Introduction to Interactive Document Search with Multi-Class Diverse Density

  • Authors:
  • Chalita Hiransoog;Lucien Murray-Pitts

  • Affiliations:
  • The University of Edinburgh, Scotland, U.K.;Celoxica Japan KK, Japan

  • Venue:
  • Proceedings of the 2006 conference on Advances in Intelligent IT: Active Media Technology 2006
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

An interactive document search is introduced. It is a document search that allows a user to report to the search engine their opinion on how relevant each document is. The search engine then uses this information to narrow down the list of documents given to a user as a result of the search. It is proposed that there are three types of keywords: general keywords; hidden keywords; and negative keywords, that the search engine needs to uncover. The problem is then redefined as Multi-Class Multiple-Instance learning problem where the Multi-Class Diverse Density method developed as part of this research was used to find the solution.