Noise increases in case the data is monitored in the real world and not only in a laboratory-like environment. This noise forms part of the output of the sensor. Any follow-up (pre-)processing step will have to cope with the noise before a clean signal is available and data mining can take place. Prof. Dr. Soria Morillo showed with the help of an example, how noise can be removed with the help of filters. In a hands-on part, the students downloaded the raw data, defined step by step a useful filter and later they visualized in each step the qualitative increase of the raw signal data. Each student could see how filtering and reduction of input parameters will lead to a simpler signal also removing redundancy. As a result, the following step of data mining can work faster even still maintain nearly 90% of accuracy at the input.