Analysis of Big Data

Description of Course Goals and Curriculum

The amount of data in our world has been exploding, and analyzing large data sets is becoming a central problem in our society. This course introduces the statistical principles and computational tools for analyzing big data: the process of exploring and predicting large datasets to find hidden patterns and gain deeper understanding, and of communicating the obtained results for maximal impact. Topics include massively parallel data management and data processing, model selection and regularization, statistical modeling and inference, scalable computational algorithms, descriptive and predictive analysis, and exploratory analysis. This course introduces the statistical principles and computational tools for analyzing big data. The focus is on developing a rigorous theoretical framework of statistics and machine learning, as they are used to study the exploding amount of data in our world. This is the best statistics class at Princeton and is taught so that students can grasp the bigger picture and how topics connect to each other. 

Learning From Classroom Instruction

The lectures were the most important learning tool in this course. Professor Liu is clear and brilliant and explains the topics amazingly well. He often presents maps of concepts, making sure the students understand how everything connects to each other and how the tools are derived. It is key to take good notes in this course, as the lecture notes he puts online are not as good as the lectures themselves. 

Learning For and From Assignments

The homework builds on the lecture and are key in developing a deeper understanding. The final is the only exam and is seen as a learning opportunity to make sure that students truly understand the key concepts.

External Resources

What Students Should Know About This Course For Purposes Of Course Selection

This is a challenging course, but one in which the challenge is worthwhile. Further, the grading distribution is generous with approximately 70% of students getting A-range grades.