Discusses automatic indexing methods and describes the development of a prototype computerized abstractor's assistant. Highlights include the text network management system, TEXNET; phrase selection that follows indexing; phrase display, including Boolean capabilities; results of preliminary testing; and availability of TEXNET software. (LRW)

This study proposes a probabilistic model for automatically extracting English noun phrases for indexing or information retrieval. The technique is based on a Markov model, whose initial parameters are estimated by a phrase lookup program with a phrase dictionary, then optimized by a set of maximum entropy parameters. (Author/LRW)

Describes an information-retrieval schema that takes into account linguistic variation based on phrases. Introduces the phrase-retrieval hypothesis to replace the keyword-retrieval hypothesis, and discusses natural-language processing systems; syntactical, morphological, and lexico-semantical normalization; and future work. (Author/LRW)

Discussion of human intellectual indexing versus automatic indexing focuses on automatic indexing. Topics include keyword indexing; negative vocabulary control; counting words; comparative counting and weighting; stemming; words versus phrases; clustering; latent semantic indexing; citation indexes; bibliographic coupling; co-citation; relevance…