User modeling with sequential data
暂无分享,去创建一个
User models are useful to automatically adapt software or information to specific users. Because crafting user models by hand is labor intensive, machine learning techniques are used to construct user models. In this paper we illustrate how via instance based techniques user modeling tasks can be performed, based on sequences of observed user actions. We compare this approach with a common classification approach on real world data and briefly sketch how this technique is also useful when dealing with non-labeled data.
[1] Ian H. Witten,et al. Data mining: practical machine learning tools and techniques with Java implementations , 2002, SGMD.
[2] Saul Greenberg,et al. USING UNIX: COLLECTED TRACES OF 168 USERS , 1988 .
[3] Vladimir I. Levenshtein,et al. Binary codes capable of correcting deletions, insertions, and reversals , 1965 .