ICTNET at Context Suggestion Track TREC 2012

Abstract

Yelp is used to collect the initial candidate web site list for every context. Then we use a spider to fetch contents of each candidate site. We also classify the candidates and examples into 6 classes and calculate preference of each profile to each class. The classes and preference are used for the ranking of the results. Finally we use several methods to filter and generate descriptions for every web site that is returned in the results.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 2012
Accession Number
ADA580906

Entities

People

  • Bingyang Liu
  • Qian Liu
  • Tong Wu
  • Xianghui Lin
  • Xueqi Cheng
  • Yanqin Zhong
  • Yue Liu

Organizations

  • Chinese Academy of Sciences

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Classification
  • Commerce
  • Demographic Cohorts
  • Governments
  • Information Operations
  • Maryland
  • Standards
  • Websites

Fields of Study

  • Computer science

Readers

  • Coastal and Marine Engineering/Sediment Transport/Hydraulic Engineering
  • Information Retrieval
  • Neural Network Machine Learning.