摘要
How to query Linked Data effectively is a challenge due to its heterogeneous datasets.There are three types of heterogeneities,i.e.,different structures representing entities,different predicates with the same meaning and different literal formats used in objects.Approaches based on ontology mapping or Information Retrieval(IR) cannot deal with all types of heterogeneities.Facing these limitations,we propose a hierarchical multi-hop language model(HMPM).It discriminates among three types of predicates,descriptive predicates,out-associated predicates and in-associated predicates,and generates multi-hop models for them respectively.All predicates' similarities between the query and entity are organized into a hierarchy,with predicate types on the first level and predicates of this type on the second level.All candidates are ranked in ascending order.We evaluated HMPM in three datasets,DBpedia,Linked MDB and Yago.The results of experiments show that the effectiveness and generality of HMPM outperform the existing approaches.
How to query Linked Data effectively is a challenge due to its heterogeneous datasets. There are three types of heterogeneities, i.e., different structures representing entities, different predicates with the same meaning and different literal formats used in objects. Approaches based on ontology mapping or Information Retrieval （IR） cannot deal with all types of heterogeneities. Facing these limitations, we propose a hierarchical multi-hop language model （HMPM）. It discriminates among three types of predicates, descriptive predicates, out-associated predicates and in-associated predicates, and generates multi-hop models for them respectively. All predicates＇ similarities between the query and entity are organized into a hierarchy, with predicate types on the first level and predicates of this type on the second level. All candidates are ranked in ascending order. We evaluated HMPM in three datasets, DBpedia, LinkedMDB and Yago. The results of experiments show that the effectiveness and generality of HMPM outperform the existing approaches.
作者
李俊娴
汪卫
王晶晶
LI Junxian;WANG Wei;WANG dingjing(School of Computer,Jiangsu University of Science and Technology,Zhenjiang 212003,Jiangsu,China;School of Computer Science,Fudan University,Shanghai 200433,China;Yangzhou Polytechnic College,Yangzhou 225009,Jiangsu,China)
