Whales, Dolphins, or Minnows? Towards the Player Clustering in Free Online Games Based on Purchasing Behavior via Data Mining Technique

• Published in: 2018 IEEE International Conference on Big Data (Big Data) pp. 4101 - 4108
• Main Authors: Yang, Wanshan, Yang, Gemeng, Huang, Ting, Chen, Lijun, Liu, Youjian Eugene
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2018
• Subjects: Clustering; Clustering algorithms; Companies; Data analysis; Data mining; Dolphins; free-to-play games; Games; k-means clustering; online games; player modeling; unsupervised learning; Whales
• DOI: 10.1109/BigData.2018.8622067

The free online game market is one of the fastest growing and most profitable markets in the electronic game industry. In order to improve players' gaming experience and increase revenue, game data analytics is usually applied to support game development, in-game event design, and game operation. In particular, the game companies strive to understand and characterize the players' buying power and design marketing strategies or incentives accordingly to increase their revenues. In this paper, we propose a novel algorithm to perform player clustering into the groups of high spenders (whales), moderate spenders (dolphins), and low spenders (minnows) based on the players' purchase records via data mining technique. Evaluations including an A/B test with a typical popular free online game show that our proposed algorithm can effectively cluster the players in terms of their purchasing behavior and the game companies can benefit from our algorithm.