Seoul National University
M1522.001400 Data Mining
Fall 2024 - U Kang

Course Information

Data mining refers to theories and techniques for finding useful patterns from massive amount of data. Data mining has been used in high impact applications including web analysis, recommendation system, fraud detection, cyber security, etc. This course covers important algorithms and theories for data mining. Main topics include finding similar items, mining frequent patterns, link analysis, link prediction, recommendation system, data stream mining, clustering, graph mining, time series prediction, and outlier detection.

Schedule

Date Topic Notice
Sep. 2 Introduction (Chapter 1)
4 Basics, Finding Similar Items (Chapter 3)
9 Finding Similar Items (Chapter 3)
11 Finding Similar Items (Chapter 3)
16, 18 Mining Data Streams (Chapter 4) HW1 out
23 Mining Data Streams-2 (Chapter 4)
25 Mining Data Streams-3 (Chapter 4)
30 Link Analysis (Chapter 5) HW2 out
Oct. 2 Link Analysis (Chapter 5)
7 Link Analysis-2 (Chapter 5)
9 Link Analysis-3 (Chapter 5)
14 Link Analysis-4 (Chapter 5)
16 Link Prediction HW 3 out
21 Midterm
23 Frequent Itemsets (Chapter 6)
28 Frequent Itemsets-2 (Chapter 6)
30 Clustering (Chapter 7)
Nov. 4 Clustering-2 (Chapter 7) HW 4 out
6 Advertising on the Web (Chapter 8)
11 Advertising on the Web-2 (Chapter 8)
13 Recommendation (Chapter 9)
18 Recommendation-2 (Chapter 9) HW 5 out
20 Mining Social-Network Graphs (Chapter 10)
25 Mining Social-Network Graphs-2 (Chapter 10)
27 Dimensionality Reduction (Chapter 11)
Dec. 2 Dimensionality Reduction-2 (Chapter 11) HW 6 out
4 Time Series Prediction
9 Anomaly Detection, Conclusion
11 Final

Grading

Late policy - for all deliverables:

Textbook

The text book is
Mining of Massive Datasets by Jure Leskovec, Anand Rajaraman, Jeff Ullman. (also available online)

Prerequisite

We expect you to have an undergraduate-level knowledge on the following topics: We provide some background, but the class will be fast paced.