earticle

논문검색

Query Suggestion Based on Theme and Context

초록

영어

Query suggestion has become one of the most fundamental features of search engines. It
attempts to suggest a series of similar queries for improving the search effectiveness. The
paper proposed a new query suggestion method which is based on themes and context.
Different from previous work, it measure similar queries from the level of semantic level and
a new similar queries metric is presented. Furthermore, how to choose similar queries for
suggestion is discussed and a new method is proposed. In the new query suggestion method,
not only theme, but also query context has been taken into considerate. Experiments show
that the new query suggestion method significantly outperformed than related works.

목차

Abstract
 1. Introduction
 2. Related Work
 3. New Query Suggestion Method Based on Theme and Context
  3.1. Similar Queries Metric
  3.2. Query Suggestion
 4. Evaluation
  4.1. Data Set
  4.2. Clustered Results Analysis
  4.3. Query Suggestion Evaluation
 5. Conclusion and Future Work
 References

저자정보

  • Lingling Meng Department of Educational Information Technology, East China Normal University, Shanghai, 200062, China
  • Runqing Huang Shanghai Municipal People's Government, Shanghai, 200003, China
  • Junzhong Gu Computer Science and Technology Department, East China Normal University, Shanghai, 200062,

참고문헌

자료제공 : 네이버학술정보

    함께 이용한 논문

      ※ 원문제공기관과의 협약기간이 종료되어 열람이 제한될 수 있습니다.

      0개의 논문이 장바구니에 담겼습니다.