Annual Meeting Contributed Papers 2009 START Conference Manager    

An Empirical Evaluation on Textual Results Clustering for Web Search

Hsiao-Tieh Pu, Shi-Yin Chen and Pei-Yi Kuo

(Submission #69)


Abstract

Clustering web search results into dynamic clusters and hierarchies provides a promising way to alleviate information overabundance typically found in the ranked list search engines. This study aims to investigate the usefulness of textual results clustering in web search by analyzing usersí search performance and satisfaction level with and without the aid of clusters and hierarchies. Usability and comprehension tests have been conducted using multiple data collection methods, and various objective and subjective measures were applied. Based on the results, though participants in the study searched slightly faster, obtained more relevant pages, and were more satisfied using the non-clustering ranked list interface, no significant difference was found. It is noted that clustering interface offers the opportunity for diverse searching, as each participant obtained relevant results not found from using the ranked list interface, and the repetitive ratio of each otherís relevant results was low. The study shows the clustering results interface provides the values of highlighting prominent concepts and offering richer context for exploring, learning and discovering related concepts; yet it also induces certain degree of information uncertainty, lost, and anxiety. Discussions on the contrast view of clustering search and suggestions for future studies are also provided.


  
START Conference Manager (V2.54.6)