Reformulated Query based Document Retrieval using optimized Kernel Fuzzy Clustering algorithm

Abstract

Clustering-based document retrieval system offers to find similar documents for a given user's query. This study explores the scope of kernel fuzzy c-means (KFCM) with the genetic algorithm on document retrieval issue. Initially, genetic algorithm-based kernel fuzzy c-means algorithm (GKFCM) is proposed to make the clustering of documents in the library. For each cluster, an index is created, which contains a common significant keywords of the documents for that cluster. Once the user enters the keyword as the input to the system, it will process the keywords with the WORDNET ontology to achieve the neighbourhood keywords and related synset keywords. Lastly, the documents inside the cluster are released at first as the resultant-related documents for the query keyword, which clusters have the maximum matching score values. Experiments results prove that GKFCM-based proposed system outperforms better performance than existing methods.