• The Korean Journal of Applied Statistics
• /
• v.37 no.4
• /
• pp.525-539
• /
• 2024

Massive multiplayer online role playing game (MMORPG) is a common type of game these days. Predicting user churn in MMORPG is a crucial task. The retention rate of users is deeply associated with the lifespan and revenue of the service. If the churn of a specific user can be predicted in advance, targeted promotions can be used to encourage their stay. Therefore, not only the accuracy of churn prediction but also the speed at which signs of churn can be detected is important. In this paper, we propose methods to identify early signs of churn by utilizing the daily predicted user retention probabilities. We train various deep learning and machine learning models using log data and estimate user retention probabilities. By analyzing the change patterns in these probabilities, we provide empirical rules for early identification of users at high risk of churn. Performance evaluations confirm that our methodology is more effective at detecting high risk users than existing methods based on login days. Finally, we suggest novel methods for customized marketing strategies. For this purpose, we provide guidelines of the percentage of accessed users who are at risk of churn.

• Journal of Korea Game Society
• /
• v.18 no.5
• /
• pp.47-58
• /
• 2018

Game bots distribute a large amount of goods or items used in a game, thereby lowering the value of game goods and items. Also, a large number of game bots hunt monsters and collect items, which hinders ordinary users from enjoying content normally. However, no research has been done on the type of user and the type of activity that the increase in bots specifically affects. Therefore, this study provides a practical implication to encourage users to use games by classifying types based on the game users' access data and analyzing the correlation with user departure due to the increase of bots.

• Journal of KIISE
• /
• v.44 no.3
• /
• pp.314-322
• /
• 2017

In MMORPG (Massively Multiplayer Online Role-Playing Game), users advance their own characters to get to the maximum (max) level by performing given tasks in the game scenario. Although it is crucial to retain users with high levels for running online games successfully, little efforts have been paid to investigate them. In this study, by analyzing approximately 60 million in-game logs of over 50,000 users, we aimed to investigate the process through which users achieve the max level and churn of such users since the moment of achieving the max level, and determine possible indicators related to churn after the max level. Based on the result, we can predict churn of the max level users by employing behavioral patterns before the max level. Moreover, we found users who are socially active and communicate with many people before the max level are less likely to leave the service (p<0.05). This study supports that communication patterns are important factors for persistent usage of the users who achieve the max level, which has practical implications to guide elite users on enjoying online games in the long run.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.6
• /
• pp.1431-1439
• /
• 2017

In the security field, log analysis is important to detect malware or abnormal behavior. Recently, image visualization techniques for malware dectection becomes to a major part of security. These techniques can also be used in online games. Users can leave a game when they felt bad experience from game bot, automatic hunting programs, malicious code, etc. This churning can damage online game's profit and longevity of service if game operators cannot detect this kind of events in time. In this paper, we propose a new technique of PNG image conversion based churn prediction to improve the efficiency of data analysis for the first. By using this log compression technique, we can reduce the size of log files by 52,849 times smaller and increase the analysis speed without features analysis. Second, we apply data mining technique to predict user's churn with a real dataset from Blade & Soul developed by NCSoft. As a result, we can identify potential churners with a high accuracy of 97%.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.179-182
• /
• 2021

게임의 운영과 관련하여 부정 사건은 유저들의 이탈을 부르며 매출에 중요한 영향을 미친다. 이러한 사건 사고는 여러 가지 게임에서 발생하고 있으며 사건에 따라서 대처도 다르고 유저의 반응도 다르다. 본 논문에서는 온라인 게임에서 일어나는 사건 사고를 분석하고 유저의 동향을 파악하여 게임 운영과 관련하여 사건 사고가 발생했을 때 잘 대처하여 유저들에게 좋은 평가를 받고 사건의 피해를 최소화하며 유저들이 다시 게임에 돌아오도록 하기 위해 중요한 점이 무엇인지 제시하고자 한다.

• Journal of Korea Game Society
• /
• v.19 no.4
• /
• pp.27-36
• /
• 2019

Approaches to find hidden values using various and enormous amount of data are on the rise. As big data processing becomes easier, companies directly collects data generated from users and analyzes as necessary to produce insights. User-based data are utilized to predict patterns of gameplay, in-game symptom, eventually enhancing gaming. Accordingly, in this study, we tried to analyze the gaming strategy and user activity patterns utilizing Battlegrounds in-game data to detect the in-game hack.

• Journal of Digital Convergence
• /
• v.18 no.2
• /
• pp.421-426
• /
• 2020

The global game market is growing evenly across a variety of platforms, but in the Korean game market is a decline in growth rates on PC and mobile platforms, which account for more than 80% of the total Korean game market, which requires an alternative. In this study, propose a multi-platform launch of IP-based games as a way to increase the growth rate of the Korean game market. It has been analyzed that multi-platform launch methods can be divided into Stand-alone Multi-platform method, Interlocking multi-platform method, and Upgrade-interlinking Multi-platform method, respectively, and the effect of expanding, providing more satisfactory game play environment and preventing them from escaping to competitive games. Given the limited case analysis in this study, further studies are needed to propose more effective multi-platform utilization measures.