Font Size: a A A

Research And Implementation Of Social Engineering-oriented Social Bot

Posted on:2021-05-14Degree:MasterType:Thesis
Country:ChinaCandidate:J M LiuFull Text:PDF
GTID:2427330632962685Subject:Computer technology
Abstract/Summary:PDF Full Text Request
Social engineering attacks are through natural,social and institutional ways,using human instincts,curiosity,trust,greed and other psychological weaknesses to establish a trust relationship with the attack target,thereby inducing or deceiving the attack target to perform specific operations,Attack methods to obtain sensitive information or defraud property.With the development of the Internet,the Internet has become the most important way for social engineering attacks.In recent years,the incidence of social engineering attacks on the Internet has posed a huge threat to the security of national and personal information and property.At present,the existing social engineering attack detection technology has the problems of high false alarm rate and untimely detection,while the existing social engineering defense technology research focuses on methods of phishing attacks,and the defense effect against other forms of attacks is not good.In order to solve this problem,this paper proposes a method based on honeypot,which uses social bot to simulate human activities on social networks,exposes the vulnerability of social engineering,and applies it to social engineering defense.The main work and innovations of this paper are as follows:(1)A method of applying social robots to active defense against social engineering attacks is proposed.This method describes the objects simulated by social robots through attributes,and quantifies people's active activities on social networks through behavior and speech characteristics.An algorithm is provided to allow social robots to act according to the characteristics of human behaviors and speech to achieve behavior and speech Two aspects simulate the active activities of human social networks.(2)A method of mapping from human attributes to active activity features using a support vector machine model.This method collects actual user data from a social network to train a model to achieve prediction of active activity characteristics.A dialogue system is adopted to implement a dialogue-based social engineering maintenance method based on scenarios.The dialogue system is trained by pre-setting dialogue corpora of typical scenarios of social engineering to achieve interaction with social engineering attackers.(3)Design and implement a social robotics-oriented social robot system.This system is based on a distributed microservices architecture and accesses Weibo and email.Attackers interact.
Keywords/Search Tags:social engineering, social bot, honeypot, dialogue system, personality traits
PDF Full Text Request
Related items