The bachelor thesis topic is the design and implementation of a solution enabling the processing and subsequent analysis of large files (tens of billions of records) of structured but incorrectly stored data. In the first part of the thesis, the student will design a tool that will allow interactive classification of these large data sets (in CSV, SQL, and JSON formats in particular), and will allow both the creation of their descriptive structure (item names, metadata, etc.) and, in particular, the iterative correction of frequently occurring errors in them (different delimiters, different number of items, different encodings, etc.). In the second part of the thesis, the student will propose and design a suitable infrastructure into which the corrected and properly described data will be indexed -- he will focus mainly on the ElasticSearch document database, for which he will create automated installation scripts for practical deployment. Moreover, he will also prepare a solution for indexing the data (or parts of it) into SQL databases and exporting them into the corrected CSV formats. When testing the data indexation into the deployed document database, he will also present several indicators of performance tests, especially in terms of indexing performance and data lookups according to different search criteria.