Clustering Algorithm for Enhanced Bibliography Visualization

A Bibliography is a list of books, publications, journals etc., with details such as authors and references. Visualization could be used as a data analysis tool to represent various types of data, analyze huge chunks of data easily and arrive at interesting results. The idea of this project is to provide a medium which eases the combination of bibliography with visualization. Though there are many sources of bibliographic data like the Digital Bibliography and Library Project (DBLP), Citeseer, Google Scholar, none of these data could be used directly for deducing relations between various entities or for visualizing the relationship between related entities. This project aims at providing a web-service that takes user queries as input and retrieves the corresponding data from a local database. Then the web-service applies a clustering algorithm to the retrieved data and then presents the clustered data as XML to the requestor. The user of the system could be any automated program that aims at providing a visual interface to bibliography. One of the main outcomes of this approach would be bringing out the hidden relationship between various related bibliographic entities and making the relationships more obvious and readable than the existing systems