debojit01/fake-review-detector

Fake Review Detection Model

Model Description

DistilBERT model fine-tuned to detect computer-generated product reviews.

Performance

Metric	Real	Fake
Precision	0.99	0.97
Recall	0.97	0.99
F1-Score	0.98	0.98

How to Use

from transformers import pipeline

# Load model (replace with your actual username)
classifier = pipeline(
    "text-classification",
    model="debojit01/fake-review-detector"
)

# Example inference
result = classifier("This product is absolutely perfect!")
print(result)  # Output: {'label': 'REAL', 'score': 0.99}

Training Data

• 20,000 real product reviews (OR)
• 40,000 computer-generated reviews (CG)
• 50/50 train-test split

Ethical Considerations

Use responsibly. May reflect biases present in training data.