Gamification of Workplace Activities - Smart Information Systems: Computational Intelligence for Real-Life Applications - page 261

9.5 Conclusion and Outlook
In this chapter, we address the challenge of applying gamification in a workplace
environment from two directions. First, we present a user study that focused on
determining users' perception of gamification and their actual interaction with a
gamified system. We conclude from this initial study that there is a relationship
between the perceived and the actual role of gamification principles in a workplace
environment.
Under the assumption that different users experience the same game design ele-
ments differently, we then focus on automatically identifying play-personas. This
will allow us to create gamified systems that adapt the application of gamification
elements based on users' types. In this context, we define the gamification problem
as the problem of assigning each user a game design element such that their expected
contribution to achieve some pre-specified goal is maximized.
One way to assign design elements to users is by means of customer segmenta-
tion. In marketing theory, segmentation aims at identifying customer groups in order
to better match the needs and wants of customers. For games these customer seg-
ments correspond to different player types based on character theory. Once a user is
classified into a customer segment, an appropriate design element for that segment
is selected and assigned to that user. The hardest part of this approach is to design
categories that correspond to various dimensions describing characteristic features
of users such as the multiple motivations of varying degrees existing simultaneously
across users and user types.
In order to avoid assignments of design elements to users via the indirection of
customer segments and user types frommarketing and character theory, respectively,
we aim at learning a predictive model based on statistical principles that directly
classify users to game design elements. Based on user interaction with game design
elements, we suggest to solve the learning problem by means of matrix factorization.
The latent factors discovered by a matrix factorization model may be interpreted as
characteristic properties of game design elements. User factors describe to which
extent a user prefers such characteristic features. Thus, the latent factors can be
regarded as a computerized alternative to the aforementioned customer segments
and user types.
Aiming to keep the gamification model simple, we ignore time dynamics of user
preferences leaving this issue open for future research. In addition, learning classifiers
based on user behavior characteristics is a second issue for future research. The
main challenges consist in constructing a useful utility function when using matrix
factorization and generating useful behavior features when learning classifiers. Due
to lack of publicly available data, empirical evaluations are currently not possible.
Therefore, this contribution aims at directing the design process of gamification to a
more principled way based on statistical grounds.
Acknowledgments The research leading to these results was performed in the CrowdRec project,
which has received funding from the European Union Seventh Framework Programme FP7/2007-
2013 under grant agreement n ⓦ 610594.