Transactions of the Japanese Society for Artificial Intelligence
Online ISSN : 1346-8030
Print ISSN : 1346-0714
ISSN-L : 1346-0714
-- Development of Web API for Triplification of Text Information --
Takahiro KawamuraAkihiko Ohsuga
Author information
JOURNALS FREE ACCESS Advance online publication

Article ID: LOD-21


Linked Open Data (LOD) is recently attracting attention as a vast amount of distributed knowledge base on the Web.Thus, semi-structured data such as tables and hierarchical data in several domains have been triplified to the LOD.In the research area, however, triplification of unstructured data such as text and sensor data is actively studied as the next target.Therefore, we developed a Web API for mainly extracting triples from text data, which is useful for the triplification of text data.We defined two steps for the text triplication.The first step is the extraction of phrases, which correspond to triple <subject, verb, object>, location and time from a natural language sentence, and the second one is a conversion of the extracted phrases to the existing (or new) resources and properties in the LOD. In this paper, we first describe the service specification corresponding to the first step, technical background, and evaluation of the current extraction accuracy, then finally introduce some use cases of the service. Although this service adopts a novel combination of a restrictive method using ontology-based rules and an example-based machine learning method using conditional random field, based on probability distribution, the main cotribution of the service is in practical aspect, that is, mash-up of several natural language processing techniques as a text triplification service, and deployment as a Web API freely available for public use so that non-expert easily use it.

Information related to the author
© The Japanese Society for Artificial Intelligence 2016
Previous article Next article