Home // IMMM 2013, The Third International Conference on Advances in Information Mining and Management // View article

Social Network-based Entity Extraction for People Ontology

Authors:
Tina Tian
Soon Ae Chun

Keywords: social networks; ontology; mining from social networks; semantic Web search

Abstract:
When users want to search people, search engines face two basic challenges. One challenge is due to the fact that there are many people (entities) with the same name, i.e. a homonym problem. The other is an entity linking issue, where several words are linked to the same person. The homonyms create the search results with a long list of hits with mingled information of the different person with the same name. The end users need to sift through the documents that fit their needs. To improve the ambiguous search arising from homonyms, we previously implemented an Ontology-Supported Web Search System (OSWS) that utilizes an ontology to disambiguate the search term and that provides search results in different possible categories that a search term may belong to. For a prototype of the OSWS system, we developed an ontology by mining person names and retrieving data from resources such as DBpedia. However, DBpedia is incomplete and often outdated. In this paper, we extend our approach to using social networks for building a People Ontology (PO). Specifically, personal profile attributes and their values of famous people are extracted from public social networks pages, cleaned and mapped to the ontology, resulting in a significant increase of the domain coverage achieved by the People Ontology to support the Ontology-Supported Web Search System.

Pages: 65 to 71

Copyright: Copyright (c) IARIA, 2013

Publication date: November 17, 2013

Published in: conference

ISSN: 2326-9332

ISBN: 978-1-61208-311-7

Location: Lisbon, Portugal

Dates: from November 17, 2013 to November 21, 2013