Mining user behavior for resource prediction in interactive electronic malls

Applications in virtual multimedia catalogs are highly interactive. Thus, it is difficult to estimate resource demands required for presentation of catalog contents. We propose a method to predict presentation resource demands in interactive multimedia catalogs. The prediction is based on the results of mining the virtual mall action log file. The log file typically contains information about previous user interests and browsing behavior. These data are used for modeling users future behavior within a session. We define heuristics to generate a start-up user behavior model as a continuous time Markov chain and adapt this model during a running session to the current user.