CENG 643

Pattern Discovery in Data Mining

This course introduces students to different aspects of pattern discovery (frequent pattern mining or pattern mining or association rule mining) which is the earliest and most popular task of data mining where the objective is to find hidden frequent patterns in massive data. Frequent patterns are important in many domains like market-basket analysis, software bug mining, web click analysis, phrase mining, image processing, recommender systems, etc. The course will cover the key aspects of the field in terms of algorithms, variations, scalability, data types and applications. The topics include frequent patterns, pattern usage in compression, patterns in dynamic data (streams, sequences and spatiotemporal objects), big data frequent pattern mining, frequent patterns in classification and clustering and privacy in pattern mining.

C.C. Aggarwal and J. Han. Frequent Pattern Mining, Springer, 2014 ,J. Han, M. Kamber, J. Pei. Data Mining Concepts and Techniques, Morgan-Kaufmann-Series-in-Data-Management-Systems, 2012 ,P-N. Tan, M. Steinbach, A. Karpatne, V. Kumar, Introduction to Data Mining (2nd Edition), Pearson, 2019

Learning Outcomes

1. Learn state-of-the-art research and industry trends in pattern mining,

2. Understand the fundamental principles of the field,

3. Be able to make design decisions in deploying pattern mining tasks in data centric applications as well as to identify the bottlenecks of such applications,

4. Learn how to read-understand-present-discuss research articles of the field.

Grading

Quizzes 20%

Assignments 20%

Class Participation 20%

Final Exam 40%