This document outlines the syllabus for a Data Mining course offered in the fall semester of 2016 at National Taiwan University of Science and Technology. The course objectives are to learn the basics of data mining methods, code existing algorithms, and apply data mining to solve real problems. Topics include classification, estimation, clustering, and association rule mining. Students will complete homework assignments involving computer programs for classification and clustering using benchmark data. Their work will be evaluated based on homework, a final exam, and a final team project applying data mining techniques to a real-world problem.
This document outlines the syllabus for a Data Mining course offered in the fall semester of 2016 at National Taiwan University of Science and Technology. The course objectives are to learn the basics of data mining methods, code existing algorithms, and apply data mining to solve real problems. Topics include classification, estimation, clustering, and association rule mining. Students will complete homework assignments involving computer programs for classification and clustering using benchmark data. Their work will be evaluated based on homework, a final exam, and a final team project applying data mining techniques to a real-world problem.
This document outlines the syllabus for a Data Mining course offered in the fall semester of 2016 at National Taiwan University of Science and Technology. The course objectives are to learn the basics of data mining methods, code existing algorithms, and apply data mining to solve real problems. Topics include classification, estimation, clustering, and association rule mining. Students will complete homework assignments involving computer programs for classification and clustering using benchmark data. Their work will be evaluated based on homework, a final exam, and a final team project applying data mining techniques to a real-world problem.
National Taiwan University of Science and Technology
Department of Industrial Management
Syllabus of Fall Semester, 2016 2016/09/22 Course Data Mining Required/ Credits 3 Optional Title Applications Optional Distinguished Professor Graduate rjkuo@mail.ntust.edu.tw Instructor IM E-mail Ren-Jieh Kuo Program Course objectives: 1. Learn the basics of data mining methods. 2. Be able to code the existing data mining algorithms. 3. Be able to apply data mining to solve real problems. Course introduction: This course intends to introduce some data mining techniques which include classification, estimation, clustering, and association rules mining. It also provides some application examples for these techniques. Textbook: Tan, P.-N., Steinbach, M., and Kumar, V., Introduction to Data Mining, Pearson Education, Inc., 2014. References: Han, J. and Kamber, M., Data Mining: Concepts and Techniques, 2nd Edition, Morgan Kaufmann, 2006. Prerequisites: Statistics, Management, and coding capability Schedule: Week 1 (09/15): Holiday Week 10 (11/17): Clustering I Week 2 (09/22): Course introduction Week 11 (11/24): Clustering II Week 3 (09/29): Intro. to data mining Week 12 (12/01): Association mining (SOM) Week 4 (10/06): Data I Week 13 (12/08): Homework 2 demo Week 5 (10/13): Data II Week 14 (12/15): Exploring data Week 6 (10/20): Classification I Week 15 (12/22): Homework 3 demo (Weka) Week 7 (10/27): Classification II Week 16 (12/29): Final exam Week 8 (11/03): Classification III Week 17 (01/05): Final project presentation I Week 9 (11/10): Homework 1 demo (ANN) Week 18 (01/12): Final project presentation II Grading: Homework (two computer programs including classification (10%), clustering (10%) and Weka application (20%)) 40% Use benchmark data to validate the programs Demo results in class Final exam 30% Final team project 30% Write a survey report for the selected problem (at least 3 journal papers from 20062016) Solve a real-world problem using data mining techniques Present in class (Every team member has to present in class) Suggestions for students: Read related academic journals. Try to code the algorithms and apply them to the real data.
Here are the association rules generated from frequent itemset {BCE} that satisfy the minimum confidence threshold of 75%:{BC} → {E} {CE} → {B}So the number of association rules generated is 2