GARS: Real-time system for identification, assessment and control of cyber grooming attacks

Abstract In this paper, the Grooming Attack Recognition System (GARS) is presented. The main objectives of GARS are the real-time identification, assessment and control of cyber grooming attacks in favor of child protection. The system utilizes the processes of document classification, personality recognition, user history and exposure time recording to calculate specific risks children are exposed to during chat conversations. The above processes are repeated after each new message and three of them feed corresponding fuzzy logic controllers that provide particular but homogenized risk values as outputs. The weighted sum of the particular risk values results in a total value that indicates the current cyber grooming risk the child is exposed to, as the conversation evolves. Depending on predefined thresholds, the total risk value can be used to trigger alarms for various scopes (children, parents, etc). The practical use of GARS is demonstrated with a case study based on real grooming dialogs. Furthermore, an evaluation of the proposed approach through the discussion of applicability and performance results is discussed.

[1]  Yuichi Kitamura,et al.  Empirical Likelihood Methods in Econometrics: Theory and Practice , 2006 .

[2]  Lilian Mitrou,et al.  Smartphone sensor data as digital evidence , 2013, Comput. Secur..

[3]  Lilian Mitrou,et al.  Smartphone Forensics: A Proactive Investigation Scheme for Evidence Acquisition , 2012, SEC.

[4]  Kam-Fai Wong,et al.  Interpreting TF-IDF term weights as making relevance decisions , 2008, TOIS.

[5]  Ioannis Mavridis,et al.  A method to calculate social networking hazard probability in definite time , 2013, Inf. Manag. Comput. Secur..

[6]  R. O'Connell,et al.  A typology of cybersexploitation and online grooming practices , 2003 .

[7]  Ophir Frieder,et al.  Repeatable evaluation of search services in dynamic environments , 2007, TOIS.

[8]  Frank Schmalleger,et al.  Crimes of the Internet , 2008 .

[9]  Irina Rish,et al.  An empirical study of the naive Bayes classifier , 2001 .

[10]  Dimitris Gritzalis,et al.  An Insider Threat Prediction Model , 2010, TrustBus.

[11]  R. Wells.  Applied Coding and Information Theory for Engineers , 1998 .

[12]  Ioannis Mavridis,et al.  Utilizing document classification for grooming attack recognition , 2011, 2011 IEEE Symposium on Computers and Communications (ISCC).

[13]  Marilyn A. Walker,et al.  Using Linguistic Cues for the Automatic Recognition of Personality in Conversation and Text , 2007, J. Artif. Intell. Res..

[14]  Wei Wu,et al.  Evaluation of normalization methods for cDNA microarray data by k-NN classification , 2005, BMC Bioinformatics.

[15]  L. Olson,et al.  Entrapping the Innocent: Toward a Theory of Child Sexual Predators’ Luring Communication , 2007 .

[16]  B. Schneider,et al.  Bullying and the Big Five , 2003 .

[17]  I. I. Hirschman,et al.  Extreme Eigen Values of Toeplitz Operators , 1977 .

[18]  Michio Sugeno,et al.  Fuzzy identification of systems and its applications to modeling and control , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[19]  Meng Joo Er,et al.  Automatic Generation of Fuzzy Inference Systems Using Unsupervised Learning , 2005, Proceedings of the 2005 IEEE International Symposium on, Mediterrean Conference on Control and Automation Intelligent Control, 2005..

[20]  Chuen-Chien Lee,et al.  Fuzzy logic in control systems: fuzzy logic controller. II , 1990, IEEE Trans. Syst. Man Cybern..

[21]  April Kontostathis,et al.  Text Mining and Cybercrime , 2010 .

[22]  Christos Douligeris,et al.  S-Port: Collaborative security management of Port Information systems , 2013, IISA 2013.

[23]  G. Bonanno,et al.  Revictimization and Self-Harm in Females Who Experienced Childhood Sexual Abuse , 2003, Journal of interpersonal violence.

[24]  New York Dover,et al.  ON THE CONVERGENCE PROPERTIES OF THE EM ALGORITHM , 1983 .

[25]  Ioannis Mavridis,et al.  Artemis: Protection from Sexual Exploitation Attacks via SMS , 2012, 2012 16th Panhellenic Conference on Informatics.

[26]  Con Stough,et al.  The big 5 dimensional personality approach to understanding sex offenders , 2001 .

[27]  Athanasios Papoulis,et al.  Probability, Random Variables and Stochastic Processes , 1965 .

[28]  Lilian Mitrou,et al.  Which side are you on? A new Panopticon vs. privacy , 2013, 2013 International Conference on Security and Cryptography (SECRYPT).

[29]  Ronald E. Riggio,et al.  Personality and deception ability , 1988 .