Automatic Indexing: An Experimental Inquiry
Journal of the ACM (JACM)
Semantic indexing of a competence map to support scientific collaboration in a research community
IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
| Hi-index | 0.00 |
As more organizations begin to deploy taxonomies for categorization and faceted search, the cost of producing these knowledge models is becoming the largest expense on a project. At a cost of 200 - 300 dollars per topic, manually developing subject area taxonomies does not scale for any but the smallest of projects. This paper will discuss an approach called Orthogonal Corpus Indexing ( OCI ). OCI leverages existing published knowledge in the subject area of the taxonomy model. This knowledge is algorithmically mapped into multiple taxonomies via the OCI algorithm. The resulting taxonomy costs are 1/ 100th of the cost of manual methods and are created with embedded rule sets for categorization engines. This paper will discuss the theory of OCI, its practical use as well as examples of knowledge management techniques that are possible when taxonomies are large, detailed and inexpensive.