Master Seminar Kognitive Systeme (WS 2011/2012)
Topic: Practical Aspects of Machine Learning (using RapidMiner)
In practice machine learning is more than algorithms classifying examples. There are many things around it:
- Data acquirement and pre-processing,
- Feature engineering,
- Model evaluation, and
- Software that performs these tasks.
In this seminary we will emphasize the practical aspects of these surroundings. Our experience will be based on RapidMiner – an open-source data mining tool. First we will learn how to use this software:
- Process design concept
- Importing, exporting, and generating data
- Basic mechanisms: loops, macros, logging, ...
Then we will center on theoretic concepts (e.g., feature generation, evaluation techniques) along with examples (PCA, bootstrap validation) and their realizations in RapidMiner. Finally we will take a look at competitive products and how to extend RapidMiner.
Possible PresentationTopics
- Feature Generation
- Disretization
- Feature Selection
- Extending RapidMiner
- Competitive products
Literature (available in the vc course)
- Witten, I. H. & Eibe, F. (2005). Data Mining. Practical Machine Learning Tools and Techniques. Elsevier. (Chapter 1)
- Han, J. & Kamber, M. (2006). Data Mining: Concepts and Techniques Elsevier. (Chapter 1)
Relevant literature for the single topics will be provided within sessions.
General Information
- You find a general course description at the corresponding pages from the WIAI module guide.
- You find administrative information at UnivIS.
- Participants should sign up for the course in the virtual campus.
- This course is open for master and advanced bachelor students.
- Prerequisites: Basic machine learning knowledge as taught in our Machine Learning course (especially the first three lectures) will be helpful.
- Presentations and theses may be given/written in German or English.