Answering Non-aggregation Questions Over Linked Open Data

With the bloom in the development of linked open data,a lot of work for various purposes have been developed by taking linked open data as the underlying knowledge base.By taking advantage of abundant knowledge stored in the linked open data,a question answering system fills the gap between user information needs expressed in natural language and a complicated query process.As a result,it is of great importance and can be further to be used for other topics,such as optimization of search result of the search engine,automatic online customer services and so on.Focusing on solving the non-aggregation questions,in this paper,we construct a subgraph of the linked open data from the entities detected in the query and propose a graph traversal method to solve both the semantic relation extraction problem and semantic item disambiguation problem in a joint way.Compared with existing work,we simplify the process of query intention understanding and pay more attention to the answer path ranking.We firstly evaluate our method on a non-aggregation question dataset and further on a complete dataset.Compared with several state-of-the-art systems,experimental results show that our method achieves best performance.