Web Search Result Clustering based on Cuckoo Search and Consensus Clustering
This work addresses issues like synonymy and polysemy in web search for users, but it is incremental as it builds on existing clustering techniques.
The paper tackles the problem of clustering web search results to improve retrieval performance by introducing WSRDC-CSCC, a method combining cuckoo search meta-heuristic and consensus clustering, which achieves higher precision, recall, and F-measure compared to an existing method.
Clustering of web search result document has emerged as a promising tool for improving retrieval performance of an Information Retrieval (IR) system. Search results often plagued by problems like synonymy, polysemy, high volume etc. Clustering other than resolving these problems also provides the user the easiness to locate his/her desired information. In this paper, a method, called WSRDC-CSCC, is introduced to cluster web search result using cuckoo search meta-heuristic method and Consensus clustering. Cuckoo search provides a solid foundation for consensus clustering. As a local clustering function, k-means technique is used. The final number of cluster is not depended on this k. Consensus clustering finds the natural grouping of the objects. The proposed algorithm is compared to another clustering method which is based on cuckoo search and Bayesian Information Criterion. The experimental results show that proposed algorithm finds the actual number of clusters with great value of precision, recall and F-measure as compared to the other method