Generationof Realistic Artificial Curricula Vitae

Abigail Paradise, Dvir Cohen, Rami Puzis, Asaf Shabtai

The curriculum vitae (CV) is an interesting resource, specially, because of its promising applications in research. CVs can be helpful for testing and system evaluation with data representing real-world scenarios. CVs can also assist in generating artificial honeypot profiles in social networks to detect potential cyber attackers. Currently, there are a number of challenges associated with the use of CVs, including their limited availability, questionable accuracy, and risk to user privacy. Given these challenges, a method of generating CVs that are difficult to distinguish from real CVs is needed. In this research we propose a method for the automated generation of realistic CVs. A field study was performed in order to evaluate the method’s ability to generate realistic artificial CVs that experts cannot distinguish from real CVs. The results indicate that participants were unable to differentiate between artificial and real CV, demonstrating that our method can be applied to generate realistic artificial CVs to a wide range of applications.