Today with the development of e-commerce websites, focusing on user experience as well as accessing products to users is extremely important. E-commerce websites are popular for the purpose of bringing the right products to users or helping users have a better experience. If you advertise products to the right users and the right products they like, the more likely the items are to be purchased, they will stay on the website longer. When they stay on the website longer, they will see more ads and generate more profit from advertising. A product recommendation system using Machine Learning will help users find the products they are looking for. With the above reasons, I decided to build an e-commerce website using a product recommendation system by Machine Learning
- Register, log in, logout
- Search function, filter results
- See product list, product details
- Compare, favorite products
- Add to cart, Order, Payment
- Customer information
- Product suggestions for customers
- Review, comment, answer product comments
- Chatbot Support Customer Counseling
- Admin Management
- HTML
- CSS
- Javascript
- PHP
User-based collaborative filtering algorithm (User-based) in collaborative filtering, also known as user-based neighborhood approach, is the most commonly used method in collaborative filtering and Follow these two steps:
-
Similarity calculation between active users and neighboring users. Euclidean Distance Similarity is a measure of the similarity between two users based on the distance between two points in Euclidean space. Where d(x,y) is the similarity value between user x and user y; xi is user x's rating value for product i; yi is user y's rating value for product i;
-
Select a subset of neighbor users (neighborhoods) similar to active users, then predict using that based on ratings of neighbor users
