Text Classification

# Text classification pipeline illustration

def classify_topic(text):
    """Simple keyword-based topic classifier."""
    text_lower = text.lower()
    
    topics = {
        "Technology": ["ai", "software", "computer", "algorithm", "data", "programming", "machine learning"],
        "Sports": ["football", "basketball", "cricket", "match", "player", "team", "score", "tournament"],
        "Finance": ["stock", "market", "investment", "price", "economic", "revenue", "profit", "bank"],
        "Health": ["disease", "treatment", "doctor", "patient", "medicine", "hospital", "health", "study"],
    }
    
    scores = {}
    for topic, keywords in topics.items():
        scores[topic] = sum(1 for kw in keywords if kw in text_lower)
    
    best_topic = max(scores, key=scores.get)
    best_score = scores[best_topic]
    
    if best_score == 0:
        return "Uncategorized", 0
    return best_topic, best_score

# Test on different texts
articles = [
    "The new machine learning model from Google achieves record accuracy on image data",
    "The cricket team won the tournament after a fantastic last-over match yesterday",
    "Stock markets fell sharply after the economic report showed higher inflation than expected",
    "Researchers published a study on the effectiveness of the new treatment for the disease",
]

print("Text Classification Results:")
print()
for article in articles:
    topic, score = classify_topic(article)
    print(f"  '{article[:60]}...'")
    print(f"  Topic: {topic} (matched {score} keyword(s))")
    print()