|
|
|
@inproceedings{DBLP:conf/vldb/MontgomeryR81,
author = {Christine A. Montgomery and
Enrique H. Ruspini},
title = {The Active Information System: A Data-Driven System for the Analysis
of Imprecise Data},
booktitle = {Very Large Data Bases, 7th International Conference, September
9-11, 1981, Cannes, France, Proceedings},
publisher = {IEEE Computer Society},
year = {1981},
pages = {376-384},
ee = {db/conf/vldb/MontgomeryR81.html},
crossref = {DBLP:conf/vldb/81},
bibsource = {DBLP, http://dblp.uni-trier.de}
}
BibTeX
The concept of Active information System is introduced. Active Information Systems are collections of autonomous, data-driven tools operating in a data base environment and seeking to produce results that enhance the understanda- bility of the contents of the data base. The activities of the corresponding multiple concurrent processes are coordi- nated by controi mechanisms that perform in a goal oriented manner. The data models supported by Active Systems are characterized by the existence of numerous data elements that are known only in imprecise or uncertain terms.
The different types of components of Active Systems are identified and their organization into a prototypical architec- ture is described. Emphasis is placed on algorithms that assist users in hypothesis generation and verification tasks. These algorithms are based on concepts from data base technology, knowledge based systems, system science and multivalued logic.
The technological approaches utilized in the development of an Active System testbed to treat problems of natural language understanding and uncertainty representation are discussed in detail.
Copyright © 1981 by The Institute of Electrical and Electronic Engineers, Inc. (IEEE). Abstract used with permission.