使用三种不同方法的推荐系统使用相关性、使用KNN和协同过滤的简单推荐。_Jupyter Notebook_下载.zip

# Recommendation System Unsupervised Learning Project ### Three Section: <ul> <li>Section A: Simple Recommendation System using python</li> <li>Section B: Recommendation System using KNN (K-Nearest Neighbours)</li> <li>Section C: Collaborative Filtering Recommendation model</li> </ul> ## Section A: Simple Recommendation System using python We will develop basic recommendation systems using Python and pandas. In this notebook, we will focus on providing a basic recommendation system by suggesting items that are most similar to a particular item. We can then use corrwith() method to get correlations between two pandas series. If we sort the dataframe by correlation, we should get the most similar movies. Code: <a href="https://github.com/shivam1808/Recommendation-System/blob/master/Recommender%20Systems%20with%20Python.ipynb">Recommender Systems with Python.ipynb</a> ## Section B: Recommendation System using KNN (K-Nearest Neighbours) Code: <a href="https://github.com/shivam1808/Recommendation-System/blob/master/Product%20Recommender%20System.ipynb">Product Recommender System.ipynb</a> ## Section C: Collaborative Filtering Recommendation model This recommendation is however based on Collaborative filtering which uses easily captured user behaviour data. The rating data is represented using a matrix where users are along the rows and products are along the columns. * Collaborative Filtering * Memory based collaborative filtering * User-Item Filtering * Item-Item Filtering * Model based collaborative filtering * Single Value Decomposition(SVD) * SVD++ Code: <a href="https://github.com/shivam1808/Recommendation-System/blob/master/Collaborative%20Filtering%20recommender%20model.ipynb">Collaborative Filtering recommender model.ipynb</a>