Dr. Ebrima N. CeesayEbrima N. Ceesay, MS, Ph.D. is Associate at Booz Allen Hamilton where he is consulting on information security, forensics, data mining, and machine learning.
Ebrima coauthored Using Type Qualifiers to Analyze Untrusted Integers and Detecting Security Flaws in C Programs, A Taxonomy for Comparing Attack-Graph Approaches, Waveform Development, and An Artificial Neural Networks Approach in Detecting Phishing Attacks.
Some of the projects he has worked on include:
- Authorship Identification Forensics: He studied unsupervised learning techniques to identify authorship of Phishing emails based on email structures and linguistic patterns found in Phishing emails.
- Kernel Feature Extraction: He studied kernel methods; Kernel Principal Component Analysis (KPCA); Kernel Linear Discriminant Analysis (KLDA); and Kernel Maximum Margin Discriminant Analysis (KKMDA) to perform online feature extraction on a Phishing repository.
- Diversity Algorithm for Worrisome Software and Networks (DAWSON): He studied how to break the vulnerability specification for the executing component code or protocol that an attacker is exploiting without breaking the functionality of the executing component or protocol. A high level abstraction of defense-in-depth.