app.py

from flask import Flask, render_template, request, jsonify
import json
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity

from utils.text_preprocessing import clean_text

app = Flask(__name__)

# ------------------ Load FAQ Data ------------------
with open("data/faqs.json", "r", encoding="utf-8") as file:
    faq_data = json.load(file)

questions = [clean_text(faq["question"]) for faq in faq_data]
answers = [faq["answer"] for faq in faq_data]

# ------------------ Vectorization ------------------
vectorizer = TfidfVectorizer(
    ngram_range=(1, 2),
    stop_words="english"
)

question_vectors = vectorizer.fit_transform(questions)

# ------------------ Routes ------------------
@app.route("/")
def home():
    return render_template("index.html")


@app.route("/get", methods=["POST"])
def chatbot_response():
    user_message = request.form.get("msg", "")
    if not user_message:
        return jsonify({"reply": "Please enter a question."})

    user_input = clean_text(user_message)
    user_vector = vectorizer.transform([user_input])

    similarity_scores = cosine_similarity(user_vector, question_vectors)
    best_match_index = similarity_scores.argmax()
    confidence_score = similarity_scores[0][best_match_index]

    if confidence_score >= 0.35:
        response = answers[best_match_index]
    else:
        response = "Sorry, I couldn't understand that. Please try asking differently."

    return jsonify({
    "reply": response,
    "confidence": round(float(confidence_score), 2)
    })



# ------------------ Main ------------------
if __name__ == "__main__":
    app.run(host="0.0.0.0", port=5000)