Aditi Agrawal, Dr. A. J. Agrawal
Abstract: This paper introduces a system for cross language information retrieval where selective term technique and query expansion will be combined to improve the retrieval of more relevant documents. The document and query processing is done using the techniques of tf-idf and cosine similarity for retrieval. Threshold value is decided to retrieve the documents relevant to query. Later a pseudo Relevance technique will be applied to further expand the query and improve relevance in future.
Keywords: Cross Language Information Retrieval, Hindi-English Dictionary, cosine similarity, tf-idf