Tools
Tools
Ayo-Akere, Nelson Seyi (2019) Towards an Effective Social Engineering susceptibility detection Model Using Machine Learning on the Online Social Network. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration manual)
Download (1MB) | Preview |
Abstract
The challenge posed by social engineering has become increasing worrisome and proven over the years to be daunting to mitigate even with recent security measures in place. Humans as it is are said to be the weakest link to security which makes online social media networks (OSN) very susceptible to social engineering attacks due to its never ending increase of users who come together for the purpose of communicating and sharing information. However, OSN users have the final say as regards dissemination of information via posts, uploads and updates on OSN consequently users become susceptible to social engineering attacks via the release of personal identifiable information (PII) on OSN. This research presents a novel social engineering machine learning prediction model (SE-MLPM) to detect and extract PII in OSN user posts using natural language processing (bag-of-words) and thereafter vectorise post text into vectors utilizing the term frequency inverse document frequency (TF-IDF) vector space modelling technique and finally classify, label and predict levels of post susceptibility to social engineering attacks in addition to revealing the PII discovered to the OSN user and recommending to the user if the post should go live or not, based on PII count recovered from the post ranging from a high susceptibility level to a no susceptibility level using the logistic regression classification algorithm. This will give the OSN users the opportunity to vet their post before disseminating to the public. By so doing, SE-MLPM will minimize the enormous volume of Sensitive Personal Information (SPI) OSN users post on OSN.
Actions (login required)
 |
View Item |