Introduction to Data Mining (3 ECTS)

Learning outcomes

Knowledge and understanding:
The students will know some basic concepts and some algorithms of data mining. They will understand what data exploration means, the difference between supervised and unsupervised learning and will become familiar with algorithms such as K-means clustering, decision trees, naïve Bayes and (feed forward) neural networks. Further, they will know how to evaluate the models produced by those algorithms. They will be able to run and evaluate those algorithms with the tool RapidMiner.

Skills:
The students are able to understand basic machine learning algorithms especially k-means clustering, decision trees, naïve Bayes and (feed forward) neural networks and make use of the tool RapidMiner to run and evaluate these algorithms on datasets.

Content

- Data, data exploration and data visualization.
- Supervised and unsupervised algorithms.
- Distance, K-means clustering, properties and limits.
- Decision trees, Gini-index and entropy.
- Confusion matrix, accuracy, precision, recall, ROC curves.
- Naïve Bayes.
- Formal neuron, multi-layer perceptron network and feed-forward neural network.
- Back-propagation algorithm.

Prerequisites

Basic mathematics, for example, linear algebra, vectors, probabilities and so on.

Further information for students

Bring your own laptop.

Evaluation scale

0-5

Assessment criteria, approved/failed

All exercises will have to be made, and a small report with answers, explanation and appropriate screenshots of the RapidMiner process should be written and handed in for each assignment. Approval for all assignments by the teacher will be necessary to obtain the 3 ECTS.