A new relevance feedback algorithm based on vector space basis change

Abstract

The idea of Relevance Feedback is to take the results that are initially returned from a given query and to use information about whether or not those results are relevant to perform a new query. The most commonly used Relevance Feedback methods aim to rewrite the user query. In the Vector Space Model, Relevance Feedback is usually undertaken by re-weighting the query terms without any modification in the vector space basis. With respect to the initial vector space basis(index terms), relevant and irrelevant documents share some terms (at least the terms of the query which selected these documents). In this paper we propose a new Relevance Feedback method based on vector space basis change without any modification on the query term weights. The aim of our method is to build a basis which optimally separates relevant and irrelevant documents. That is, this vector space basis gives a better representation of the documents such that the relevant documents are gathered and the irrelevant documents are kept away from the relevant ones. © 2014 Springer-Verlag Berlin Heidelberg.