Model Klasifikasi Review Mobile App 🇮🇩

Model ini digunakan untuk mengklasifikasikan ulasan pengguna aplikasi di Mobile Store ke dalam 4 kategori:

Bug: crash, error, gagal fungsi
Feature: permintaan fitur baru
UX: pengalaman pengguna buruk (lambat, sulit, membingungkan)
Noise: spam, ambigu, sangat singkat

🧠 Base Model

Model ini dilatih berdasarkan pretrained model IndoBERT (indobenchmark/indobert-base-p1) dari IndoNLU.

📚 Sitasi IndoBERT

@inproceedings{wilie2020indonlu, title={IndoNLU: Benchmark and Resources for Evaluating Indonesian Natural Language Understanding}, author={Bryan Wilie and Karissa Vincentio and Genta Indra Winata and Samuel Cahyawijaya and X. Li and Zhi Yuan Lim and S. Soleman and R. Mahendra and Pascale Fung and Syafri Bahar and A. Purwarianti}, booktitle={Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing}, year={2020} }

Dataset

Dilatih pada dataset berlabel berbahasa Indonesia (custom dataset).

Akurasi

F1 Score terbaik: 0.64

Cara Penggunaan

from transformers import AutoTokenizer, AutoModelForSequenceClassification

model = AutoModelForSequenceClassification.from_pretrained("gilangsp/01-classification-review-mobilestore-model-f1-064")
tokenizer = AutoTokenizer.from_pretrained("gilangsp/01-classification-review-mobilestore-model-f1-064")

inputs = tokenizer("aplikasi sering crash pas buka kamera", return_tensors="pt")
outputs = model(**inputs)