Prediction outcome for massive multiplayer online games using data mining

• Main Authors: Diah, N.M (Author), Ismail, M. (Author), Kamal, N.A.M (Author), Samsurim, S. (Author)
• Format: Article
• Language: English
• Published: Institute of Advanced Engineering and Science 2018
• Subjects: Classification; Games; IBK; J48; MMOG; Naïve bayes
• Online Access: View Fulltext in Publisher; View in Scopus
• ISBN: 25024752 (ISSN)
• DOI: 10.11591/ijeecs.v11.i1.pp248-255

Massive Multiplayer Online (MMO) game is one of the famous game genres among teenagers nowadays. MMO games allow gamers to interact and play with up to thousand players. Rainbow Six Siege (RSS) belongs to MMO type of game. However, due to many operators that are available in this game, the player needs to choose the right operator to counter the enemy operator. Therefore, based on the characteristic of the selected operator, this paper attempted to predict the outcomes of the game. In our prediction model, characteristics for these operators were extracted from 120 live stream replays. Three classification algorithms were utilized to predict the outcome of the game. Among these algorithms, IBK had obtained outstanding performance in the dataset. The accuracy of the model is 93.75%, applying 5-fold cross-validation test. The success rate reveals that our proposed model is suitable to predict the outcome of the game. © 2018 Institute of Advanced Engineering and Science. All rights reserved.