ISR11 Design Criteria for Automatic Information Systems chapter M. E. Lesk G. Salton Harvard University Gerard Salton Use, reproduction, or publication, in whole or in part, is permitted for any purpose of the United States Government. V-35 I. Term Weights Weighted Word Stems >[OCRerr] Logical Stems Weighted Synonym Classes » Logical Synonym Classes 2. Document Length Full Summaries (2000 words) > Abstracts (150 words) Abstracts (ISO words) » Titles Only 3. Synonym Recognition: Abstracts with Thesaurus )> Abstracts Null Summaries with Thesaurus > Summaries Null 4. Phrase Recognition: Synonym and Phrase > Synonym Recognition Recognition (Thesaurus) only 5. Syntactic Analysis: Syntactic Analysis » Word Stem Match Syntactic Analysis > Synonym Recognition Syntactic Analysis - Statistical Phrase Recognition 6. Term-Term Associations: Stem-Stem Associations > Simple Word Stems Concept-Concept (rhesaurus - Synonym Recognition Class) Associations 7. Manual Indexing: Abstract Stem Matching Index Term Match lndex Term with Thesaurus > Abstracts with Thesorus »> ::much greater than" greater than81 Overoll Evaluation Results - "about equal to" (based on experiments with 4 collections in 3 topic areas) Fig. 16