¡@

Web searching has become a necessary activity for most Internet users. People can reach a specific website or collect desired information by using search engines. Nevertheless, when a user is unfamiliar with his or her search target, e.g., learning a previously unknown topic, browsing and understanding the web pages in the corresponding search result requires extra effort. In such an informational web search, the user accumulates information cues by reading web pages to approach the search target. For example, a user can gradually understand ¡§cloud computing¡¨ after obtaining some information about relevant terms, e.g., distributed computing, Google App Engine, and Amazon EC2. In this work, we propose an ¡§immediacy of interest¡¨ indicator, measured in terms of information freshness, to evaluate the possible user interest in a web page contained in a search result. Moreover, to address relevance in search sessions, the topic map is also utilized in our proposed scheme. Empirical studies show that informational web searching tasks can be completed efficiently and effectively with our approach.

Keywords: web search, browsing history, information cue, relevance feedback, visualization

Retrieve PDF document (201201_11.pdf)

Received February 28, 2011; revised August 27, 2011; accepted August 30, 2011.
Communicated by Chun-Nan Hsu.
^* This work was supported in part by the National Science Council of Taiwan, Project No. NSC 100-2221-E-006-261-.
⁺ Corresponding author.