Recommender Systems with Machine Learning

Overview

This course starts with the theoretical concepts and fundamental knowledge of recommender systems, covering essential taxonomies. You'll learn to use Python to evaluate datasets based on user ratings, choices, genres, and release years. Practical approaches will help you build content-based and collaborative filtering techniques. As you progress, you'll cover necessary concepts for applied recommender systems and machine learning models, with projects included for hands-on experience. Key learnings include AI-integrated basics, taxonomy, overfitting, underfitting, bias, variance, and building content-based and item-based systems with ML and Python, including KNN-based engines. The course is suitable for beginners and those with some programming experience, aiming to advance ML skills and build customized recommender systems. No prior knowledge of recommender systems, ML, data analysis, or math is needed, only basic Python. By the end, you'll relate theories to various domains, implement ML models for real-world recommendation systems, and evaluate them.

Syllabus

Introduction

In this module, we will introduce you to the field of AI Sciences and recommender systems. You will meet the instructor, explore the course layout, understand the basics of recommender systems, and preview the exciting projects you will undertake.

Motivation for Recommender System

In this module, we will delve into the motivations behind recommender systems. You will learn about their processes, historical evolution, and the critical role AI plays. We'll also cover practical applications and the challenges faced in real-world scenarios.

Basic of Recommender Systems

In this module, we will cover the foundational aspects of recommender systems. You will study the taxonomy, data matrices, evaluation techniques, and filtering methods, equipping you with a solid understanding of how these systems function and are assessed.

Machine Learning for Recommender System

In this module, we will focus on leveraging machine learning for recommender systems. You will gain insights into data preparation, explore filtering methods, and implement machine learning algorithms like tf-idf and KNN, enhancing the recommendation process.

Project 1: Song Recommendation System Using Content-Based Filtering

In this module, we will guide you through building a song recommendation system using content-based filtering. You will work on dataset management, genre exploration, and implement advanced techniques like tf-idf and FuzzyWuzzy to create effective song recommendations.

Project 2: Movie Recommendation System Using Collaborative Filtering

In this module, we will take you through developing a movie recommendation system using collaborative filtering. You will learn to analyze user and movie data, create collaborative filters, and apply KNN to generate accurate movie recommendations, culminating the course with practical applications.