Wednesday, April 6, 2011

Paper Reading #20 - Supporting Exploratory Information Seeking by Epistemology-based Social Search

Comments

Patrick

Paoloa


Reference

Supporting Exploratory Information Seeking by Epistemology-based Social Search

Yuqing Mao, Haifeng Shen, Chengzheng Sun

IUI 2010 - Hong Kong


Summary

This paper looks at new techniques for improving exploratory search strategies. There is a large focus on reusing searches to aide other users in their quests for information. The system proposed, Baijia, uses an epistemology-based search in addition to exploratory information seeking techniques to add a new dimension of relevance assurance to search. When a search is run a user can select pages to add to the epistemology. Other users rank these epistemologies. As use of a particular epistemology increases it becomes more reliable and helpful to new searchers. In effect, users benefit from previous searches. The system was compared against data which AOL released regarding their search engine. It was demonstrated that this system creates a better performance than the AOL system.


Analysis

This is one of many approaches to creating more relevant search results. Over the past few years we have learned both the beneficial and negative power of crowd sourcing information. While sites like Wikipedia for the most part enjoy great benefits from polling users for data, they are also occasionally attacked and caused to show incorrect data. I could imagine a system like this being attacked in order to augment search results for financial or political purposes. I think future work should focus on noticing trends which seem oddly motivated in systems like this.

No comments:

Post a Comment