Earticle

초록

영어

Foraging information from web forums is still one of the most challenging information retrieval tasks due to various combinations of auto-generated page structural information and user-created contents. Traditional information extraction methods employ either duplicated subtree pattern detection methods, or machine learning methods. Due to the periodical update of forum templates and diversity of page contents, aforementioned approaches do not work very well on forum sites. In this paper, we present a page-segmentation based wrapper specially designed for mining data pattern of web forums, which combines a novel page segmentation algorithm and decision tree classifier together to detect the data pattern in forum. In the segmentation phase, a novel page segmentation algorithm is proposed to identify the records areas in a page, then a classifier is adopted to identify the detailed pattern of each record in the extraction phase. Extensive experiments on various types of web forums are conducted and the results conclude that our wrapper is a more generalized one which requires only few labeled training data.

목차

Abstract
 1. Introduction
 2. Related Works
 3. Extraction for Web Forum Data
  3.1. Concept Definition
  3.2. System Architecture
  3.3. Segmentation for Posts Extraction
  3.4. Classification Based Records Extraction
 4. Experiments and Evaluation Results
  4.1. Experiment Design
  4.2. Page Segmentation Comparison Results
  4.3. Data Extraction Comparison Results
  4.4. Time Consume Comparison Results
 5. Conclusions
 Acknowledgements
 References

키워드

저자

• Chunshan Li [ Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen Key Laboratory of Internet Information Collaboration ]
• Jingjing Chen [ Department of Computer Science, Hong Kong Baptist University, Hong Kong SAR, China, School of Science and Technology, China Jiliang University, Hangzhou, China ]
• Dianhui Chu [ Department of Computer Science, Harbin Institute of Technology ]
• Ge Song [ Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen Key Laboratory of Internet Information Collaboration ]
• Haijun Zhang [ Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen Key Laboratory of Internet Information Collaboration ]
• Yunming Ye [ Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen Key Laboratory of Internet Information Collaboration ]
• Jianliang Xu [ Department of Computer Science, Hong Kong Baptist University, Hong Kong SAR, China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Database Theory and Application

• 간기

  격월간

• pISSN

  2005-4270

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605