To determine if a product is fake, we can implement a rule-based system that checks multiple indicators. Here's a Python function that evaluates a product based on key attributes:

def is_fake_product(product):
    """
    Determines if a product is fake based on predefined rules.
    Args:
        product (dict): A dictionary containing product details with keys:
            - 'price': float
            - 'seller_rating': float (0-5 scale)
            - 'description': str
            - 'category': str
            - 'avg_category_price': float (optional)
    Returns:
        bool: True if the product is likely fake, False otherwise.
    """
    # Rule 1: Unusually low price (below 50% of category average)
    if 'avg_category_price' in product:
        if product['price'] < 0.5 * product['avg_category_price']:
            return True
    # Rule 2: Suspicious seller rating (below 3.0)
    if product['seller_rating'] < 3.0:
        return True
    # Rule 3: Fake keywords in description
    fake_keywords = {'fake', 'replica', 'imitation', 'counterfeit', 'knockoff'}
    description = product['description'].lower()
    if any(keyword in description for keyword in fake_keywords):
        return True
    # Rule 4: Unusual price-to-rating ratio
    # Low-rated sellers with very low prices are high-risk
    if product['seller_rating'] <= 2.0 and product['price'] < 20:
        return True
    return False

How It Works:

1. Price Check: Compares the product's price to the average price for its category. If it's below 50% of the average, it's flagged as fake.
2. Seller Rating: Checks if the seller's rating is below 3.0 (on a 5-point scale), indicating potential unreliability.
3. Description Keywords: Scans the product description for terms like "fake," "replica," or "imitation."
4. Combined Risk Factor: Flags products from low-rated sellers (≤2.0) priced under $20, as this combination strongly suggests fakes.

Example Usage:

    'price': 99.99,
    'seller_rating': 4.5,
    'description': "Authentic leather wallet",
    'category': 'accessories',
    'avg_category_price': 120.0
}
print(is_fake_product(genuine))  # Output: False
# Fake product
fake = {
    'price': 15.0,
    'seller_rating': 2.0,
    'description': "Fake designer imitation bag",
    'category': 'accessories',
    'avg_category_price': 150.0
}
print(is_fake_product(fake))  # Output: True

Key Notes:

• Customizable Rules: Adjust thresholds (e.g., price percentage, rating cutoff) based on your data.
• Data Requirements: For accurate results, include avg_category_price (calculated from historical data).
• Scalability: For large datasets, integrate this with a machine learning model trained on labeled fake/genuine products.

This approach provides a quick, rule-based initial screening to identify high-risk products before deeper analysis.