Federated Search: Results Clustering

Abstract

Federated searching applications are supposed to help users find and access relevant content for a given search. Librarians and library users are now looking at federated search tools and expect easier to use results sets. However, with the "avalanche" of results, the patron can be easily overwhelmed by the sheer size of the results sets. Present day federated search tools do very little to aid in "exploring" the results set – rather they just present a very long list of citations.

New results clustering technology will improve the user experience to the satisfaction of both users and librarians. Attendees to this session will discover how results clustering can make it easy to "drill down" into the results and reduce the amount of time "searching", providing users more time for analysis and interpretation.

Topics presented in this session are directly relevant to librarians that are considering purchasing a federated search engine or those that have already purchased a system and looking to see what is on the horizon.

Content Outline 1) An overview of federated searching What does a patron experience prior to discovering federated searching applications? What is federated searching? How do patrons use federated searching? 2) Challenges of relevance in federated searching What are librarians and users saying about the results sets? 3) Explanation of clustering What does it do? What are the advantages? What are the benefits? How can it be implemented? 4) Demonstration and walkthrough