WindMine: Fast and Effective Mining of Web-Click Sequences

Download Now Date Added: Mar 2011
Format: PDF

Given a large stream of users clicking on web sites, how can the people find trends, patterns and anomalies' The authors have developed a novel method, WindMine, and its fine-tuning sibling, WindMine-part, to find patterns and anomalies in such datasets. Their approach has the following advantages: it is effective in discovering meaningful "Building blocks" and patterns such as the lunch-break trend and anomalies, it automatically determines suitable window sizes, and it is fast, with its wall clock time linear on the duration of sequences. Moreover, it can be made sub-quadratic on the number of sequences (WindMine-part), with little loss of accuracy.