Data Mining | University of Utah CS 5955/6955

Explore efficient algorithms and models for finding patterns in data sets. Covers topics like similarity search, clustering, and link analysis. Suitable for students with basic programming and probability knowledge.

Introduction

Data mining is the study of efficiently finding structures and patterns in data sets. This class will take a two-pronged approach to the topic, understanding the model and then exploring efficient algorithms to find them. It may differ greatly from many data mining classes offered elsewhere, focusing more on how to use and provide explanations (but often not proofs) of correctness.

Highlights

• Covers topics such as similarity search, clustering, regression/dimensionality reduction, link analysis (PageRank), and small space summaries
• May also discuss anomaly detection, compressed sensing, and pattern matching
• Utilizes randomized algorithms that are often extremely simple to use, but more difficult to analyze
• Provides access to all lectures through online videos

Recommendation

This course is suitable for students who are comfortable with basic probability, basic big-O analysis, and simple programming. It is recommended for undergraduates who have taken CS 3505 and CS 2100, and it is also highly recommended for those who have taken CS 3130. The class has attracted a diverse group of students, including undergraduates, masters, and PhD students from various backgrounds, and most have been able to keep up with the material, though it can be challenging.