In this Project, we will explore the topic of ECommerce Recommendation Systems using Machine Learning, specifically focusing on personalized fashion recommendations based on previous purchases. We will be using a dataset from the H&M Personalized Fashion Recommendations competition on Kaggle. Throughout the project, we will cover various goals such as data collection, data analysis, and creating visual reports. We will evaluate different classification models including Decision Trees, Random Forests, and Linear Regression, as well as deep learning models such as CNN and RNN. Based on the outcomes, we will design and implement one or more deep learning systems, experimenting with different algorithms to maximize their learning capability. The performance of these systems will be evaluated, and the findings will be thoroughly documented. One crucial aspect of this project is the careful consideration and justification of cost functions. These functions play a vital role in optimizing the recommendation system. Our problem statement revolves around creating personalized product recommendations for each customer. We aim to recommend seven products to each customer using the provided datasets. To achieve this, we will group the characteristics of customers who have previously purchased products from the store. By analyzing these characteristics, we can predict the inclination of customers in that group towards certain products. We will continue this approach by classifying our products based on the characteristics of customers who buy them and making estimations for each customer. To make these estimations, we will identify which group the customer belongs to based on their past purchases and recommend the most frequently purchased products from that group.