Abstract: The article is devoted to the problem of ensuring the quality of results in information processing systems, where some operations are performed with the involvement of people, interaction with whom is carried out via the Internet. Such systems are widely used in solving various tasks, but the involvement of a person in information processing tasks is associated with a set of fundamental limitations inherent in a person: low speed of information processing, the need for motivation, the possibility of errors or purposeful distortion of information. Thus, the development of methods and tools for managing the quality of results obtained with the help of such systems is an urgent task. The article proposes a model of data aggregation to improve the quality of results obtained using large-scale human-machine computing. The application of the model is considered by the example of solving the problem of marking and searching for images obtained as part of mass athletics events (runs). The assessment of the effect of aggregation is carried out on the basis of simulation modeling. The results of the study of the proposed approach have shown that integration is especially effective in conditions of poor-quality markup. However, even in conditions of high-quality markup, the use of aggregation allows you to increase the completeness of search results. In general, it can be concluded that the use of data aggregation in the processing of human-machine computing results is a promising approach, and the use of probabilistic graphical models for aggregation allows you to smoothly increase the accuracy of the results of the system with an increase in the amount of available information.