Saturday, November 22, 2014

Preprint: Mining Various Semantic Relationships from Unstructured User-Generated Web Data



Pei-Ling Hsu, Hsiao-Shan Hsieh, Jheng-He Liang and Yi-Shin Chen, Mining Various Semantic Relationships from Unstructured User-Generated Web Data, Web Semantics: Science, Services and Agents on the World Wide Web, to appear.

Absract: With the emergence of Web 2.0, the amount of user-generated web data has sharply increased. Thus, many studies have proposed techniques to extract wisdom from these user-generated datasets. Some of these works have focused on extracting semantic relationships through the use of search logs or social annotations, but only hierarchical relationships have been considered. The goal of this paper is to detect various semantic relationships (hierarchical and non-hierarchical) between concepts using search logs and social annotations. The experimental results demonstrate that our proposed approach constructs adequate relationships.