Investigating Stream Clustering and Correlation for Heterogeneous Gaming Data

In today’s world it is important to make data-driven decisions. However, as the data volume increases and as it evolves continuously and dynamically, in other words as we get more and more streaming data, it becomes more challenging to use such data for decision making and support. Thus in this research, we aim to develop and evaluate the stream clustering algorithms to study how they can assist the decision making process in online gaming platforms. To this end we will also study correlation techniques in conjunction with clustering to address the issues with hereogenous data sources. Our overall aim is to investigate the feasibility of such tecniques in identifying player behaviours for tactical online marketing strategies.

Faculty Supervisor:

Dr. Nur Zincir-Heywood

Student:

S. Baran Tatar

Partner:

GTech Canada ULC

Discipline:

Computer science

Sector:

Information and communications technologies

University:

Dalhousie University