ELE 3909 Cluster Analysis for Business
ELE 3909 Cluster Analysis for Business
Clustering methods help to design strategies based on customer segments rather than the entire population, in order to better meet customer expectations. We can, for example, design marketing campaigns, avoid customer leakage or assess credit risk more accurately by analyzing clusters in a given population and designing specific strategies for each cluster. Clustering methods are also useful for revealing which similar characteristics subgroups of a population share.
This course presents different clustering techniques for revealing subgroups within a larger population, as well as methods for analyzing and understanding what characteristics distinguish each group. In addition, the course introduces the idea of dynamic clustering, or time-dependent clustering, in which we learn to track movements between clusters over time.
The focus of the course is practical and analytical, preparing students for jobs as data analysts/scientists or for a master's degree in Data Science for Business or Business Analytics.
In this course the students will:
- Learn the difference between supervised and unsupervised learning
- Learn how to prepare and/or transform a data set for cluster analysis
- Get a thorough introduction to different clustering algorithms
- Learn how to select the optimal number of clusters
- Learn to analyze the clustering results
- Understand the differences between clustering and dynamic clustering
After the completion of this course, students will be able to:
- Prepare data sets, including data representations, for cluster analysis
- Apply different cluster techniques for a given task
- Choose the number of clusters using an appropriate metric
- Use statistical methods to analyse the results
Students should be able to understand the principles of unsupervised learning and how clustering methods can be applied in different applications. A successful student will be able to select a clustering algorithm, carry out an analysis that reveal hidden patterns in the data, or a representation of it, and bring added-value into the application of the analysis.
- Unsupervised vs Supervised learning
- Key clustering algorithms, including: Hierarchical clustering, K-means, and density-based clustering methods
- Cluster analysis on different feature spaces
- Statistical methods to analyze the clusters
- Clustering vs Dynamic clustering
Lectures and practical exercises that must be solved with a computer using Python. The practical exercises are shared via GitHub.
Please note that while attendance is not compulsory in all courses, it is the student’s own responsibility to obtain any information provided in class.
Higher Education Entrance Qualification
Disclaimer
Deviations in teaching and exams may occur if external conditions or unforeseen events call for this.
EBA 1180 Mathematics for Data Science or EBA 2904 Statistics with programming.
Students who do not have EBA 1180 Mathematics for Data Science or EBA 2904 Statistics with programming, can compensate the requirement with any other similar course in mathematics or statistics.
| Assessments |
|---|
Exam category: Submission Form of assessment: Written submission Weight: 40 Grouping: Group (2 - 3) Duration: 1 Week(s) Comment: Written home exam. Work in groups of up to three. Requires Python. All exams must be passed to obtain a final grade in the course. Exam code: ELE 39091 Grading scale: ECTS Resit: Examination when next scheduled course |
Exam category: Submission Form of assessment: Written submission Invigilation Weight: 60 Grouping: Individual Support materials:
Duration: 3 Hour(s) Comment: Written school exam. All exams must be passed to obtain final grade in the course. Exam code: ELE 39092 Grading scale: ECTS Resit: Examination when next scheduled course |
All exams must be passed to get a grade in this course.
| Activity | Duration | Comment |
|---|---|---|
Teaching | 36 Hour(s) | |
Seminar groups | 9 Hour(s) | |
Student's own work with learning resources | 130 Hour(s) | |
Group work / Assignments | 25 Hour(s) |
A course of 1 ECTS credit corresponds to a workload of 26-30 hours. Therefore a course of 7,5 ECTS credit corresponds to a workload of at least 200 hours.