Deep Learning

Designing Protective Measures against Malware Attacks
  • Devised a gradient based, greedy attack strategies to modify malicious applications to fool detection systems
  • Designed a novel defensive strategy to protect detection systems from malware attacks
  • Validated the strategies for twelve classification algorithms against 50,000 android applications
Designing Protective Measures against Malware Attacks
Detection of Malaria from Segmented Human Cell Images
  • Built detection models using convolutional neural networks in Pytorch implementing a custom architecture and transfer learning from state-of-art models
Profiling Fake News: Learning the Semantics and Characterisation of Misinformation
  • Proposed Fake News detection model using extracted syntactic, semantic, temporal features
  • Evaluated the hypothesis against four benchmark datasets of varying article lengths
Supervised Contrastive Learning for Interpretable Long Document Comparison