Describes a method for automatically selecting key frames that can be used to represent the total image sequence of visual documents. An abstracting algorithm developed at the National Aeronautics and Space Administration is explained, and extensive examples of abstracted motion sequences are presented. (eight references) (LRW)

Describes research that examined the feasibility of automatically deriving name access points from machine readable title pages of English language monographs. An algorithm for automatic derivation is presented and results using the algorithm are compared to traditionally cataloged name access points. The implications for bibliographic standards…

Reviews the use of signatures to improve the performance of text searching and considers the use of signature trees to further accelerate text searching. Based on the analysis of experimental results, it is concluded that the improvement in text searching performance using signature trees outweighs the increase in required storage. (12 references)…

Describes a project that developed two expert systems for library cataloging using the second edition of the Anglo American Cataloging Rules (AACR2) as a knowledge base. The discussion covers cataloging as interpretation, the structure of AACR2, and the feasibility of using expert systems for cataloging in traditional library settings. (26…

Describes a bottom-up procedure for discovering multivalued dependencies in observed data without knowing a priori the relationships among the attributes. The proposed algorithm is an application of technique designed for learning conditional independencies in probabilistic reasoning; a prototype system for automated database schema design has…

Describes the development and application of an algorithm that generates non-linear representations of English text. It appears that the representation it produces could be quite useful in automatic language processing. (JB)