Implementasi Algoritma K-Means untuk Menganalisa Pemain Video Game Mobile Legend untuk Mengetahui Tipe Hero dan Role yang Sering Digunakan pada Setiap Kalangan

Abstract

Video game adalah dimasyarakat. Bahkan Esports (kompetisi video game ) semakin diakui sebagai olahraga rekreasi. Contoh Esport yang sedang digemari dikalangan saat ini adalah Mobile Legend s. Pada smartphone Android dan IOS keduanya dapat menjalankan aplikasi video game Mobile Legend s. Namun, belum pernah ada yang membahas penggunaan Role - playing bawaan developer sebagai hero alternatif untuk menyederhanakan panduan kombinasi Role - playing dalam game , mengelompokkan karakter sesuai karakteristiknya, dan menggunakan karakter sesuai keinginan pemain. Untuk itu, penelitian ini menggunakan algoritma k-means untuk menyusun pengelompokan hero dan Role playing dalam video game Mobile Legend . Dari hasil penelitian yang dilakukan, terdapat tiga kelompok hero yang memiliki Abstract Video games are one of the technologies in the entertainment field that are growing rapidly in the community. Even Esports (video game competitions) are increasingly recognized by the industry as a recreational sport. An example of an esport that is currently popular among people is Mobile Legends. Both Android and IOS smartphones can run the Mobile Legends video game application. However, no one has ever discussed using the developer's built-in Role-playing as an alternative hero to simplify the Role-playing combination guide in the game, grouping characters according to their characteristics, and using characters according to the player's wishes. For this reason, this study uses the k-means algorithm to compile the grouping of heroes and Role-playing in the Mobile Legend video game. From the results of the research conducted, there are three groups of heroes that have similar characteristics, namely Mage, Marksman, and Fighter. The K-Means algorithm used can help determine the order of hero launches. Based on the results of clustering testing using the K-Means method, data in cluster 1 was obtained with an average accuracy of 5% hero support; heroes Tanks 10%; Fighter heroes 20%; hero Mage 25%; hero Assassin 20%; Marksman heroes 20%; then for the average age of 25.1 ; win rate 67.3; total matches 5863.4 and total MVP 1664,1. Players with hero support, tank, fighter and mage types have Roles as supporting tools to develop a core Role or Marksman and assassin heroes who have high capabilities as winners in the team, with an average user who has a supporting Role of 0.15 and an average user with 0.2 . Role core