Detection of the too rapid decline of cognitive functions in time is pivotal, because the decline can be slowed much better in the early stadiums, than in the later ones. Prevention is also of key importance, for example, doing mental exercises is efficient at reducing the chance of dementia. However, tests that doctors make, take lots of time, so the vast majority of the risk group cannot be examined. The M3W (Maintaining and Measuring Mental Wellness) project was started in order to solve these problems, which focused on prevention and detection as well.
Logic games were created on a website, where users can play them. The goal is to be able to detect the too rapid cognitive decline from the results and other characteristic data of the games.
It is also important to find other useful information in the data. For data analysis, clustering also should be used among many methods.
The goal of the cluster analysis is to group the data such way, that data in the same group are more similar to each other, than those in other groups. In the first part of my thesis I wrote about clustering in general, described several clustering algorithms, and showed their advantages, disadvantages. The algorithms were run on smaller data sets, and the results were visualized. In the second part of my thesis the results and data of the M3W games were clustered, and the clusters created by the algorithms were analyzed.