Investigates the feasibility of using a distance measure for automatic sequential document classification. This property of the distance measure is used to design a sequential classification algorithm which classifies key words and analyzes them separately in order to assign primary and secondary classes to a document. (VT)

Reviews aspects of the relationship between machine learning and information retrieval. Highlights include learning programs that extend from knowledge-sparse learning to knowledge-rich learning; the role of the thesaurus; knowledge bases; artificial intelligence; weighting documents; work frequency; and merging classification structures. (78…

Presents a general mathematical framework for information retrieval, in which query evaluation is seen as a mapping from the free term algebra built by query descriptors and weights to the semantic algebra of evaluation value. Competing mathematical models that try to describe how retrieval systems behave are reviewed. (Author/MBR)

Presents principles for determining profile of a system user based on analysis of co-occurrence of index terms in queries and pertinent documents. A procedure for determining index term weights, the information value of which depends on the degree of homogeneity of system users, is introduced. Three references are cited. (EJS)

This paper deals with quantitative evaluation of the effects of erroneous performance by stemming and similar clustering algorithms in information retrieval systems. (Author)

An information retrieval system which aspires to retrieval of relevant documents should have second stage which will order topical set to provide maximum informativeness to requestor. A two-stage system concerned only with topicality can be iterated to generate a high recall set and discard imprecise documents without user input. (EJS)

Explains the use of genetic algorithms to combine results from multiple query evaluations to improve relevance in information retrieval. Discusses niching techniques, relevance feedback techniques, and evolution heuristics, and compares retrieval results obtained by both genetic multiple query evaluation and classical single query evaluation…

Describes SCISOR (System for Conceptual Information Summarization, Organization and Retrieval), a prototype intelligent information retrieval system that extracts useful information from large bodies of text. It overcomes limitations of linguistic coverage by applying a text processing strategy that is tolerant of unknown words and gaps in…

Highlights semantic links and shows how the semantic content of hypertext links can be used for information retrieval. Discussion includes indexing and retrieval algorithms that exploit link content and node content; retrieval strategies exploiting semantic links, including conventional retrieval and constrained spreading activation techniques;…

Discusses end-user searching in Boolean information retrieval systems considers the role of search intermediaries and proposes a model of user preferences that incorporates a user's profile. Highlights include document representation; information queries; document output ranking; calculating user profiles; and selecting documents for a local…

Describes a process of morpho-syntactic language analysis for information retrieval. Tree Structured Analytics (TSA) used for text representation is summarized; the matching process developed for such structures is outlined with an example appended; and experiments carried out to evaluate the effectiveness of TSA matching are discussed. (26…

Describes a project that improved the quality of BCIN, the bibliographic database of the Conservation Information Network, by correcting data fields and removing duplicate records. The database is described, deduplication theory is discussed and relevant literature is reviewed, algorithm development and testing is described, and results of…

Discusses information retrieval and the use of binary trees as a fast access method for search strategies such as hashing. Proposes new methods based on compact binary trees that provide faster access and more compact storage, explains the theoretical basis, and confirms the validity of the methods through empirical observations. (LRW)

Presents a heuristic, attribute-based, noise-tolerant data mining program, HCV (Version 2.0) based on the newly-developed extension matrix approach. Outlines some techniques implemented in the HCV program for noise handling and discretization of continuous domains; an empirical comparison shows that rules generated by HCV are more compact than the…

Large-scale, on-line information storage and retrieval systems pose numerous problems above those encountered by smaller systems. A step toward the solution of these problems is presented along with several demonstrations of feasibility and advantages. The methodology on which this solution is based is that of a posteriori automatic classification…