Indoroberta Base Sentiment Indonesian Social Media

🚀 Sentiment Classification Model

This project focuses on sentiment classification, and the proposed model achieves high - accuracy results.

🚀 Quick Start

For reference on model card metadata, see the spec: Model Card Metadata Spec Doc / guide: Model Cards Documentation

✨ Features

The proposed sentiment classification model has achieved excellent performance. The accuracy and F1 - score values for the model resulted in 98% accuracy and 97.4% F1 - score (Epoch = 10, Batch Size = 32).

📚 Documentation

Citation

• Plain Text: Y. O. Sihombing, R. Fuad Rachmadi, S. Sumpeno and M. J. Mubarok, "Optimizing IndoRoBERTa Model for Multi - Class Classification of Sentiment & Emotion on Indonesian Twitter," 2024 IEEE 10th Information Technology International Seminar (ITIS), Surabaya, Indonesia, 2024, pp. 12 - 17, doi: 10.1109/ITIS64716.2024.10845566.
• BibTex:

@INPROCEEDINGS{10845566,
  author={Sihombing, Yogie Oktavianus and Fuad Rachmadi, Reza and Sumpeno, Surya and Mubarok, Moh. Jabir},
  booktitle={2024 IEEE 10th Information Technology International Seminar (ITIS)}, 
  title={Optimizing IndoRoBERTa Model for Multi - Class Classification of Sentiment & Emotion on Indonesian Twitter}, 
  year={2024},
  volume={},
  number={},
  pages={12 - 17},
  keywords={Seminars;Sentiment analysis;Accuracy;Social networking (online);Blogs;Text categorization;Predictive models;Emotional responses;Information technology;Optimization;IndoRoBERTa;classification;sentiment;emotion;Indonesian;Twitter},
  doi={10.1109/ITIS64716.2024.10845566}
}

Acknowledgment

The authors express their gratitude to all internal and external sources that consistently provide support for our study, especially to Wilson Wongso, Steven Limcorn, and Cahya Wirawan for their openly usable models on HuggingFace.