Download Now Free registration required
This paper proposes a personalized URL re-ranking method based on psychological characteristics of users browsing. The characteristics are classified into three groups, which are "Common-mind," "Uncommon-mind," and "Extremely uncommon-mind." The authors' personalization method constructs an index of the anchor text retrieved from the web pages that the user has clicked during his/her past searches. Their method provides different weights to the anchor text according to the psychological characteristics for re-ranking URLs. In the experimental section, they show that their method can provide better performance than Google and another web personalization method in terms of the average rank.
- Format: PDF
- Size: 294.09 KB