recommendation engine 75% polished

recommondation-engine/app.py

@@ -4,6 +4,7 @@ import mysql.connector
 from sklearn.metrics.pairwise import cosine_similarity
 import numpy as np
 import logging
+from unittest import result
 
 def database():
     db_connection = mysql.connector.connect(
@@ -83,27 +84,57 @@ def get_user_history(user_id):
     return final
 
 
-def get_recommendations(user_id, top_n=40):
+def get_recommendations(user_id, top_n=10):
     try:
         # Get all products and user history with their category vectors
         all_products = get_all_products()
         user_history = get_user_history(user_id)
-
         # if not user_history:
         #     # Cold start: return popular products
         #     return get_popular_products(top_n)
-
         # Calculate similarity between all products and user history
         user_profile = np.mean(user_history, axis=0)  # Average user preferences
         similarities = cosine_similarity([user_profile], all_products)
-
         # finds the indices of the top N products that have the highest
         # cosine similarity with the user's profile and sorted from most similar to least similar.
         product_indices = similarities[0].argsort()[-top_n:][::-1]
         print("product", product_indices)
 
+        # Get the recommended product IDs
+        recommended_products = [all_products[i][0] for i in product_indices]  # Product IDs
+
+        # Upload the recommendations to the database
+        history_upload(user_id, product_indices)  # Pass the indices directly to history_upload
+
         # Return recommended product IDs
-        return [all_products[i][0] for i in product_indices]  # Product IDs
+        return recommended_products
     except Exception as e:
         logging.error(f"Recommendation error for user {user_id}: {str(e)}")
         # return get_popular_products(top_n)  # Fallback to popular products
+
+def history_upload(userID, anrr):
+    db_con = database()
+    cursor = db_con.cursor()
+
+    try:
+        for item in anrr:
+            #Product ID starts form index 1
+            item_value = item + 1
+            print(item_value)
+            # Use parameterized queries to prevent SQL injection
+            cursor.execute(f"INSERT INTO Recommendation (UserID, RecommendedProductID) VALUES ({userID}, {item_value});")
+
+        # Commit the changes
+        db_con.commit()
+
+        # If you need results, you'd typically fetch them after a SELECT query
+        # results = cursor.fetchall()
+        # print(results)
+
+    except Exception as e:
+        print(f"Error: {e}")
+        db_con.rollback()
+    finally:
+        # Close the cursor and connection
+        cursor.close()
+        db_con.close()

recommondation-engine/server.py

@@ -1,6 +1,8 @@
 from flask import Flask, request, jsonify
 from flask_cors import CORS
-from example1 import get_recommendations
+from app import get_recommendations
+from app import history_upload
+
 import time
 
 
@@ -18,8 +20,7 @@ def handle_session_data():
         if not user_id or not email or is_authenticated is None:
             return jsonify({'error': 'Invalid data'}), 400
 
-        print(get_recommendations(user_id))
-        time.sleep(2)
+        get_recommendations(user_id)
 
         print(f"Received session data: User ID: {user_id}, Email: {email}, Authenticated: {is_authenticated}")
         return jsonify({'message': 'Session data received successfully'})