Extrakce pravidel ze znalostních grafů

This bachelor thesis deals with the rule mining from databases, with special focus on knowledge graphs. Knowledge graphs store data by creating entities and relations between them which allows for storing a large amount of data in a complex, connected way. When implementing knowledge graphs, the RDF format is frequently used. Data in databases can be mined for rules which can be useful namely for data abstraction or for finding new associations. With many tools generally used for rule mining, the ones described in this work are: AMIE and its extension RDFRules, WARMR, SWARM, RuDik and Aleph. In the practical part of this work RDFRules is used for KG-COVID-19 knowledge graph analysis. This graph contains data about the covid-19 disease caused by SARS-CoV-2 virus. This disease caused a pandemic in 2020. Analysis of the graph contains three tasks. All of them focus on the ACE2 receptor which is used by SARS-CoV-2 to enter a host cell. Output of the first task are rules referring to ACE2 and a validation of the results in literature. In the second task potential options for repurposing of ACE1 drugs for ACE2 are explored. Output of this task are five drugs with two of them being more closely analyzed. Last task compares the rules found for ACE2 and Neuropilin. Neuropilin is another receptor that could possibly serve as an entry for the virus into a host cell. Output of this task is the comparison of proteins found by the analysis that interact with both receptors.