onlincourse

Recommender Systems in Python

Learn what recommender systems are, why they’ve become so popular, and how AI could help you implement your own, with this online course from National Tsing Hua University.

Duration

6 weeks

Weekly study

3 hours

100% online

How it works

Unlimited subscription

Learn more

Established

1956

Location

Taiwan

World ranking

Source: QS World University Rankings 2021

Build a recommender system with National Tsing Hua University

If you’ve ever watched a recommended film on Netflix or listened to a suggested playlist on Spotify, you have used a recommender system.

On this six-week course from National Tsing Hua University, you’ll learn why so many platforms incorporate recommender systems, and how you can use Python to build your own.

Learn what recommender systems are and why so many platforms are using them

Recommender systems use complex data sets and machine learning to bring you tailored recommendations for your consumption.

The course will start with an introduction to the concept and influence of recommender systems, reviewing some of the most popular models and explaining why they have become so popular among big tech platforms.

Explore different approaches to building a recommender system

Once you’ve understood the concept and influence of recommender systems, you’ll get stuck in analysing different approaches to building them.

In Weeks 2, 3, and 4 of the course, you’ll learn how to build a recommender system in Python, using each of a variety of different approaches.

Discover the role of AI in developing recommender systems

The last three weeks of the course will explore the role AI and machine learning play in developing and enhancing recommender systems.

You’ll learn how algorithmic data can be used to make more sophisticated recommendations.

By the end of the course, you’ll have the expertise and programming skills you need to start building your first recommender system.

Week 1
Recommender systems and their applications
- Introduction to Recommender Systems
  Define a recommender system and identify why we need it.
- Recommendation Approaches
  Identify different recommendation approaches.
- Recommender Implementation and Evaluation
  Identify steps of building a recommender. Recognize how to evaluate a recommender.
- Python Practice
  Install Python development environment and run Python programs.
- Datasets
  Explore what a dataset is and why it is important to build a recommender system.
- Lecture Notes and Source Code
  Download lecture notes and source code and explore them.
Week 2
Fundamental Recommenders
- Data Collection
  Collect the data that we use to build a recommender. Look into the details of the data items.
- Data Organization and Metrics
  Organize and prepare data for a recommender. Identify and design metrics for recommenders.
- A Recommender based on Certain Metrics
  Build a recommender based on a certain metric.
- A Recommender based on User’s Preferences
  Build a recommender based on user’s preferences.
- A Recommender based on Similarities
  Build a recommender based on similarities.
Week 3
Content-based Recommender
- Content-based Filtering
  Explore content-based filtering. Explore the dataset used to illustrate the content-based filtering.
- A Content-based Recommender
  Explore the dataset again for a recommender. Design metrics for a recommender.
- TF-IDF for a Recommender
  Calculate TF-IDF for a recommender.
- A Content-based Recommender using TF-IDF
  Calculate cosine similarity for a recommender. Build a content-based recommender using TF-IDF.
Week 4
Collaborative Filtering Recommender
- Collaborative Filtering
  Explore collaborative filtering
- A User-Based CF Recommender
  Build a user-based collaborative filtering recommender
- An Item-based CF Recommender
  Build a item-based collaborative filtering (CF) recommender
- Matrix Factorization
  Explore matrix factorization and its role in collaborative filtering (CF) recommenders.
- A Model-Based CF Recommender
  Build a model-based collaborative filtering recommender
Week 5
Artificial Intelligence (AI) and Machine Learning (ML)
- AI, Machine and Deep Learning
  Explore AI, machine learning, and deep learning.
- Machine Learning: Regression
  Use linear regression for prediction tasks.
- Machine Learning: K-Means
  Use K-means to cluster data points.
- Machine Learning: K-Nearest Neighbors (KNN)
  Use K-Nearest Neighbors (KNN) to classify data points.
- Deep Learning
  Build a deep learning application.
Week 6
Machine Learning Recommender
- Recommenders using Machine Learning
  Explore recommenders using machine learning.
- A Recommender using Linear Regression
  Build a recommender using linear regression.
- A Recommender using K-means
  Build a recommender using K-means.
- A Recommender using K-Nearest Neighbors (KNN)
  Build a recommender using K-Nearest Neighbors (KNN)
- A Recommender using Deep Learning
  Build a recommender using Neural Networks.