Information Retrieval System with Pseudo Relevance Feedback using Rocchio algorithm

Information retrieval is concerned withorganizing and retrieving relevant documentsfrom a set of unstructured collections ofdocuments.In most collections, the same conceptmay be referred to using different word.This issueis known as synonymy which has an impact on therecall of most information retrieval systems.Theproblem can be eliminated by means of relevancefeedback where user judges the result of the initialquery, and system refines the query vector toimprove the recall.But in normal relevancefeedback, users have to give manual feedbackwhich is inflexible, so the proposed system usespseudo relevance feedback where top-k retrieveddocuments are assumed to be relevance and queryis rewrite using Rocchio algorithm.Finally thesystem analyzes with precision and recall ofinformation retrieval result