Amazon Fine Food Reviews - RoBERTa Sentiment Analysis
Model Description
This is a fine-tuned RoBERTa model for Sentiment Analysis, trained on the Amazon Fine Food Reviews dataset. It classifies English food reviews into three categories: positive, neutral, and negative.
This model is part of an academic Web Mining project developed at EMSI Marrakech.
Performance
- Accuracy: ~84%
- F1-Score (Macro): ~0.84
Intended Uses
- Intended Use: Analyzing customer sentiment in e-commerce food reviews.
- Language: English
How to use
You can use this model directly with a pipeline for text classification:
from transformers import pipeline
# Load the model
sentiment_classifier = pipeline("text-classification", model="aablaess/amazon-reviews-roberta")
# Analyze a review
result = sentiment_classifier("This product is absolutely delicious, I will buy it again!")
print(result)
# [{'label': 'positive', 'score': 0.98...}]
- Downloads last month
- 33