A Data Mining Approach to Learning Probabilistic User Behavior Models from Database Access Log

Forecasting economic result of business logic improvements using Game Theory for modeling user scenarios

“…However, ( 30) is explicit enough to understand domination between two strategies in the practical study clearly. As shown by many pieces of research [2,[5][6][7][8], the modern technology stack proposes various tools, enabling science and industry professionals to collect accurate data about users' actions. Considering this fact, we will not focus on tools and methods of data collection.…”

“…In shorter form r[]. Such (or similar) representation is widely used in Markov models [7] to describe the sequence of users' actions.…”

“…From this perspective, executing the same behavioral act using different amounts of time could be considered various strategies. [7] and [8] show the effective use of time utilized by users for a certain step to amend the model of their behavior. Let us take i ∈ N as an amount of time used to perform strategy s i u .…”

“…Besides applying the same strategy within a different time, users may use different strategies, covering r[]. This statement was described in (7). Practically it will case the utilization of various amounts of time in most of the cases so that it will be covered by (26).…”

“…A detailed review of the group of works applying different approaches in clickstream analysis and brief evaluation was proposed in [6]. In [7] a comprehensive probabilistic model for the impact of time on users' actions and its value was built. More insights from users' behavior are highlighted in [8,9]:…”