ExtMiner: Combining multiple ranking and clustering algorithms for structured document retrieval

Abstract

This paper introduces ExtMiner, a platform and potential tool for information management in SMEs (Small& Medium-size Enterprise), or for organizational workgroups. ExtMiner supports interactive and iterative clustering of documents. It provides users with a visual clusterand list views at the same time, supporting iterative search process. ExtMiner may also be applied as a platform for research on retrieval fusion, since it combines search, clustering and visualization algorithms. ExtMiner was evaluated with three document collections. Although the findings were encouraging the user interface and performance with large document repositories need further development.