Effective Data Retrieval Using Relevance Feedback and Local Search Approximation Algorithms

Abstract: In search engines, data/Information Retrieval (IR) is affected with indexing and retrieving documents with user�s requirement. The documents which are retrieved should most relevant to the users need. In order to improve the retrieval efficiency, RF (Relevance Feedback) methods are used. The query expansion with effective data retrieval is the main aim of our proposal. In this paper, we propose a local search approximation algorithm to re-weight the query terms and to re-rank the document retrieved by an IR and it solves the computational hard optimization problems. Additionally, we propose a new indexing method named as Set Inverted Index, which is a semantic extracted term based inverted index. This helps to summarize the documents and sum up with its semantic similarity. This indexing system outperforms than other inverted index methods.

Keywords: Information Retrieval systems, search engine, relevance feedback, local search approximation, inverted index.