Sentiment Analysis Genshin Impact Game to Know Players Reaction and Expectation Using Support Vector Machine

Sugianto, Muhammad Refansyach and Qur’ania, Arie and Delli W, Fajar (2023) Sentiment Analysis Genshin Impact Game to Know Players Reaction and Expectation Using Support Vector Machine. Skripsi thesis, Universitas Pakuan.

[img] Text
Skripsi Refansyach 065119226.pdf

Download (3MB)

Abstract

Sentiment Analysis Genshin Impact Game to Know Players Reaction and Expectation Using Support Vector Machine Arie Qur’ania1 , Fajar Delli Wihartiko2 , Muhammad Refansyach Sugianto3 1Program Studi Ilmu Komputer, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Pakuan, Bogor, Jawa Barat, 16143, Indonesia Abstract As the popularity of online games increases, it becomes important to know the sentiment of players towards a particular game title. This study aims to determine the reactions and expectations of players in community of game titled Genshin Impact, based on aspect of character, event and game as a whole in game version 3.6. In this case, sentiment analysis is used to classify players opinion on YouTube comment and determine whether the opinion is positive, neutral or negative. The data used are player comments on 6 YouTube videos uploaded by the official Genshin Impact Channel that promoting the new version of the game, version 3.6. The process involved several steps like Scraping data comment, Cleaning, Case Folding, Tokenization, Filtering, Lemmatization, Feature Extraction using TF-IDF and Classification using Support Vector Machine (SVM). The results showed that the overall sentiment of players towards version 3.6 was positive. However, there are some player expectations and complaints that can be taken into consideration by developers. In the character aspect, players expect the new characters, Kaveh and Baizhu, to have a stronger appeal. On the game aspect, players want long-term content to keep them motivated to keep playing. On the event aspect, players complain that the events in this version seem very long which makes players quickly get bored. The classification model with the SVM Algorithm succeeded in providing a fairly high accuracy of 83% in the game and character aspects, while in the event aspect the model provided a fairly good accuracy of 69%. Keywords: Sentiment Analysis; Genshin Impact; Version 3.6; expectations; SVM

Item Type: Thesis (Skripsi)
Subjects: Fakultas Ilmu Pengetahuan Alam dan Matematika > Ilmu Komputer
Divisions: Fakultas Matematika dan Ilmu Pengetahuan Alam > Ilmu Komputer
Depositing User: PERPUSTAKAAN FAKULTAS MATEMATIKA DAN ILMU PENGETAHUAN ALAM UNPAK
Date Deposited: 17 May 2024 03:25
Last Modified: 17 May 2024 03:25
URI: http://eprints.unpak.ac.id/id/eprint/7677

Actions (login required)

View Item View Item