One of the most topical problems of information security is the prevention of successful incidents of social engineering attacks - psychological manipulation of people to get access to confidential information. It is necessary to assess personal characteristics of an information system user to estimate his vulnerability to some type of attack. Assessment the degree of openness of a social media user is perspective this way (openness is considered according to the factor of 'openness to experience', which is part of the personality model 'The Big Five'). In this paper, we consider an approach to assessing the openness of an online social network user using the bayesian network. Based on the five-factor personality questionnaire, the characteristics that can be extracted from the user's page, potentially affecting his openness to accepting information, are considered. The structure of the Bayesian network for solving this problem is proposed.