Full-text

URL: http://www-eksl.cs.umass.edu/papers/98-20.ps

Cached: PDF-602K    PS-2455K    PS.gz-962K

SAVE AS an easy-to-recall long filename:

Filename format: author--year--title   PDF-602K    PS-962K    :: About GZip'd PS

Filename format: author--year--title--journal|proceedings|...--pages   PDF-602K    PS-962K

Related links

CiteSeer: http://citeseer.ist.psu.edu/cohen98growing.html

Web search: Google Web Search   ::   Google Scholar

Within this site: References (13)

Paper at a Glance

2
Growing Ontologies
Paul R. Cohen
Computer Science Technical Report 98­20
Experimental Knowledge Systems Laboratory
Computer Science Department, Box 34610
Lederle Graduate Research Center
University of massachusetts
Amherst, MA 01003­4610

Abstract Conceptual structures or ontologies are usually built by hand by skilled knowledge engineers. This paper presents a theory of how conceptual structure may be acquired by an intelligent agent interacting with its environment in an unsupervised way. Categories of activities are learned, then abstractions over these categories result in concepts. The entire conceptual structure is based on activities. The meanings of concepts and of conceptualizations of activities are discussed. Systems that implement aspects of the theory are presented, and their general characteristics described.  3
Introduction The subject of this paper is the foundation of conceptual systems. In artificial intelligence, conceptual systems are sometimes called ontologies and they are usually built by highly­trained knowledge engineers. It's less clear how humans acquire conceptual systems, or rather, the scope of the innate endowment is unclear. In any case, if machines could acquire conceptual knowledge with the same facility as humans, then AI would be much better off. Many people think that our conceptual system is a prerequisite for all sorts of intelligent processes, from analogical reasoning to natural language understanding, from reasoning under uncertainty to computer­assisted cooperative work. Lenat and Feigenbaum (1987) promised us that a sufficiently large corpus of common sense knowledge would "go critical" and start learning, by reading, autonomously. That hasn't happened yet, but there's no denying the dream of a machine that knows roughly what we know, organized roughly as we organize it, with roughly the same values and motives as we have. It makes sense, then, to ask how this knowledge is acquired by humans and how might it be ...

@techreport{cohen98growingOntologies,
  author={Paul R. Cohen},
  title={Growing Ontologies},
  year={1998},
  institution={Computer Science Department, University of Massachusetts at Amherst},
  note={Technical Report 98-20},
  url={http://www.isrl.uiuc.edu/~amag/langev/paper/cohen98growingOntologies.html}
}

HOME   ::  Technical Report List   ::   Technical Report	Comments to: junwang4 you-know-at gmail.com	Last update: 10/27/09