Výběr vhodných AI nástrojů pro řešení vybraných databázových úloh

The diploma thesis focuses on the selection of suitable AI tools for solving selected database tasks. The aim is to analyse the capabilities of free versions of AI tools, propose a procedure for selecting a suitable AI tool for solving selected database tasks, and verify this procedure. Based on a review of literature and online sources, ten tasks are specified within a case study of a film database inspired by the ČSFD portal. The tasks include requirements specification, data modelling, SQL script generation, implementation of database logic, SQL query optimization, SQL dialect migration, and technical database documentation. For the verification, a reference solution, evaluation criteria, checklists, and standardized prompts are prepared. For outputs that can be verified in a database environment, control scripts are also prepared. The outputs of AI tools are evaluated using a knock-out criterion, point-based scoring, and conversion into banded evaluation. The verification results made it possible to determine a recommended AI tool for most database tasks and, for some of them, also other suitable tools. At the same time, the degree of output usability and the extent of required human intervention differ. Based on the results, the tasks are divided into the categories AI solves independently, AI helps, and AI is insufficient. The highest usability was found in script generation, implementation of database logic, SQL query optimization, and technical database documentation. By contrast, the conceptual database schema and physical data model required substantial reworking. The contribution of the thesis is a practical comparison of selected AI tools, recommendations of specific tools for individual database tasks, and materials enabling the verification to be repeated using the same procedure.