The inspiration for this diploma thesis is a stereotype that some individuals are more talented to program than others. We investigate the innate and acquired origin of abilities in general and in a more specific domain - whether is the ability of solving algorithmically based exercises, stated in literature as a necessary skill for successful learning how to program, innate or acquired by learning. The first part of the theoretical part of diploma thesis is therefore a summary and comparison of read literature on the subject.
In the second part of the theoretical part of this diploma thesis we focus on the field of data mining, especially on decision tree induction, being the model later used in the empirical part.
In the empirical part of this diploma thesis we then describe the method of acquiring needed data through designing a two-part questionnaire, one part of which was used to evaluate the successfulness of individuals at solving algorithmically based exercises. The other part was used to acquire the data about student characteristics, which we believed could be linked to individuals’ success at solving algorithmically based exercises. We also describe the process of data modeling, with the use of machine learning and data mining methods, and then discuss the findings, which indicate the connection between certain characteristics of individuals (for example: success at mathematics matura exam, the selection of matura exam subjects, the frequency of solving logical puzzles) and the success at a algorithmically based exercises solving test.