Game Data Mining Competition on Churn Prediction and Survival Analysis using Commercial Game Log Data

02/07/2018 ∙ by Kyung-Joong Kim, et al. ∙ 0

Usually, game companies avoid sharing their game data with external researchers and only a few number of research groups were granted limited access to gam data so far. Such reluctance of the companies to make data publicly available closes doors on the wide use and development of the data mining techniques and AI research specific to the game industry. In this work, we propose an international competition on game data mining using the commercial game log data from one of the major game companies in Korea: NCSOFT. It provides opportunities to many researchers who wish to develop and apply state-of-the-art data mining techniques to game log data by making the data open. The data has been collected from Blade & Soul, Action Role Playing Game, from NCSoft. The data comprises of approximately 100GB of game logs from 10,000 players. The main aim of the competition was to predict whether a player would churn and when the player would churn during two different periods between which its business model was changed to the free-to-play model from monthly fixed charged one. The final result of the competition reveals that the highly ranked competitors used deep learning, tree boosting, and linear regression.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 2

page 11

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.