Bricking Semantic Wikipedia by Relation Population and Predicate Suggestion
Semantic Wikipedia aims to enhance Wikipedia by adding explicit semantics to the links between Wikipedia entities.However,we have observed that it currently suffers the following limitations: lack of semantic annotations and semantic annotators.In this paper,we resort to relation population to automatically extract relations between any entity pair to enrich semantic data, and predicate suggestion to recommend proper relation labels to facilitate semantic annotating.Both tasks leverage relation classification which tries to classify the extracted relations into the predefined ones.However,since there are too few labeled data but there exists excessive noise in SemanticWikipedia,existing approaches cannot be directly applied to these tasks in order for high-quality annotations.In this paper, to tackle the above problems brought by Semantic Wikipedia,we use a label propagation algorithm and exploit semantic features (e.g.Domain and range constraints on categories) as well as linguistic features (e.g.Dependency trees of the context sentences) in Wikipedia articles.The experiment results on 7 typical relation types show the effectiveness and efficiency of our approach in dealing with both tasks.
Semantic Wikipedia Relation Population Predicate Suggestion Relation Classification
Haofen Wang Linyun Fu Yong Yu
Dept.of Computer Science & Engineering,Shanghai Jiao Tong University,200240 Shanghai,China
国际会议
南京
英文
1-11
2009-08-29（万方平台首次上网日期，不代表论文的发表时间）