Extracting semantic relationships between terms from PC documents and its applications to web search personalization

Authors:
Hiroaki Ohshima;Satoshi Oyama;Katsumi Tanaka
Affiliations:
Department of Social Informatics, Graduate School of Informatics, Kyoto University, Kyoto, Japan;Department of Social Informatics, Graduate School of Informatics, Kyoto University, Kyoto, Japan;Department of Social Informatics, Graduate School of Informatics, Kyoto University, Kyoto, Japan
Venue:
APWeb'06 Proceedings of the 8th Asia-Pacific Web conference on Frontiers of WWW Research and Development
Year:
2006

Citing 4
Cited 0

Deriving concept hierarchies from text

Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval
Haystack: per-user information environments

Proceedings of the eighth international conference on Information and knowledge management
Inferring hierarchical descriptions

Proceedings of the eleventh international conference on Information and knowledge management
WorkWare: WWW-based Chronological Document Organizer

APCHI '98 Proceedings of the Third Asian Pacific Computer and Human Interaction

Quantified Score

Hi-index	0.00

Visualization

Abstract

A method is described for extracting semantic relationships between terms appearing in documents stored on a personal computer; these relationships can be used to personalize Web search. It is based on the assumption that the information a person stores on a personal computer and the directory structure in the PC reflect, to some extent, the person’s knowledge, ideology, and concept classification. It works by identifying semantic relationships between the terms in documents on the PC; these relationships reflect the person’s relative valuation of each term in a pair. The directory structure is examined to identify the deviations in the appearance of the terms within each directory. These deviations are then used to identify the relationships between the terms. Four relationships are defined: broad, narrow, co-occurrent, and exclusive. They can be used to personalize Web search through, for example, expansion of queries and re-ranking of search results.