A social structure comprised of individual and organizational groups, a social network is used to express social structure with different interests. Various studies have shown that social network capacity can be used at many individual and social levels to identify and solve issues, establish social relationships, manage organizational affairs, create public policy, and guide people on the path to achieving goals. Social networks play a key role in business successes and career advancements. In fact, networks present opportunities to organizations in order to collect data, establish healthy competition and even compromise with each other to regulate prices and policies. Social network monitoring has emerged as a crucial and important technique and a popular subject for thinking and study in areas of sociology, anthropology, social communication sciences, organizational studies, modern economics and biology, and health. Up to this point, we have mentioned the importance of social networks in society, and it is completely clear that social networks will play a more decisive role in human society in the future. Meanwhile, practical and targeted policymaking of social networks requires study, monitoring, and analysis. Monitoring is a technique that can improve the effectiveness of the social network. One of the most applicable areas is control and monitoring of community mental and clinical health. Therefore, the health problems of society can be analyzed by monitoring the number of communications, and policymaking can be done by taking appropriate measures. The present study focuses on the use of the social network in the health area and monitors the number of connections between diabetic and hypertensive patients and patients with hyperlipidemia by using a new approach. Notably, access to accurate information about people is sometimes impossible, but there is access to classified information. In addition, the large dimension of these networks complicates and causes errors in the estimation of a model’s parameters. To solve this issue, attempts are made in the current research to monitor the number of connections between social networks and categorical variables at large dimensions by using control charts. First, data obtained from the social network are converted into a contingency table, followed by monitoring the table through WALD and SST statistics. The change point is estimated by the maximum likelihood estimation (MLE) method provided that the foregoing statistics send an alert signal about getting out of control. Ultimately, the statistics are compared in terms of their performance on real data obtained from the health network.

Social Network Monitoring, Contingency Tables, Control Charts, Change Point