Unsupervised machine learning focuses on uncovering hidden structures in datasets without relying on predefined labels. This repository includes exercises, portfolio assignments, and study cases designed to provide hands-on experience with these techniques.
-
Clustering: This technique identifies patterns or structures within data, allowing us to group data points into clusters based on similarities.
-
Dimensionality Reduction: This approach simplifies data by using structural characteristics to retain the most important information while reducing complexity.
- Cell Type Annotation in PBMCs Using scRNA-Seq Data
- Anomaly detection in sensor data
- Identifying Biological Substitutes for Synthetic Compounds
- Identifying Biomarkers for Pre-Eclampsia Using LC-MS Data
- Cluster text to extract insights from clinical case reports
Contact: f.feenstra@pl.hanze.nl