation methodology with examples; and a statistical inference methodology called bootstrap.…

Describes the EurekaSeek bibliometric technique for automated linked-literature analysis. EurekaSeek uses medical subject headings instead of text analysis in a fully automated search process, thereby eliminating the reliance on expert input during the process of linking literatures. In this paper, EurekaSeek is tested by retroactively examining…

Examines the differences between Web image and textual queries, and attempts to develop an analytic model to investigate their implications for Web image retrieval systems. Provides results that give insight into Web image searching behavior and suggests implications for improvement of current Web image search engines. (AEF)

A task-oriented, comparative evaluation between four Web retrieval systems was performed; two using query-biased summarization, and two using the standard ranked titles/abstracts approach. Results indicate that query-biased summarization techniques appear to be more useful and effective in helping users gauge document relevance than the…

Discussion of text analysis and information retrieval and measurement of document similarity focuses on a new text manipulation system called FA (field association)-Sim that is useful for retrieving information in large heterogeneous texts and for recognizing content similarity in text excerpts. Discusses recall and precision, automatic indexing…

Document retrieval tasks composed of human relevance judgments need test sets that allow comparison of human retrieval and automated retrieval and that place absolute limits on performance by any method. Results suggest that it should be possible to improve retrieval system performance within recommended limits by incorporating specific subject…

To determine if known-item searching was possible with four of the Web's most popular search engines (Alta Vista, Excite, Infoseek Guide, and Lycos A2Z), searches for 39 known Web sites were performed. The search engines were then ranked according to retrieval of relevant and exact matches and the order in which the search results appeared. (PEN)

Describes a theoretic process to apply Bahadur-Lazarsfeld expansion (BLE) to general probabilistic models and the state-of-the-art 2-Poisson model. Through experiments on two standard document collections, one in Korean and one in English, it is demonstrated that incorporation of term dependences using BLE significantly contributes to performance…

Reports on a test of the applicability of the theory of General Scalability to documents in terms of transitivity, substitutability, and independence. Implications for the construction of test collections in information retrieval research are examined. (24 references) (EAM)

Describes an examination of the effectiveness of retrieving and ranking records from citation indexes based on the number of references in common with an identified and relevant record. It is concluded that this retrieval method is potentially useful, especially in those searches backward and forward in time that are enabled by the design of…

Discusses the interpretation of nominal compounds in the context of natural language processing for information retrieval. Knowledge-intensive algorithms that can successfully interpret compounds found in technical documents are described, and experiments are reported that indicate these algorithms may be unnecessary for improving retrieval…

Describes a study that supported two hypotheses concerning the use of anaphors in information retrieval: first, that anaphors tend to refer to integral concepts rather than peripheral concepts; second, that various term weighting schemes are affected differently by anaphoric resolution. The implications for the use of anaphoric resolution for…

Reports on two experiments investigating the relationship between topical subject expertise and the assessment of relevance. Results indicate that topical subject expertise enhances effects of the semantic distance model (SDM), and that relevance assessment is a contingent, psychological phenomenon that is part of the human-perception task of…

Proposes an argumentation-based method for information retrieval that involves representing a scholarly article by means of rhetorical structure rather than by a semantic representation of content. Topics addressed include problems with indexing practices, the argumentative nature of articles, rhetoric as an information categorizing schema, and…

The fifth in a series of articles on microcomputer software for information storage and retrieval describes characteristics of and presents test results for seven programs in the following categories: personal information managers, including 3by5/RediReference, askSam, Dayflo Tracker, and Ize; Personal Librarian, a relevance ranking program; and…