Social bots are intelligent programs that control the behavior of fake accounts in an online social network(OSN). They pass themselves off as human accounts and manipulate the health of the ecosystem of OSNs. Thus, it is crucial to distinguish social bots from human accounts. Despite recent advances in social bot detection, state-of-the-art methods face challenges in scalability, generalization, and robustness. Considering these drawbacks, in this article, a new category of features, called friendship preference, is proposed. Friendship preference features are extracted from the profile attributes of the followers. The proposed feature extraction formula is designed to be scalable as the number of followers grows. Two classifiers are developed to evaluate the proposed category of features in terms of efficiency and usefulness. The classifiers are benchmarked on several real datasets. The experimental results indicate that the classifiers outperform Botometer in terms of classification performance. Scalability is assessed by inspecting the detection performance of classifiers when the attributes of a limited number of the accounts in the neighborhood of the users with large egonets are available. Furthermore, generalization is validated by crossover validation on various datasets. Finally, robustness and the ability for early detection of social bots are discussed.

Socialbots are intelligent software that controls all behaviour of fake accounts in an online social network. Since they are armed with detection evasion techniques, it is valuable to be able to determine the effectiveness of these techniques. In this study, an analytical model is developed to estimate a lower bound for the cost of automatic establishment of a socialbot network. Moreover, by considering fake accounts purchasing as an establishment strategy, an upper bound is suggested for acceptable costs. These two boundaries are compared to decide on the economic feasibility of a socialbot network design strategy. To demonstrate the practicality and effectiveness of the model, two case studies are investigated. They show that although designing a fully stealthy socialbot network is economically feasible, the infiltration time would be unacceptable. Thus, this ideal situation in which the establishment is fully stealthy, performs in a tolerable time, and satisfactory infiltration scale, is impractical. A possible solution could be achieved by reducing the time and cost in exchange for less stealthy behaviour while the infiltration scale kept unchanged. Since the model presents a trade‐off between stealthiness, time, and cost, it is a useful tool facilitating the design of a possible strategy.