Formalized Answer Extraction Based on Pattern Learning and Confidence Estimation

LI Peng; QIAO Pei-li; WANG Xiao-long; WANG Bao-xun

您当前的位置：

首页 >

文章列表页 >

Formalized Answer Extraction Based on Pattern Learning and Confidence Estimation

更新时间：2025-07-16

- Formalized Answer Extraction Based on Pattern Learning and Confidence Estimation
- Acta Electronica Sinica Vol. 36, Issue 12, Pages: 2339-2343(2008)
- 作者机构：
  
  1. 哈尔滨理工大学计算机科学与技术学院,黑龙江,哈尔滨,150001
  2. 哈尔滨工业大学计算机科学与技术学院,黑龙江,哈尔滨,150001
  3. 哈尔滨理工大学计算机科学与技术学院黑龙江哈尔滨,150001
  4. 哈尔滨工业大学计算机科学与技术学院黑龙江哈尔滨,150001
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP391.2
- Published Online：25 December 2008，
  
  Published：2008
- 稿件说明：
移动端阅览
LI Peng, QIAO Pei-li, WANG Xiao-long, et al. Formalized Answer Extraction Based on Pattern Learning and Confidence Estimation[J]. Acta Electronica Sinica, 2008, 36(12): 2339-2343.
DOI：

LI Peng, QIAO Pei-li, WANG Xiao-long, et al. Formalized Answer Extraction Based on Pattern Learning and Confidence Estimation[J]. Acta Electronica Sinica, 2008, 36(12): 2339-2343. DOI：

摘要

本文提出一种基于模式学习的形式化答案抽取方法

区别于传统基于特征的答案抽取方法

通过问题模式和答案模式的自动匹配

直接获取问题答案.本文通过机器学习的方法自动生成用于答案抽取的形式化模板

克服了人工方法费时、费力以及覆盖率低等问题.本文创造性地采用逻辑回归的方法对所学习到的模式进行置信度评价.对比实验表明

本文的方法取得了比较好的答案抽取效果.本文方法实际应用于国际TREC QA评测

评测结果证明本文的方法与传统基于特征的答案抽取方法具有很好的互补性.

Abstract

This paper presents a method of formalized answer extraction based on pattern learning.Our method extracts answers directly by matching the question patterns and the answer patterns automatically

which is different from the other traditional feature-based ones.The method gains formalized patterns automatically by using a machine learning strategy.Therefore

huge work load and low coverage of manually methods are avoided.Creatively

we apply logistic regression to estimate the confidence of the learned patterns.The contrast experiments give our method encouraging results

and it achieves fair and objective test in TREC QA track

the result of which shows that the method is a supplementary to the traditional feature based answer-extracting approaches.

关键词

Keywords

references

Views

1099

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Advances in Open-Domain Question Answering

Chinese Question Answering Based on Syntax Analysis and Answer Classification

State-Difference Counting and Its Confidence Estimate

Predicting Microblog User Retweet Behaviors Based on Energy Optimization

Related Author

ZHANG Zhi-chang

ZHANG Yu

LIU Ting

LI Sheng

SUN Ang

JIANG Ming-hu

HE Yi-fan

CHEN Lin

Related Institution

IR Laboratory of Computer Science and Technology School,Harbin Institute of Technology.

Mathematic and Information Science School,Northwest Normal University

IR Laboratory of Computer Science and Technology SchoolHarbin Institute of Technology.HarbinHeilongjiang 150001China

Mathematic and Information Science SchoolNorthwest Normal UniversityLanzhouGansu 730070China

Lab of Computational Linguistics,School of Humanities and Social Sciences,Tsinghua University

⁰