aaron-rae-nicolas/product-specific-model-llm-annotated

Product Specific Aspect Identifier Model on Taglish Product and Service Reviews

This model is a fine-tuned version of Gemma-3-1b-pt for multi-label classification of product review specific aspects written in Taglish (Tagalog-English code-switched text). It was trained using the Gemini LLM annotated dataset of "Product and Service Reviews", focusing on various product qualities such as color, durability, functionality, and etc. The training incorporates LoRA for efficient adaptation and uses class weighting to handle label imbalance.

The model supports automated analysis of Taglish product reviews by identifying multiple relevant aspects simultaneously.

Citations

All model details and training setups can be found in our papers. If you use our model or find it useful in your projects, please cite our work:

@article{dionela2025aspectextraction,
  title={Aspect Extraction from E-Commerce Product and Service Reviews},
  author={Valiant Lance Dionela, Fatima Kriselle Dy, Robin James Hombrebueno, Aaron Rae Nicolas, and Charibeth Cheng},
  journal={ },
  year={2025}
}

Data and Other Resources

The training and evaluation datasets, as well as label definitions, can be found in the respective CSV files under the datasets/ directory. Additional resources and related benchmark data may be available on the project website.