The ability to control player retention has been long sought after by both game designers and game researchers.With casual games becoming more popular, it has become increasingly important to have a good understanding of how to influence player retention. In the game industry, there are many accepted rules about how to best maximize player retention in a game environment (Fields, 2014). These rules are based on expert knowledge and were created based on years of experience making games, but have not been explicitly validated. In academia, many researchers have studied what contributes or detracts from player retention in various game environments with varying amounts of success. Typically, those studying retention choose to examine how well games can retain players over the entire lifetime of the game. In my work, I examine the problem of session-level retention.
Given the recent rise in popularity of social and casual games, I believe that learning to influence session-level retention can be just as useful as learning to influence long-term retention. In this dissertation, I propose an analytics-driven technique for increasing session-level retention through the use of dynamic game adaption. This technique leverages two different types of analytics, vanity analytics and actionable analytics, to create models of player retention and then use them to dynamically alter game worlds to increase session-level retention. Vanity analytics are those that hold a great deal of predictive power but are difficult to directly affect, whereas actionable analytics are difficult to directly use in modeling player behavior but can be manipulated in real time.
This technique offers several benefits to both game designers as well as game researchers. Since this technique is data-driven, the insights gained are grounded in data which gives them more strength than some of the insights gained from expert knowledge. Also, this technique uses the inherent strengths of both vanity and actionable analytics and provides researchers with a method for incorporating models of player retention into game environments.
My research will proceed in three phases. First, I will use data-driven techniques to create computational models of session-level retention using sets of vanity analytics. Then, I will use these models to determine how to adapt two different game worlds in an attempt to influence session-level retention. Finally, these adaptions will be evaluated based on how well they actually do influence session-level retention.
In addition, I evaluate the side-effects that dynamic game adaption has on player experience to make sure that any gains in session-level retention do not come at the cost of player experience. During this evaluation, I measure the effects of dynamic game adaption on two measures of play experience, intrinsic motivation and player engagement.
|School:||North Carolina State University|
|School Location:||United States -- North Carolina|
|Source:||DAI-B 76/05(E), Dissertation Abstracts International|
|Keywords:||Data mining, Machine learning, Player retention, Session-level retention, Video games|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be