Retrieval result presentation and evaluation

  • Authors:
  • Shengli Wu;Yaxin Bi;Xiaoqin Zeng

  • Affiliations:
  • School of Computing and Mathematics, University of Ulster, Northern Ireland, UK;School of Computing and Mathematics, University of Ulster, Northern Ireland, UK;School of Computer Science, Hohai University, Nanjing, China

  • Venue:
  • KSEM'10 Proceedings of the 4th international conference on Knowledge science, engineering and management
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

In information retrieval systems and digital libraries, result presentation is a very important aspect. In this paper, we demonstrate that only a ranked list of documents, thought commonly used by many retrieval systems and digital libraries, is not the best way of presenting retrieval results. We believe, in many situations, an estimated relevance probability score or an estimated relevance score should be provided for every retrieved document by the information retrieval system/digital library. With such information, the usability of the retrieval result can be improved, and the Euclidean distance can be used as a very good system-oriented measure for the effectiveness of retrieval results. The relationship between the Euclidean distance and some ranking-based measures are also investigated.