Development research of highly accurate Question Answering system doing precise matching of meaning

语义精准匹配的高精度问答系统开发研究

• 批准号: 16500085
• 负责人: HARADA Minoru
• 金额: $ 2.3万
• 依托单位: Aoyama Gakuin University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 2004
• 资助国家: 日本
• 起止时间: 2004 至 2006
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-16500085/
• 关键词: Question answering; Graph matching; Semantic analysis; Semantic graph; Answer extraction; 内容検索; SVM; 回答抽出; 自然語処理; グラフ推論; 照応解析; インターネット検索

We have developed Metis, a question-answering system that finds an answer by matching a question graph with the knowledge graphs. The question graph is obtained as a result of semantic analysis of a question sentence, the knowledge graphs are similarly analyzed from knowledge sentences retrieved from a database using keywords extracted from the question sentence. In retrieving such knowledge sentences, the system searches for and collects them using Lucene, a search engine, based on search keywords extracted from the question graph. To extract the answer, Metis calculates the degrees of similarity between the question and knowledge graphs to conduct precise matching. In this matching, the system calculates the degrees of similarity, which is the relative size of the similarity co-occurrence graph to the question graphs with respect to all combinations of nodes in the knowledge graph corresponding to those in the question graph. The system then chooses the knowledge graph with the highest degree of similarity and extracts from it the portion that corresponds to the given interrogative word. The system presents this portion as the answer. The evaluation experiment was done by using 100 quiz millionaire questions. The precision to obtain the correct answer within the first three answers was 74%. Moreover, it participated in the evaluation contest in NTCIR, and the precision was 18% in QAC which asked factoid question, the precision was 22% in CLQA which asked the reasons, the method and etc.

我们开发了 Metis，一个问答系统，通过将问题图与知识图匹配来找到答案。问题图是对问题句进行语义分析的结果而获得的，知识图谱同样是使用从问题句中提取的关键词从数据库中检索的知识句进行分析的。在检索此类知识句子时，系统根据从问题图中提取的搜索关键字，使用搜索引擎Lucene进行搜索和收集。为了提取答案，Metis 计算问题与知识图谱的相似度，进行精准匹配。在该匹配中，系统计算相似度，即对于知识图谱中与问题图中的节点相对应的所有节点组合，相似度共现图与问题图的相对大小。然后系统选择相似度最高的知识图谱，并从中提取与给定疑问词对应的部分。系统将这部分作为答案呈现。评估实验是使用100个百万富翁问答题进行的。前三个答案中获得正确答案的准确率为 74%。另外，参加了NTCIR的评估竞赛，在询问factoid问题的QAC中精度为18%，在询问原因、方法等的CLQA中精度为22%。

项目成果

QA System Metis Based on Semantic Graph Matching

基于语义图匹配的问答系统Metis

• 发表时间: 2007
• 期刊: Proc.of the 6th International Conference on NII Test Collection for IR Systems(NTCIR6) (In printing)
• 作者: Minoru Harada;Yuhei Kato;Kazuaki Takehara;Masatsuna Kawamata;Kazunori Sugimura;Junichi Kawaguchi
• 通讯作者: Junichi Kawaguchi

Summary generation centered on important words

以重要单词为中心的摘要生成

• 发表时间: 2005
• 期刊: LNCS 3411 (information Retrieval Technology)
• 作者: Dongli Han;Takashi Noguchi;Tomokazu Yago;Minoru Harada
• 通讯作者: Minoru Harada

意味解析を踏まえた自動要約システムABISYS

ABISYS，基于语义分析的自动摘要系统

• 发表时间: 2006
• 期刊: 言語処理学会論文誌 Vol.13,No.1
• 作者: 田中信彰;面来道彦;野口貴;矢後友和;韓東力;原田実
• 通讯作者: 原田実

Determination of voice, tense, aspect and moodin the Japanese semantic analysis system SAGE

日语语义分析系统SAGE中的语音、时态、体态、语气的判定

• 发表时间: 2005
• 期刊: Proceedings of the 67th National Conference of Information Processing Society of Japan. No.2
• 作者: Satoh Naomi;Dongli Han;Minoru Harada
• 通讯作者: Minoru Harada

Research of Anaphoric analysis based on Semantic analysis

基于语义分析的照应分析研究

• 发表时间: 2007
• 期刊: Proceedings of the 69th National Conference of Information Processing Society of Japan. No.2
• 作者: Kazunori Sugimura;Yoshitatsu Matsuda;Minoru Harada
• 通讯作者: Minoru Harada