Many labor-intensive problems can be automatized by modeling the behavior of the computer system manually. Nevertheless, for many problems this is not possible because the system has to adapt to feedback on actions or the problem size is simply too large.
Machine learning (also known as data mining, pattern recognition, data analysis or classification) is an area of research at the edge of computer science, artificial intelligence, mathematics and statistics which addresses these problems. It covers general methods and techniques which can be applied on problems like customer behavior prediction, controlling of robotics, spam recognition or Protein folding prediction.
In this lecture we study simple learning problems starting with regression and classification tasks (supervised learning). We study different models like linear models, decision trees, neural networks and Support Vector Machines. Later we study cluster analysis and dimension reduction (unsupervised learning).
Textbooks:
- Richard O. Duda, Peter E. Hart, David G. Stork (2001): Pattern Classification, 2nd edition.
- Christopher M. Bishop (2006): Pattern Recognition and Machine Learning.
- Trevor Hastie, Robert Tibshirani, Jerome Friedman (2001): The Elements of Statistical Learning.
- Brian D. Ripley (1996): Pattern Recognition and Neural Networks.
- Tom Mitchell (1997): Machine Learning.
- Ian Witten, Eibe Frank (2005): Data Mining: Practical Machine Learning Tools and Techniques, Second Edition.
Lecturer: Prof. Dr. Dr. Lars Schmidt-Thieme
Tutorial: Martin Wistuba
