Combining Multiple Evidence from Diierent Relevance Feedback Methods

It has been known that using diierent representations of a query retrieves diierent sets of documents. Recent work suggests that signiicant improvement in retrieval performance can be achieved by combining multiple representations of an information need. In this paper, we rst investigate a fully automatic way of generating multiple query representations for a given information problem. We produce multiple query vectors by expanding an initial query vector with various relevance feedback methods. We then describe the eeect of combining the multiple query vectors on retrieval eeectiveness. Experimental results show that signiicant improvements can be obtained by the combination of multiple query vectors expanded with diierent relevance feedback methods.