A novel approach for classification and clustering of biomedical citations

Citation refers the information of a published paper with its author and publication details. It is used by various authors for referring the research works published in other research articles. Citations play a crucial role in several scientific publications digital libraries (DLs), like Cite Seer, arXiv e-Print, DBLP, and Google Scholar. Users usually use citations to seek out data of interest in DLs, while researchers relay on citations to see the impact of a specific article. Citation mining is the area where in the citation databases are mined for performing various mining tasks such as classification and clustering to retrieve citations efficiently and accurately. Citations have additionally been used as auxiliary support in information retrieval tasks. Citation classification is the process of classifying the citation data by means of topic, author, paper name, and journal category. Clustering involves the categorization of papers based on content similarity or functional similarity. At present the size of databases in the web is massive hence the quantity of records in a dataset will vary from some thousands to thousands of millions. Authors or scholars are spending their precious time in searching the papers especially in bio medical field. So to provide more accurate retrieval of biomedical citations we have proposed a citation mining system with a combined approach of clustering. Our experiments conducted with the citations from the web database shows an effective retrieval of biomedical citations.