The code was tested using Python version 3.9. For other necessary libraries please use requirements.txt
pip install -r requirements.txtPlease download the data from Kaggle here
This project works with Amazon Reviews Data and focusses on:
- Sentiment Analysis using Large Language Model (LLM)
- Fine-Tuning the Large Language Model
- Data Augmentation using Back-Translation technique
- Data Balancing (working with imbalanced Dataset)
- Random Oversampling
- Weighted Loss
sentiment-analysis-llm/
├── amazon reviews sentiment analysis.ipynb # the notebook with data augmentation, data balancing, sentiment analysis and fine-tuning using (HuggingFace)
├── README.md # readme file
├── requirements.txt # all necessary libraries
├── Emoticon.py # dictionary with emojis as a symbol with the text description
The main findings of the code can be found at the post:
"Sentiment Analysis of Customer Product Reviews using Large Language Model (LLM)" available here.
Must give credit to Kaggle for providing the data.