Slabchenko O. Methods, models and an information technology for reconstruction of missing values from social networks' accounts

The thesis is devoted to the solution of an actual scientific problem that consists in development of special methods and models of imputation as elements of an information technology for reconstruction of missing values in order to improve the quality of the process of imputation of data from social networks' accounts in the course of automated treatment and preprocessing of information for the further analysis. To achieve this goal the methods for construction of an extended matrix of attributes, for decreasing the number of unique values in attributes on the basis of pre-clustering were developed and the method of searching for association rules was improved. Using the suggested methods, the models of imputation on the basis of association rules, a random forest, a support vector machine, a neural network and an EM-algorithm were developed. The method of imputation on the basis of an ensemble of machine learning models, which is adapted to the analysis of data from social networks, was further developed. The information technology for reconstruction of missing values from social networks' accounts was developed.