SEEM4630 E-Commerce Data Mining

Teacher

TA

GAO Chang	1155154491@link.cuhk.edu.hk	William M.W. Mong Engg Bldg, Rm 711
GUO Qintian	1155149734@link.cuhk.edu.hk	William M.W. Mong Engg Bldg, Rm XXX
WANG Chaojun	1155170172@link.cuhk.edu.hk	William M.W. Mong Engg Bldg, Rm 709D
XU Weiwen	1155149726@link.cuhk.edu.hk	William M.W. Mong Engg Bldg, Rm 615
YANG Sen	1155184038@link.cuhk.edu.hk	William M.W. Mong Engg Bldg, Rm 711

Annoucement

2024/01/08: Please determine the tutorial schedule here, the deadline is 23:59 Jan 11 (Thu).
2024/01/15: The tutorials will be conducted in ERB 612 every Monday 18:30-19:00.
2024/01/17: Assignment 1 has been posted. The deadline is Jan 31 (Wed) 17:00.
2024/01/30: The deadline of Assignment 1 has been extended to Feb 7 (Wed) 12:00 (noon).
2024/02/15: Assignment 2 has been posted. The deadline is Mar 05 (Tue) 12:00 Noon.
2024/02/19: Mid-term exam will be held on March 12 during lecture time and classroom. It is definitely easier than final exam. Please attend to get some marks!
2024/02/27: The scope of mid-term exam covers the lecture notes from the beginning to "model evaluation and selection". Only lecture notes are tested. No Weka will be tested. No cheat sheet is allowed. No calculators are allowed. Most mathematical formulae that are needed will be given on the question paper.
2024/02/27: We will prepare answer booklet for you for the mid-term exam. To submit mid-term answer sheets, you will need to take photo of each written page, e.g. in jpg format, using your mobile phone. After that, you will convert and combine the photos of all written pages into one single PDF file. The file conversion can be done by apps such as CamScanner or AdobeScan. Then you will submit the single PDF file to the Blackboard system. Therefore you should now install such apps in your mobile phone.
2024/03/04: The grades for Assignment 1 are now available on Blackboard. If you have any questions, please reach out via email to gaochang@se.cuhk.edu.hk.
2024/03/04: The deadline of Assignment 2 has been extended to March 8 (Fri) 12:00 Noon.
2024/03/21: Assignment 3 has been posted. The deadline is Apr 11 (Thu) 17:00.
2024/03/26: Tutorial 6 is re-scheduled to Apr 2 18:30-19:00, in ERB 612.
2024/04/08: The deadline of Assignment 3 has been extended to Apr 18 (Thu) 17:00. Please also include pre-processed new client data in your submission.
2024/04/11: Grading criteria for Assignment 3: 60% based on data mining process and report quality, 40% based on accuracy rank of new client prediction (higher-ranked submissions in the upper half will receive significantly more rewards, while lower-ranked submissions in the lower half will receive approximately 10 out of 40 points).

Course Outline and Schedule

Week	Lecture Date	Lecture Topic	Tutorial Topic
1	Jan 9, 11	introduction; data pre-processing	No tutorial
2	Jan 16, 18	data pre-processing; Weka installation	data pre-processing; Weka demo on installation
3	Jan 23, 25	decision tree learning	Weka demo on installation and data pre-processing; explain assignment 1
4	Jan 30, Feb 1	decision tree learning; model evaluation & selection	Weka demo on installation and data pre-processing; explain assignment 1
5	Feb 6, 8	model evaluation & selection	No tutorial due to Spring Festival travel
6	Feb 13, 15 Spring Festival	Spring Festival No lectures	No tutorial due to Spring Festival holiday
7	Feb 20, 22	Weka demo on assgn 1 solution, decision tree, and model evaluation and selection; lecture on model evaluation & selection	explain assignment 2
8	Feb 27, 29	practical consideration	explain assignment 2
9	Mar 5, 7 Reading Week	Reading Week No lectures	Reading Week No tutorials
10	Mar 12, 14	Mid-term exam on Mar 12 neural networks	Weka demo on practical consideration
11	Mar 19, 21	neural networks; Weka demo on neural networks	Weka demo on Assignment 2 solution
12	Mar 26, 28	neural networks; logistic regression	explain assignment 3
13	Apr 2 Apr 4 holiday	logistic regression; clustering	Since Apr 1 is a holiday, the tutorial for this week has been re-scheduled to Apr 2 (Tue) same time and venue. Weka demo on logistic regression; Explain assignment 3
14	Apr 9, 11	clustering; loan assessment case study; association rule mining	Weka demo on clustering
15	Apr 16, 18	association rule mining; Python libraries	demo on Python libraries for data science

