Research On The Ontology-based Internet Chinese Question And Answer System And Its Implementation

Along with the expansion of information on the Internet, it is becoming more and more difficult to find the information needed on the Internet. Although the current search engines such as Google and Baidu has provided tools to effectively search information from enormous web pages, the modern search engines are generally based on keywords matching therefore there are lots of redundant or useless information in the results and nevertheless influence the accuracy of the results. In order to solve the problem, this paper proposed a question and answer system, which can not only utilizing the enormous information more effectively on the Internet but also making the results more abundant and more accurate by utilizing ontology knowledge.According to the domestic researches, the quality of question and answer system is not very satisfying. There are two reasons. First, the categorizing of questions is not accurate so that the answer may deviate from the original question. Second, the current answer extraction approaches are usually based on statistic method, which ignores the semantic of sentences and influences the accuracy of the result. To solve the problems, this paper focus on the research of categorizing of questions and the extraction answer candidates. First of all, this paper proposed a Chinese question categorizing method base on domain Ontology in order to categorizing the questions more accurate. Second, this paper also put forward a multi-strategy answer extraction method base on domain Ontology to improve the quality of the generated answers.All in all this paper mainly concentrates on the following three aspects:1) Put forward a Chinese question categorizing method based on domain Ontology. By making use of the hierarchy of ontology as well as the thesaurus, the various and complicated character of Chinese has been completely satisfied.2) A multi-strategy answer extraction algorithm has been proposed in this paper. By utilizing the variety and complexity of thesaurus, the calculation method of concept similarities in the ontology has been optimized and been combined with pattern matching. All these efforts have improved the access rate and the accuracy of answer extraction.3) In order to prove the theory proposed in the paper, the author has implemented a prototype system and done lots of comparison experiments. All the experiments have demonstrated the algorithms are effective.