Chinese natural language interrogative sentence semantization knowledge base automatic question-answering method
Hu Wei,Jiang Chengyue,Cheng Gong,Qu Yuzhong
2016-01-01
Abstract:The invention discloses a Chinese natural language interrogative sentence semantization knowledge base automatic question-answering method. The method includes the following steps that Chinese natural language processing is performed on a fact type question input by a user, word segmentation, part-of-speech tagging and identification and expanding of a named entity are achieved, and a semantic dependency tree is generated; a generalization template and a semantic analysis technology are used for acquiring time, space, a fact entity, a fact object and the like in an interrogative sentence, then semantic processing is performed, composition element attributes relevant to all events in the interrogative sentence and values of the attributes are extracted, a plurality of 'attribute-value' pairs are generated, to-be-answered elements are substituted by interrogatives, and a complex fact triple set is formed; after a triple where a to-be-answered part is located is combined with other relevant fact triple sets to form knowledge base query with conditional constraints, and query matching based on similarity calculation is performed in a knowledge base, a result is extracted from the knowledge base, and a final answer is obtained. Fast and accurate query response to the knowledge base is achieved.