The research aims to check the hypothesis and identify signs of labor resources in the periphery of the digital economy using a machine learning model. The research is based on the analysis of a vacancy from the HeadHunter recruitment platform in terms of the requirements stated by the employer for the applicant and the identification of signs of labor resources using a machine learning model. The main methods in the study are general scientific methods, semantic text analysis methods, machine learning methods, principal component methods, statistical methods, and others. Based on the proposed machine learning model, the authors defined the characteristics of labor resources in the periphery of the digital economy in the context of aggregated categories of economic activity. The profiles of the required employees are individual for each type of economic activity due to the specific features of the activities of enterprises in the periphery of the digital economy. The authors identified factors that influence the decision-making process in the periphery of the digital economy. The authors emphasize that the signs of labor resources on the periphery of the digital economy can be differentiated depending on the industry of the enterprise.