Unsupervised analysis provides descriptive models of great utility in the process of knowledge acquisition and system modeling. Mastery of non-predictive learning techniques—pattern discovery, variable generation (feature engineering), observation partitioning (clustering), anomaly analysis, or data transformations—is, therefore, essential for these purposes. The main objective of the Data Science and Unsupervised Analysis course is to introduce these unsupervised tasks to different data structures. In addition to the most representative and promising techniques, particular emphasis will be placed on the evaluation of the results obtained.
This course is part of the Data Science for Engineers Specialisation Programme as an optional course, and this edition will be taught remotely via Zoom.
Objectives:
- Perform data partitioning (clustering)
- Discover the most important patterns (pattern mining and biclustering)
- Identify anomalies (anomaly detection)
- Perform data transformations (including Principal Component Analysis)
- Evaluate the discovered information
Target audience:
This course is intended for all engineering graduates, especially those interested in working with large volumes of data, deepening critical analysis, and addressing problems whose solutions are not easily mapped using supervised learning.
Course information (2026 Edition):
Start Date: January 10, 2026
End Date: February 14, 2026
Schedule: Saturdays, 9:30 AM to 4:00 PM
Price: €1,080
Format: Remote (live sessions via Zoom)
Duration: 25 hours
