Claim Missing Document
Check
Articles

Found 2 Documents
Search

Game Real Time Action RPG Online Berbasis Flash: Alternatif Bermain Game dengan Spesifikasi Perangkat Keras Minimal Elidjen Elidjen; Paulus Adi Adi Purnomo; Budi Prasetyo
ComTech: Computer, Mathematics and Engineering Applications Vol. 3 No. 2 (2012): ComTech
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/comtech.v3i2.2304

Abstract

Various games are available for a variety of community as one means of entertainment which is getting more popular these days. However, sometimes several games can not be played on computer due to insufficient hardware specifications support. This situation developed the making of ChainKiller Game, a Flash based real time action RPG online game to provide an alternative as well as interactive and interesting game with a minimum hardware specification. The sequential linear process model is used to develop the overall game. Besides, a literature study was implemented to enrich the insight to develop the interesting mini game. The real-time action RPG online game which is based on Flash can be easyly used with minimum hardware specification. 
A Comparative Analysis of MultinomialNB, SVM, and BERT on Garuda Indonesia Twitter Sentiment Budi Prasetyo; Ahmad Yusuf Al-Majid; Suharjito
PIKSEL : Penelitian Ilmu Komputer Sistem Embedded and Logic Vol. 12 No. 2 (2024): September 2024
Publisher : LPPM Universitas Islam 45 Bekasi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33558/piksel.v12i2.9966

Abstract

This study investigates customer sentiment towards Garuda Indonesia Airlines (GIA) using sentiment analysis of Twitter data. The research aims to identify prevailing sentiments, uncover common themes in customer feedback, and provide recommendations for improving customer satisfaction and brand loyalty. A dataset of 1,250 tweets from March 2007 to July 2024 was collected and pre-processed, including cleaning, language detection, and tokenization. Sentiment analysis was conducted using three models: MultinomialNB, SVM, and BERT.The results indicate that BERT outperformed both MultinomialNB and SVM in sentiment classification accuracy, achieving 75.6%. This highlights the effectiveness of BERT in capturing contextual meaning within customer reviews. The findings of this research will contribute to a deeper understanding of customer sentiment towards GIA and inform strategies for enhancing customer experience and brand image.