STAT243: Data Mining using Machine Learning
1
1. Module Coordinator and Lecturer:
Prof Faustin Habyarimana: [email protected]
•OFFICE : F18
2. Module Administrator:
• Andile Shangase: S h a n g a s e A @ u k z n . a c . z a
• OFFICE: F14.
2
3. Lecture Timetable
Type Days Periods Times Venue
Tuesday 6 12h20-13h05 MSB PHYS L1
Lectures:
Thursday
2-3 08h40-10h20
MSB PHYS L1
Friday
5 07:45-08:30 MSBPHYL1
Practicals/
Monday 8-11 14h10-16h45 G24
• Tutorials
Students must attend at least 80% of all tutorials and practicals
4. Pre-requisite: STAT140
5. Aim:
• To introduce machine learning and data mining techniques
that will be needed to discover structure inside unstructured
data, extract meaning from noisy data, discover patterns,
correlations, and ultimately predict customer/market
behaviour and industrial as well as natural processes.
• This module will provide hands on experience in dealing with
data and using various software.
4
6. Course Content
Chapter 1: Introduction to Data Mining
1) Why Data Mining?
2) What is Data Mining?
3) Data Mining Steps
4) What is/is not Data Mining?
5) Data Mining tasks
6) What is Machine Learning?
7) Types of Machine Learning
Chapter 2: Getting to know your data
1) Data Objects and Attribute Types
2) Basic Statistical Descriptions of Data
3) Data Visualization
4) Measuring Data Similarity and Dissimilarity
5
Chapter 3: Data Preprocessing
1) Data Preprocessing: An Overview
2) Data Cleaning
3) Data Integration
4) Data Reduction
5) Data Transformation and Data Discretization
Chapter 4: Classification – Basic concepts and methods
1) What is Classification and General Approach to Classification
2) Decision Trees
3) Bayes Classification Methods
4) Rule-Based Classification
5) Model Evaluation and Selection
6) Techniques to Improve Classification Accuracy
6
Chapter 5: Classification – Advanced Methods
1) Bayesian Networks
2) Logistic Regression
3) Support Vector Machine
4) Neural network
5) K-Nearest Neighbour (KNN)
6) Ensemble Methods
7
7. Moodle
https://learn2025.ukzn.ac.za/ : Log onto this site daily. The
following will be made available:
❖ Course Material (lecture slides and all extra examples)
❖ Announcements
❖ Instructions for the practicals
❖ Questions for the tutorials
❖ Solutions for all practicals, tutorials, and tests
8
8. Course Material
❖ There is no prescribed text book but a list of recommended
books will be provided.
❖ All lecture slides/materials will be made available on Moodle.
❖ A scientific calculator, with statistical functions, is required.
9
9. Practicals
❖ The practicals will consist of a mixture of tutorial questions that
will be based on hand calculations and practical questions to
be answered using computer software.
11
10. Class Tests/Assessments
❖ The class tests (and exam)will consist of theory and
practicals.
❖ Test1: Thursday 13th March (Chapters 1-3, 35% weight)
❖ TEST 2: Monday 5th May (Chapters 4 to 5.3, 45% weight)
❖ Excel Practical: Monday 14th or Thursday 24th April (20%
weight).
❖ Only for students with valid reasons for missing tests
11. Aegrotat Test ❖ Nature: Oral or written
❖ Date: will be written soon after the original test.
12
12. DP Requirement
❖ A class mark of at least 35%
❖ The class mark is made up of the tests.
❖ At least 80% attendance.
13. Final Mark
❖ 35% of two test and practical/assignment
and 65% of exam.
14. Class Rep
A student can either volunteer or be selected by their peers.
Issues of a general nature should be raised through the class
representative.
Role of a class rep:
❖ A middle person between lecturer and students
❖ Looks great on your CV
❖ Must be someone who attends lectures regularly
17
15. Copying and Plagarism
Read the UKZN Policy on Copying and Plagiarism
COMMUNICATION PROTOCOL FOR ADDRESSING STUDENT GRIEVANCES RELATING TO ACADEMIC MATTERS
The issue affects an
individual student.
NB: Proactive is preferable. Course
Not resolved
Lecturer
Not resolved
Not resolved
Not resolved
The issue affects a group
or class Module
Class Rep
NB: Proactive is preferable. coordinator
Academic
leader
(Discipline)
Academic leader
College DVC (Teaching &
(To be avoided as much Learning)
as possible through Referral College Dean
efficiency.) Necessary referral and consultation Dean & HoS
17. Special Needs
If you have any form of disability or special educational need which
requires extra learning support, please see UKZN Disability Support
Unit as soon as you register so that extra support can be arranged as
necessary.
http://studentservices.ukzn.ac.za/DisabilitySupportUnit.aspx
19
18. Excel Self Assessment
To help us gauge your level of understanding, take the
quick self assessment survey - Link can be found on Moodle
https://forms.gle/DtLT6wAraeLRza2N8
20
