Lecture: Data Mining II
- Introduction
- High Dimensional learning
- Stream mining
- Learning under label-scarcity
- Enseble Learning
- Lecturer: Prof. Dr. Eirini Ntoutsi
- Teaching assistants: Vasileios Iosifidis || Amir Abolfazli
- Participation is equivalent to 5 ECTS credits (1 ECTS credit equals 30 hours of study).
- We will use StudIP for discussions/announcements/material/schedule.
- Kick-off lecture: Friday 18/10/2019
- Lecture 1 (18/10/2019, 13:00-14:30). Introduction
- Lecture 2 (25/10/2019, 13:00-14:30). Feature selection
- Tutorial 1 (25/10/2019, 14:30-16:00). Feature selection (Vasilis)
- Lecture 3 & 4 (1/11/2019, 13:00-14:30 & 14:30-16:00). Dimensionality Reduction & Subspace Clustering
- Lecture 5 & 6 (8/11/2019, 13:00-14:30 & 14:30-16:00). Label scarsity & Ensemble learning
- Tutorial 2 & 3 (15/11/2019, 13:00-14:30 & 14:30-16:00). Dimensionality Reduction & Subspace Clustering (Vasilis)
- Tutorial 4 & 5 (29/11/2019, 13:00-14:30 & 14:30-16:00). Label scarcity & Ensemble learning (Vasilis)
- Lecture 7 & 8 (6/12/2019, 13:00-14:30 & 14:30-16:00). Data stream classification
- Tutorial 6 & 7 (13/12/2019, 13:00-14:30 & 14:30-16:00). Data stream classification (Amir)
- Lecture 9 & 10 (20/12/2019, 13:00-14:30 & 14:30-16:00). Data stream clustering
- Tutorial 8 & 9 (10/01/2020, 13:00-14:30 & 14:30-16:00). Data stream clustering (Amir)
- Lecture 11 (17/01/2020, 13:00-14:30). Industrial Lecture - TBA
- Student presentations 1/2 (24/01/2020, 13:00-14:30 & 14:30-16:00). Oral and poster student presentations
- Student presentations 2/2 (31/01/2020, 13:00-14:30 & 14:30-16:00). Oral and poster student presentations
About the lecture
In the Data Mining I course, basic data mining/machine learning tasks and techniques are introduced. However, the characteristics of the data generated by modern applications introduces challenges which go beyond these techniques. The focus of this course is on the challenges introduced due to the modern data characteristics and on the methods and techniques for handling modern complex data. This includes both (i) adaptation of old data mining techniques to deal with modern data challenges and also (ii) new methods and techniques that were explicitly introduced for such sort of data.
Course content
Organisation
Schedule (note that most lectures/tutorials take place in blocks)