In the context of this bachelor thesis, the automated evaluation of application documents for the master's program Autonomous Systems at the University of Stuttgart is considered. The existing application process requires a high workload for university employees, since many of them are involved in the evaluation. Manually processing also leads to a subjective evaluation.

The goal is to develop an efficient and objective process that verifies the content requirements of applicants for the target university in order to reduce the workload for both staff and applicants and to admit qualified applicants faster. The Thesis focuses on the development of a system that verifies the content overlap between the courses specified by the applicant in an additional form and the requirements of the target university using data from a module handbook. The module handbook contains detailed information about the lectures and their contents.

The module handbook is evaluated using a system called "ESCA", which uses the metadata of the module handbook and the applicant's input on an additional form. This is important because accurate module descriptions ensure that only relevant content is credited. After the module descriptions have been segmented, a language classifier and natural language processing are used to check whether there is any overlap in content between the applicant's entries and the requirements. The user interface was designed to be intuitive and efficient.

Various metrics were used for the evaluation, for example, the Average Mean Percentage Error was calculated for the accuracy of the assessments. For the accuracy of the module handbooks it was examined how many different universities can be correctly evaluated with the parser.

We were able to correctly analyze about half of the module manuals from different universities. In the evaluation compared to the official evaluation, assuming unscanned module handbooks and despite the presence of spelling errors in the additional forms, the system achieved an accuracy of 74.5\%. The evaluation shows that positive results are observed in the majority.