文献类型：会议论文

英文题名：A SVM and Co-seMLP Integrated Method for Document-based Question Answering

作者：Liu Xiaoan[1];Peng Tao[2]

第一作者：Liu Xiaoan

通讯作者：Liu, XA[1]

机构：[1]Beijing Union Univ, Coll Intellectualized City, Beijing, Peoples R China;[2]Beijing Union Univ, Coll Robot, Beijing, Peoples R China

第一机构：北京联合大学智慧城市学院

通讯机构：[1]corresponding author), Beijing Union Univ, Coll Intellectualized City, Beijing, Peoples R China.|[1141734]北京联合大学智慧城市学院;[11417]北京联合大学;

会议论文集：14th International Conference on Computational Intelligence and Security (CIS)

会议日期：NOV 16-19, 2018

会议地点：Hangzhou, PEOPLES R CHINA

语种：英文

外文关键词：component; Feature; Word Vector; styling; Model Integration

摘要：In this paper, we describe our features and models for Chinese Open-Domain Question Answering DBQA shared task in NLPCC-ICCPOL 2017. After the analysis of task and dataset, 8 features were extracted, and then 4 models were trained. Finally, our model achieves a result, in which MRR score is 0.494292 and MAP score is 0.491736.