Koncepce zpracování závěrečných zpráv z výměnných pobytů s využitím nástrojů AI

This bachelor thesis examines the possibilities of using artificial intelligence tools for processing final reports written by students after exchange stays at the Prague University of Economics and Business. The aim of the thesis is to design and practically verify a procedure that transforms extensive and difficult to compare text reports into clear summary outputs that can be used by students and by the International Office. The thesis first compares the approaches of selected Czech and foreign universities to student final reports. It then identifies the requirements of students, the International Office, and the Centre for Informatics regarding the form of the outputs and the rules for working with data. It also analyses selected artificial intelligence tools and practically tests five of them on anonymized data prepared from the InSIS information system. The results of the practical testing show that general conversational AI tools proved to be suitable for this purpose. ChatGPT achieved the best results in the practical testing. The verification also showed that the proposed outputs correspond to the main requirements of the monitored groups. For students, the outputs provide clearer information about foreign universities and make the selection process easier. For the International Office, they represent a possible way to work faster with a larger number of final reports and to make better use of the data that have been collected so far. The main contribution of the thesis is the proposal of a concrete procedure for preparing anonymized inputs, selecting suitable tools, and creating two types of summary outputs that can later be extended through partial automation of the whole process.