Anggraeni, Aprisa (2022) Prediksi Jumlah Pemain Video Game Pada Platform Steam dengan Menggunakan Model ARIMA. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 05211840000105-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
05211840000105-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2024. Download (3MB) | Request a copy |
|
Text
05211840000105-Undergraduate_Thesis.pdf Restricted to Registered users only Download (3MB) | Request a copy |
Abstract
Berkembangnya penggunaan internet secara global telah mendorong industri video game untuk terus berkembang dan berinovasi pula. Pengembang video game bukan hanya perlu untuk membuat produk mereka sebagus mungkin, namun mereka juga perlu untuk memenuhi keinginan para pemain game. Pengembang perlu untuk mengetahui video game seperti apa yang saat ini tengah banyak dimainkan oleh pemain, sehingga pengembangan game yang dilakukan dapat memberikan produk yang sesuai. Jumlah pemain video game menjadi salah satu acuan pengembang video game dalam membuat dan memasarkan game yang mereka kembangkan. Platform besar seperti Steam mencakup spektrum pemain yang luas dengan berbagai perilaku pemain yang dapat berubah dengan drastis. Peramalan perlu dilakukan sehingga pengembang video game dapat memperkirakan perilaku dan pengalaman pemain yang akan membantu dalam memastikan pengembangan video game yang sukses. Pengerjaan Tugas Akhir ini menggunakan model Autoregressive Integrated Moving Average (ARIMA) untuk memperkirakan jumlah pemain video game pada platform Steam. ARIMA telah digunakan dalam banyak kasus peramalan karena keakuratan, efisiensi, dan fleksibilitas peramalannya dalam merepresentasikan beberapa jenis deret waktu yang berbeda. Hasil pemodelan menggunakan ARIMA rata – rata memberikan hasil dengan nilai akurasi prediksi yang tinggi dan nilai error
vi
yang relatif rendah. Data yang digunakan dala pengerjaan Tugas Akhir ini adalah Steam Game Dataset yang berisi mengenai riwayat jumlah pemain video game dalam platform Steam. Sedangkan genre video game yang akan menjadi fokusan utama adalah Action, RPG, Massively Multiplayer, Sports, dan Strategy dimana kelimanya merupakan game dengan jumlah pemain terbanyak dalam platform Steam. Genre Strategy yang diwakili oleh game DOTA 2 menggunakan model ARIMA(5,1,5) dengan MAPE 3.7%. Genre Action diwakili Counter Strike: Global Offensive menggunakan ARIMA(5,1,5) dengan MAPE 1.7%. Genre RPG diwakili oleh Elder Scroll V: Skyrim menggunakan ARIMA(5,1,5) dengan MAPE 5.05%. Genre MMO diwakili oleh Final Fantasy XIV Online menggunakan ARIMA(4,1,5) dengan MAPE 5.05%. Terakhir genre sports diwakili oleh Football Manager 2015 menggunakan model ARIMA(2,1,1) dengan MAPE 4,2%. Tujuan akhir dari pengerjaan Tugas Akhir ini adalah untuk mengetahui tren dari masing – masing genre video game yang digunakan dalam platform Steam dalam 35 hari kedepan dihitung dari hari terakhir data dikumpulkan. Diberikan pula saran mengenai waktu yang tepat untuk melakukan pembaruhan dalam game berdasarkan dari hasil peramalan yang telah dilakukan
==================================================================================================
The growing use of the internet globally has encouraged the video game industry to continue to grow and innovate as well. Video game developers not only need to make their products as good as possible, but they also need to cater to the desires of gamers. Developers need to know what kind of video games are currently being played by many players, so that the game development carried out can provide the appropriate product. The number of video game players is one of the references for video game developers in creating and marketing the games they develop. A large platform like Steam covers a wide spectrum of players with a wide range of player behavior that can change drastically. Forecasting needs to be done so that video game developers can predict player behavior and experience which will help in ensuring successful video game development. This Final Project uses the Autoregressive Integrated Moving Average (ARIMA) model to estimate the number of video game players on the Steam platform. ARIMA has been used in many forecasting cases because of its accuracy, efficiency, and forecasting flexibility in representing different types of time series. The results of modeling using ARIMA on average give results with a high predictive accuracy value and a relatively low error value. The data used in this Final Project is the Steam Game Dataset which contains the history of the number of video game players on the Steam platform. While the video game genres that will be the main focus are Action, RPG, Massively Multiplayer, Sports, and
viii
Strategy where the five are the games with the highest number of players on the Steam platform. The Strategy genre represented by the DOTA 2 game uses the ARIMA(5,1,5) model with a MAPE of 3.7%. The Action genre is represented by Counter Strike: Global Offensive using ARIMA(5,1,5) with a 1.7% MAPE. The RPG genre is represented by Elder Scroll V: Skyrim using ARIMA(5,1,5) with a MAPE of 5.05%. The MMO genre is represented by Final Fantasy XIV Online using ARIMA(4,1,5) with a 5.05% MAPE. Finally, the sports genre is represented by Football Manager 2015 using the ARIMA(2,1,1) model with a MAPE of 4.2%. The ultimate goal of working on this Final Project is to find out the trend of each video game genre used on the Steam platform in the next 35 days calculated from the last day the data was collected. Suggestions are also given regarding the right time to make changes in the game based on the forecasting results that have been done.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Uncontrolled Keywords: | video game, peramalan, Steam, ARIMA video game, forecasting, Steam, ARIMA |
| Subjects: | T Technology > T Technology (General) T Technology > T Technology (General) > T57.5 Data Processing |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis |
| Depositing User: | Aprisa Anggraeni |
| Date Deposited: | 11 Feb 2022 03:43 |
| Last Modified: | 11 Feb 2022 03:43 |
| URI: | http://repository.its.ac.id/id/eprint/93685 |
Actions (login required)
 |
View Item |