Lecture Notes

introduction Ref: Provost's book Chapter 1 and Han's book Chapter 1
Industry Case - Walmart Data Mining Applications
Industry Case - 5 ways Amazon and Alibaba use AI and data mining
data pre-processing Ref: Han's book Chapter 2
Weka installation and data pre-processing The data set can be found below.
classification and decision tree induction Ref: Han's book Chapters 8.1 and 8.2
decision tree case study - Boston Globe
decision tree sample application - Analysis of Covid-19
model evaluation and selection Ref: Han's book Chapter 8.5 and Provost's book Chapter 5
Industry Case - Expedia Group's Customer Lifetime Value Prediction Model
revision on decision tree learning Recording of Weka demo on Assignment-1, revision lecture, and Assignment-2
practical considerations for classification learning Ref: Han's book Chapter 3.4.4, 8.2.3, and 8.2.4.
classification - neural networks Ref: Han's book Chapter 9.2
Industry Case - Real-Life and Business Applications of Neural Networks
classification - logistic regression Ref: Speech and Language Processing, Chapter 5.1-5.7
Industry Case - credit scoring project at Deloitte using logistic regression
clustering Ref: Han's book Chapters 2.4, 10.1, and 10.2.
loan assessment case study article
loan assessment case study lecture note
association rule mining Ref: Han's book Chapters 6.1 and 6.2.
case study on association rule mining article
Python libraries for data analysis and machine learning

Tutorial Notes

Tutorial 1 (Jan 15): Data Preprocessing Slide Recording
Tutorial 2 (Jan 22): Weka Demo on Data Preprocessing Slide Recording
Tutorial 4 (Feb 19): Decision Tree Slide Recording
Weka Demo (Mar 22): Neural Network Slide Recording
Tutorial 5 (Mar 25): Client Subscription Prediction Slide Recording
Tutorial 6 (Apr 02): Logistic Regression Slide Recording
Tutorial 7 (Apr 08): Weka Demo on Clustering Slide T_shape.csv arrow_shape.csv bank.csv Recording
Tutorial 8 (Apr 15): Demo on Python Libraries for Data Science Slide python_demo Recording

Assignments and Weka Data Mining Package

[Weka] Please follow the instruction on the website to install the stable version (3.8) of Weka. It provides different links to suit different OS. Please select the one you are using (if possible, download it before the lab class).
The bank marketing data set used in Weka notes: bank.csv
Assignment 1 (Due: Jan 31 (Wed) 17:00): Data Preprocessing Using Weka Specification bank-additional.csv
Assignment 2 (Due: Mar 05 (Tue) 12:00 Noon): Decision Tree Specification dataset.zip
Assignment 3 (Due: Apr 18 (Thu) 17:00): Client Subscription Prediction Specification dataset.zip

Required Readings

" Data Mining - Concepts and Techniques " , Jiawei Han, Micheline Kamber, and Jian Pei, 3rd edition, Elsevier Science, 2011. (e-book can be accessed online via CUHK library)

Final Exam

The scope of the normal final exam covers the lecture topics "model evaluation & selection", "practical considerations for classification learning", "classification - neural networks", "classification - logistic regression", and "clustering".
The scope of the make-up final exam covers ALL lecture topics.
Calculators are allowed.
You cannot bring anything except calculators. Therefore, books, notes, cheat sheets are NOT allowed.

Assessment Scheme

Assignments 50%
Mid-term Exam 20%
Final Exam 30%

Evaluation

CSV File Upload

Accuracy:

SEEM4630 E-Commerce Data Mining

Teacher

TA

Annoucement

Course Outline and Schedule

Lecture Notes

Tutorial Notes

Assignments and Weka Data Mining Package

Required Readings

Recommended Readings

Final Exam

Assessment Scheme

Evaluation