The certification of the Compact Muon Solenoid (CMS) particle detector data, as useable for physics analysis, is a crucial task to ensure the quality of all physics results published by CERN. Currently, the certification conducted by human experts is labor intensive and can only be segmented on a long period of time basis. The goal of the thesis is to automate data certification process of the CMS at the Large Hadron Collider (LHC). The student should first find the right representation of the data and develop a new analyzer on top of existing CMSSW framework that would produce a new dataset for the problem. The most suitable machine learning model will be developed to address all the characteristics of the problem, namely sparsity of anomalies, high dimensionality of the data and changing configuration of both CMS and LHC. Finally, it will be evaluated and compared to the previous data certification process.