Electives

Machine learning (ML) and data-driven methods have become essential to analyzing complex data. This course will introduce the fundamental concepts and algorithms that enable computers to uncover the underlying patterns from different types of behavioral data. Students will learn about commonly used learning techniques including dimension reduction (PCA, MDS, and UMAP), unsupervised learning algorithms (K-means, mean shift), supervised learning algorithms (k-nearest Neighbors, Naïve Bayes), time series analysis (HMMs) and neural networks (CNNs).

This class offers a hands-on approach to machine learning and data science, focusing on giving the students the basic knowledge behind these machine learning methods and the ability to utilize them with behavioral data. The students will be required to perform experimentation on datasets either by programming or using off-the-shelf packages, with the goal to understand the practical issues when applying many of the machine learning algorithms. When the students have completed the course, they should be able to understand and explain a set of machine learning algorithms, and to apply machine learning tools to their own research. Thus, the two overall goals of the class are to help students to build a conceptual framework of ML and to introduce a variety of ML methods.

Through taking this course, students are expected to achieve learning goals:

1. Understanding fundamental concepts and knowledge in machine learning and data mining
2. Mastering a wide range of machine learning algorithms
3. Learning to think in a machine learning way and translate real-world applications into machine learning problems
4. Completing data analysis projects to apply learned knowledge to real-world data