Li-ion batteries are the most wide-spread electrochemical energy storage systems. Cathode materials are a major focus because they are considered the limiting element of overall battery performance. Li-rich layered oxides are among the actively investigated cathode materials designated for using in Li-ion batteries of new generation and approaching to the imposed high requirements of high energy and power densities. However, the inherent complexity of their structure resulting in the well-known problems with the structural stability has hindered the widespread use of Li-rich layered oxides as the cathode materials. In this study, we have used machine learning (ML) methods to obtain the models able to predict the electrochemical and structural characteristics of these materials. The data from XPS and XRD spectra were used as the additional descriptors to evaluate the target functional characteristics: initial discharge capacity, coulombic efficiency and capacity fade. One of the capacities of ML to impute the missing data were exploited to solve the problem of incomplete experimental data. The obtained models have shown the reasonable predictive performance. The analysis of the experimental data and the results of modeling have feed to further considerations concerning the complexity of the multifold processes that determine the target functional characteristics of these materials.