Recommendation Systems

Course Description

This course will give you a broad knowledge of what a recommendation system is, what type of problems it can solve, the main algorithms used and how to evaluate such systems. We will start with basic methods for making recommendations when using and predicting ratings (user-based and item-based collaborative filtering), to then move to more state-of-the-art techniques used by companies such as Netflix and Spotify (matrix factorization).

Course topics

Recommendation Systems, Collaborative Filtering, Matrix Factorization.

Course tools

Python

Prerequisites

Linear Algebra, Python

Lecturer

Juan Pablo Figueroa Senior Data Scientist at N-iX, Ukraine Juan Pablo is a data scientist at N-iX, where he works in projects related to recommendation systems, predictive modeling and time series forecasting. He has over 5 years of experience building predictive models for different industries, including Ad Tech, Healthcare and Retail, among others. Juan Pablo has a background in Statistics and holds a MSc in Machine Learning from University College London, and his most recent area of focus is Automated Machine Learning (AutoML). Fields of interests: Machine Learning, Applied Predictive Modeling, Recommendation Systems, Basketball Analytics. Contacts: jpfigueroa@n-ix.com linkedin.com/in/juan-pablo-figueroa-74003482/