The work utilizes evolution computation techniques to induce association rules based on example data stored in big datasets. The main focus is especially on genetic algorithms, which represent a generic population-based metaheuristic optimization algorithm that uses solution space search mechanisms inspired by biologic evolution, such as recombination, mutation and evolutionary selection. The goal is to describe the process of designing and implementation of own genetic algorithm that will mine the association rules. This includes the definition of solution representation, their evaluation and specification of whole evolutionary cycle. The work composes of 5 main chapters. In the first chapters we specify the overall topic of association rules as a part of data mining and knowledge discovery. Later on we focus on evolutionary algorithms and current trends of their usage in data mining. We describe theoretical principles, methodology and techniques for mining association rules and outline common problems and challenges that are related to this topic. This work also includes snippets of implemented code in C#, which demonstrates the actual implementation of proposed algorithm.