Home // International Journal On Advances in Software, volume 7, numbers 3 and 4, 2014 // View article

A Social Approach for Natural Language Query to the Web of Data

Authors:
Takahiro Kawamura
Akihiko Ohsuga

Keywords: Web of Data; Linked Open Data; Query Answering; User Context; Plant.

Abstract:
The `Web of Data' aims to enable people to share structured data as easily as they can share documents on the Web. Accordingly, a search engine for the data is becoming important for promoting data-intensive services. However, the data are represented and interlinked by a triple format in the Resource Description Framework, and thus full-text search is unsuitable for structured data, and formal query languages are difficult for ordinary users. Therefore, we propose a query answering system, which accepts natural language queries in Japanese, translates them to triples and sends formal queries to the `Web of Data'. The proposed system is implemented on a mobile phone, and evaluated in the context of gardening advice for people. The use is within a social system, and combines user feedback and user context information obtained by sensors, in order to improve search accuracy for open-schema mapping and data acquisition. Then, we confirmed that the search accuracy of 18.2% was improved by the user feedback, and the useful triples have been increased 3.4 times by the context information.

Pages: 446 to 455

Copyright: Copyright (c) to authors, 2014. Used with permission.

Publication date: December 30, 2014

Published in: journal

ISSN: 1942-2628