Discovering temporal hidden contexts in web sessions for user trail prediction

摘要: In many web information systems such as e-shops and portals, predictive modeling is used to understand user's intentions based on their browsing behaviour. User behavior inherently sensitive various hidden contexts. It has been shown in different experimental studies that exploitation of contextual can help improving prediction performance significantly. reasonable assume users may change intents during one session changes are influenced by some external factors switch temporal context e.g. 'users want find about a specific product' after while 'they buy this product'. A be represented sequence actions where ordered time. The generation might several Each concatenation independent segments, each which corresponding context. We show how learn apply models for segment work. define the problem discovering contexts way we optimize directly accuracy (e.g. users' trails prediction) process acquisition. Our empirical study real dataset demonstrates effectiveness our method.