| 摘要: |
| 置信度指的是一个问题回答系统(QA系统)对其所作回答的自信程度.描述了一种基于最大熵模型的算法.首先,从训练语料中提取若干因素来训练最大熵模型;然后应用训练好的模型在测试集上计算置信度.在2002年度的文本检索会议(TREC)中,QA系统用该算法计算每个问题答案的置信度,并依此排序,获得了显著的成绩. |
| 关键词: 自然语言处理 信息检索 问答系统 最大熵模型 置信度 |
| DOI： |
| 分类号: |
| 基金项目:Supported bythe National Natural Science Foundation of China under Grant No.60435020(国家自然科学基金);Key Project of Shanghai Science andTechnologyCommitteeofChinaunderGrantNo.035115028(上海市科委重点项目) |
|
| A Maximum Entropy Model Based Confidence Scoring Algorithm for QA |
|
YOU Lan,ZHOU Ya-Qian,HUANG Xuan-Jing,WU Li-De
|
| Abstract: |
| Confidence score describes how confident a question-answering system is about its response. This paper presents a Maximum Entropy Model based algorithm which uses several factors to train an ME model, and then the ME model is used to calculate the confidence of other questions. Efficiency of this method has been proved by the TREC11’s QA evaluation, where the performance of the system has been improved dramatically after confidence ranking. |
| Key words: natural language processing information retrieval question-answering system maximum entropy model confidence score |
