CS700:Graduate Seminar in Computer Science & Informatics

Discovering Common Motifs in Cursor Movement Data
Dmitry Lagun, Department of Mathematics and Computer Science

Mouse cursor interaction behavior can provide valuable information on how searchers examine and engage with the web search results. This interaction data is far richer than traditional search click data, and can be used to improve search ranking, evaluation, and presentation. Unfortunately, the diversity and complexity inherent in this interaction data pose significant challenges to capturing the salient behavior characteristics through feature engineering. To address this problem, we introduce a novel approach of automatically discovering frequent subsequences, or motifs, from mouse cursor movement data, which could then be used as features for ranking, evaluation, and other search improvements. In order to scale our approach to realistic datasets, we introduce novel optimizations for motif discovery, specifically designed for mining cursor movement data. As a practical application, we show that by encoding the motifs discovered from thousands of real web search sessions as features for relevance estimation and result re-ranking, we achieve improvements of up to 27%, over a previous state-of-the-art baseline that relies on manually engineered features alone. These results, complemented with visualization and qualitative analysis, indicate that our motif discovery algorithm is able to capture key characteristics of mouse cursor movement behavior, providing a significant step forward in online behavior analysis. In addition to the application of motifs to Web search, we demonstrate that similar technique can be successfully applied in medical domain for the task of predicting future decline of memory function and subsequent development of the Alzheimer Disease.