faq_system.py

# CodeBasics FAQ System
# Smart FAQ retrieval using TF-IDF and cosine similarity

import pandas as pd
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
import numpy as np

class CodeBasicsFAQ:
    def __init__(self, csv_path='codebasics_faqs.csv'):
        """Initialize FAQ system from CSV file"""
        # Load FAQ data
        encodings = ['utf-8', 'latin-1', 'iso-8859-1', 'cp1252']
        df = None
        
        for encoding in encodings:
            try:
                df = pd.read_csv(csv_path, encoding=encoding)
                print(f"✅ Loaded {len(df)} FAQs")
                break
            except:
                continue
        
        if df is None:
            raise Exception("Could not load FAQ CSV")
        
        self.questions = df['prompt'].tolist()
        self.answers = df['response'].tolist()
        
        # Create TF-IDF vectorizer
        self.vectorizer = TfidfVectorizer(
            lowercase=True,
            stop_words='english',
            ngram_range=(1, 2)
        )
        
        # Fit on all questions
        self.question_vectors = self.vectorizer.fit_transform(self.questions)
        print(f"✅ FAQ System ready!")
    
    def find_best_match(self, query, threshold=0.2):
        """Find best matching FAQ"""
        query_vector = self.vectorizer.transform([query])
        similarities = cosine_similarity(query_vector, self.question_vectors)[0]
        
        best_idx = np.argmax(similarities)
        best_score = similarities[best_idx]
        
        if best_score >= threshold:
            return {
                'question': self.questions[best_idx],
                'answer': self.answers[best_idx],
                'confidence': best_score
            }
        return None
    
    def answer(self, query):
        """Get answer for a query"""
        result = self.find_best_match(query)
        
        if result:
            return {
                'status': 'success',
                'confidence': f"{result['confidence']*100:.1f}%",
                'matched_question': result['question'],
                'answer': result['answer']
            }
        else:
            return {
                'status': 'no_match',
                'message': 'No matching FAQ found. Try rephrasing your question.'
            }
    
    def search_keyword(self, keyword):
        """Search FAQs by keyword"""
        keyword_lower = keyword.lower()
        matches = []
        
        for i, q in enumerate(self.questions):
            if keyword_lower in q.lower() or keyword_lower in self.answers[i].lower():
                matches.append({
                    'question': q,
                    'answer': self.answers[i]
                })
        
        return matches
    
    def list_all_questions(self):
        """Return all FAQ questions"""
        return self.questions


# ============================================================================
# USAGE EXAMPLE
# ============================================================================

if __name__ == "__main__":
    # Initialize
    faq = CodeBasicsFAQ('codebasics_faqs.csv')
    
    # Example questions
    test_questions = [
        "Can I take this bootcamp without programming experience?",
        "Why should I trust Codebasics?",
        "What are the prerequisites?",
        "Do I need a laptop?"
    ]
    
    print("\n" + "="*70)
    print("TESTING FAQ SYSTEM")
    print("="*70 + "\n")
    
    for question in test_questions:
        print(f"❓ {question}")
        result = faq.answer(question)
        
        if result['status'] == 'success':
            print(f"✅ Match: {result['confidence']}")
            print(f"📝 Q: {result['matched_question']}")
            print(f"💡 A: {result['answer'][:100]}...\n")
        else:
            print(f"❌ {result['message']}\n")
    
    # Interactive mode
    print("\n" + "="*70)
    print("INTERACTIVE MODE")
    print("="*70)
    print("Type 'quit' to exit\n")
    
    while True:
        user_q = input("❓ Your question: ").strip()
        
        if user_q.lower() in ['quit', 'exit', 'q']:
            print("👋 Goodbye!")
            break
        
        if not user_q:
            continue
        
        result = faq.answer(user_q)
        
        if result['status'] == 'success':
            print(f"\n[Confidence: {result['confidence']}]")
            print(f"\n📌 {result['matched_question']}")
            print(f"\n✨ {result['answer']}\n")
        else:
            print(f"\n❌ {result['message']}\n")