Optimize user behavior in social games through identification and dynamic adaptation to player attributes

In this project, the candidate will gather new & utilize existing data in conjunction with machine learning algorithms to optimize various aspects of how one (or more) social games operates. Experiments will be proposed and run in order to gather operational data. This data will be used in conjunction with existing player & game data in order to identify, test, and apply different machine learning algorithms and statistical models. An adaptive platform will be used to apply the models/algorithms, and continually update and optimize player behavior. Optimial behavior shall determined by measuring impact on key game metrics such as number of days game is played, number of actions taken in game, revenue per user, number of social interactions and quality of social interactions (duration, frequency).

Faculty Supervisor:

Eugene Fiume

Student:

Partner:

Zynga Game Network Canada Inc

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

University of Toronto

Program: