Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal METHODIKA: Jurnal Teknik Informatika dan Sistem Informasi Jurnal Info Sains : Informatika dan Sains
Ahmad Fadhil N
Unknown Affiliation
Computer Science & IT

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Sentiment Analysis Using Transformers Ahmad Fadhil N
Jurnal Info Sains : Informatika dan Sains Vol. 14 No. 03 (2024): Informatika dan Sains , Edition July - September 2024
Publisher : SEAN Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

This study examines how transformer-based models, such as BERT and DistilBERT, can be used for sentiment analysis of IMDb movie reviews. The goal of the experiment was to find a balance between accuracy and computational efficiency, evaluating how well both models performed with different training parameters. BERT was able to reach a peak accuracy of 91.39% in three epochs, taking a total of 54 minutes to train. On the other hand, DistilBERT achieved a similar accuracy of 91.80% in only 38 minutes and 25 seconds. Although there was a slight variance in accuracy, DistilBERT proved to be a much more efficient option for training, thus becoming a feasible substitute for environments with limited resources. The findings were contrasted against R. Talibzade's (2023) research, which obtained a 98% accuracy rate using BERT but needed 12 hours of training, illustrating the balance between accuracy and training duration. Potential upcoming tasks involve refining further, testing with bigger datasets, investigating alternative transformer models, and utilizing more resource-efficient training methods to improve performance without sacrificing efficiency.
EVALUATING THE EFFECTIVENESS OF DISTILBERT FOR SENTIMENT ANALYSIS OF PLAYER FEEDBACK IN GAME DEVELOPMENT Ahmad Fadhil N; Saragih, Eka Parima
METHODIKA: Jurnal Teknik Informatika dan Sistem Informasi Vol. 11 No. 2 (2025): Volume 11 Nomor 2 Tahun 2025
Publisher : Universitas Methodist Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46880/mtk.v11i2.4498

Abstract

Real-time sentiment analysis (SA) plays an increasingly vital role in enhancing player experience through emotion-aware game design. By enabling systems such as dynamic difficulty adjustment, adaptive non-playable character (NPC) behavior, and responsive narrative progression, SA allows games to respond intelligently to player emotions. This study investigates the effectiveness of DistilBERT, a lightweight transformer-based language model, for multi-label emotion classification using the GoEmotions dataset, which includes 27 fine-grained emotion categories. The model’s performance was evaluated in terms of classification accuracy and computational efficiency. Experimental results reveal that DistilBERT delivers surprisingly strong performance despite its reduced size, making it a viable candidate for real-time applications in resource-constrained environments. These findings indicate that lightweight transformer models can support emotionally adaptive gameplay without significant trade-offs in latency or accuracy. Future work will focus on integrating DistilBERT into a live game environment to assess its impact on player engagement and real-time system responsiveness.
EVALUATING THE EFFECTIVENESS OF DISTILBERT FOR SENTIMENT ANALYSIS OF PLAYER FEEDBACK IN GAME DEVELOPMENT Ahmad Fadhil N; Eka Parima Saragih
METHODIKA: Jurnal Teknik Informatika dan Sistem Informasi Vol. 11 No. 2 (2025): Volume 11 Nomor 2 Tahun 2025
Publisher : Universitas Methodist Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Real-time sentiment analysis (SA) plays an increasingly vital role in enhancing player experience through emotion-aware game design. By enabling systems such as dynamic difficulty adjustment, adaptive non-playable character (NPC) behavior, and responsive narrative progression, SA allows games to respond intelligently to player emotions. This study investigates the effectiveness of DistilBERT, a lightweight transformer-based language model, for multi-label emotion classification using the GoEmotions dataset, which includes 27 fine-grained emotion categories. The model’s performance was evaluated in terms of classification accuracy and computational efficiency. Experimental results reveal that DistilBERT delivers surprisingly strong performance despite its reduced size, making it a viable candidate for real-time applications in resource-constrained environments. These findings indicate that lightweight transformer models can support emotionally adaptive gameplay without significant trade-offs in latency or accuracy. Future work will focus on integrating DistilBERT into a live game environment to assess its impact on player engagement and real-time system responsiveness.
Co-Authors Eka Parima Saragih Saragih, Eka Parima