In the theoretical part, compile an overview of current methods used for searching for personal data in unstructured texts in the Czech language and evaluate their performance. Then, present a summary of anonymization methods both for full-text and tabular data.

In the practical part, implement software for automatic identification and anonymization of personal data based on the FUTURE project requirements (https://irtis.muni.cz/future), in which the data of Android device users are collected. Furthermore, create a dataset of testing data that corresponds to the real data. Design the categories for the personal data classification, which are going to be used for tagging, and evaluate the performance of the implemented anonymization software on the labeled dataset.