Indonesian three-way sentiment output
According to the model card, the model classifies Indonesian text as negative, positive, or neutral after fine-tuning IndoBertweet-base-uncased.
Open Source Model Profile · Aardiiiiy
indobertweet-base-Indonesian-sentiment-analysis
indobertweet-base-Indonesian-sentiment-analysis is a BERT text-classification fine-tune from Aardiiiiy for Indonesian sentiment with negative, positive, and neutral labels.
- Publisher
- Aardiiiiy
- Task
- text-classification
- Model type
- bert
- License
- mit
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
indobertweet-base-Indonesian-sentiment-analysis is published by Aardiiiiy as a text-classification fine-tune. Captured configuration identifies BertForSequenceClassification with a bert model type, and Safetensors metadata reports 110,560,515 parameters under MIT licensing. According to the model card, it fine-tunes IndoBertweet-base-uncased on Twitter and social-media reactions covering politics, disasters, and education.
Recorded capabilities
Optuna-tuned training setup
The model card describes Optuna hyperparameter tuning, up to 10 epochs, batch size 16, evaluation every 100 steps, accuracy-based checkpoint selection, and early stopping with patience 3.
BERT sequence-classification base
Captured configuration records BertForSequenceClassification with a bert model type for text classification.
Documented bias limits
The model card warns of possible socio-cultural bias, weaker accuracy on recent uncovered events, and limits of three sentiment categories for complex emotions.
Use cases in the source record
- Indonesian social-media sentiment classification into negative, positive, and neutral categories as documented by the publisher.
- Accuracy, F1-score, precision, and recall experiments following the card's documented evaluation approach.
Limitations and unknowns
- The publisher notes possible training-data bias, weaker handling of recent events, and that three categories may not capture complex emotional context.
- Provider state is historical snapshot data and should be refreshed before being presented as current availability.
- No context-window, pricing, VRAM, or supported-deployment values were extracted from this record.
Source and provenance
Source: Aardiiiiy/indobertweet-base-Indonesian-sentiment-analysis
Captured: Unknown. Processed: 2026-09-07T19:34:27.969300+00:00.
Model Card for Model ID Model Details Model Description This model is a fine-tuned version of IndoBertweet-base-uncased for Indonesian sentiment analysis. The model is designed to classify sentiment into three categories: negative, positive, and neutral. It was trained on a diverse dataset comprising reactions from Twitter and other social media platforms, covering various topics, including politics, disasters, and education. The model is optimized using Optuna for hyperparameter tuning and evaluated using accuracy, F1-score, precision, and recall metrics. Bias and Limitations Do consider that this model is trained using certain dat…
F001F002F003F004F005F006F007F010F012F013F014