Perhitungan Kenaikan Atribut Gameplay Untuk Pemain Dan Non-Player Character Pada Permainan Role-Playing Game Berbasis K-NN Dan Naive Bayes

Widiyanto, Nur Rohman (2020) Perhitungan Kenaikan Atribut Gameplay Untuk Pemain Dan Non-Player Character Pada Permainan Role-Playing Game Berbasis K-NN Dan Naive Bayes. Masters thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 07111650052005-Master_Thesis.pdf]
Preview
Text
07111650052005-Master_Thesis.pdf

Download (15MB) | Preview

Abstract

Permainan dengan genre Role-Playing Game (RPG) merupakan permainan yang bersifat kompetitif, antara pemain melawan pemain ataupun melawan musuh yang berupa Non-Player Character (NPC). Banyak pengembang permainan dalam pembuatan permainan itu sendiri masih menggunakan cara manual dalam penentuan atribut gameplay untuk karakter pemain ataupun musuh. Terlebih lagi, saat sebuah permainan memiliki banyak karakter, seperti hanyalnya banyak karakter pemain (contohnya pada JRPG dan TRPG) dan juga banyak musuh. Pada penelitian ini diimplementasikan beberapa pendekatan seperti halnya k-NN, Distribusi Normal, dan Naive Bayes yang akan digunakan dalam program penghitung kenaikan atribut gameplay pada karakter pemain dan musuh secara otomatis. Pada program tersebut dibutuhkan parameter masukan yang akan menentukan atribut gameplay yang akan dihasilkan. Untuk karakter pemain, dibuatlah skenario atribut gameplay berdasarkan peran setiap karakter yang ingin dibuat seperti knight, priest, assassin, dan lain lain. Sedangkan pada karakter musuh, dalam distribusi atribut gameplay juga dibagi menjadi beberapa tipe musuh yang dicontohkan seperti mixed, hard strength, hard magic, dan lain-lain. Setelah itu, hasil atribut gamepla diklasifikasi dengan menggunakan Neural Network Multiclass Classification. Hal tersebut bertujuan untuk menghitung tingkat kesesuaian atribut gameplay dari karakter pemain dan musuh yang dihasilkan. Operasi tersebut dilakukan secara terpisah pada atribut gameplay pemain dan musuh, karena tidak adanya hubungan saat pembuatan atau perhitungan. Masing-masing dibagi kedalam data training dan testing, dengan perbandingan 70% dan 30%. Proses tersebut menghasilkan keluaran berupa tipe karakter pemain dan musuh pada data testing, hal tersebut diperoleh dari proses training. Presentase kesesuaian tersebut diperoleh dengan membandingkan tipe pada karakter yang diperoleh dari hasil klasifikasi pada data testing dengan tipe dari karakter yang sebenarnya, maka diperolehlah sebuah presentase banyaknya karakter yang berhasil terklasifikasi.
=====================================================================================================
The game with the Role-Playing Game (RPG) genre was a competitive game, between players against other players or enemies in the form of a Non-Player Character (NPC). Many game developers in making the game itself still use manual methods in determining gameplay attributes for player or enemy characters. Especially, when the game had many player characters (for example in JRPG and TRPG) and also the enemy. In this research, several approaches were implemented such as k-NN, Normal Distribution, and Naive Bayes which will be used in the program to automatically calculate the growth in gameplay attributes of the player and enemy characters. The program requires input parameters that will determine the result of gameplay attributes. For the player character, the gameplay attribute scenario was created based on the role of each character that user want to make such as knight, priest, assassin, and others. Whereas for enemy characters, the distribution of gameplay attributes also divided into several types of enemies that were exemplified, such as mixed, hard strength, hard magic, and others. After that, the results of the gameplay attributes classified using the Neural Network Multiclass Classification. It aims to calculate the level of suitability of the gameplay attributes of the resulting player and enemy characters. The operation is carried out separately on the player's and enemy's gameplay attributes because there is no correlation during creation or calculation. Each result divided into training and testing data, with a ratio of 70% and 30%. This process produces output in the form of player's and enemy's character types on the testing data, that obtained from the training process. The percentage of suitability obtained by comparing the types of characters from the classification results in the testing data with the types of the actual characters, so the percentage of the number of characters that were classified had obtained.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Role-Playing Game, Gameplay Attributes, k-NN, Naive Bayes, Neural Network, Classification, Role-Playing Game, Atribut Gameplay, k-NN, Naive Bayes, Neural Network, Klasifikasi.
Subjects: G Geography. Anthropology. Recreation > GV Recreation Leisure > GV1469.2 Computer games
H Social Sciences > HD Industries. Land use. Labor > HD108 Classification (Theory. Method. Relation to other subjects )
Q Science > QA Mathematics > QA76.6 Computer programming.
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User: Nur Rohman Widiyanto
Date Deposited: 29 Aug 2020 08:11
Last Modified: 09 Jan 2024 08:41
URI: http://repository.its.ac.id/id/eprint/81618

Actions (login required)

View Item View Item