Asymmetric Information Distances for Automated Taxonomy Construction
Wei Lee Woon
Masdar Institute of Science and Technology (MIST)
Massachusetts Institute of Technology (MIT) - Sloan School of Management
August 25, 2008
MIT Sloan Research Paper No. 4712-08
A novel method for automatically constructing taxonomies for specific research domains is presented. The proposed methodology uses term co-occurence frequencies as an indicator of the semantic closeness between terms. To support the automated creation of taxonomies or subject classifications we present a simple modification to the basic distance measure, and describe a set of procedures by which these measures may be converted into estimates of the desired taxonomy. To demonstrate the viability of this approach, a pilot study on renewable energy technologies is conducted, where the proposed method is used to construct a hierarchy of terms related to alternative energy. These techniques have many potential applications, but one activity in which we are particularly interested is the mapping and subsequent prediction of future developments in the technology and research.
Number of Pages in PDF File: 13
Keywords: Taxonomy Construction, Asymmetric Informationworking papers series
Date posted: August 27, 2008
© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo4 in 0.235 seconds