Sandareka Wickramanayake
Scholar

Sandareka Wickramanayake

Google Scholar ID: 6mlKqLgAAAAJ
Senior Lecturer, University of Moratuwa
Explainable AIDeep LearningMachine LearningArtificial Intelligence
Citations & Impact
All-time
Citations
267
 
H-index
7
 
i10-index
7
 
Publications
20
 
Co-authors
15
list available
Resume (English only)
Academic Achievements
  • Published 'Explainable Artificial Intelligence for Enhanced Living Environments: A Study on User Perspective' in Advances in Computer, Vol. 134, 2023.
  • Published 'DeFiTrust: A transformer-based framework for scam DeFi token detection using event logs and sentiment analysis' in Expert Systems with Applications (ESWA), 2024.
  • Published 'Spatio-Temporal Dual-Attention Transformer for Time-Series Behavioral Biometrics' in IEEE Transactions on Biometrics, Behavior, and Identity Science (T-BIOM), 2024.
  • Published 'BUGSPHP: A dataset for Automated Program Repair in PHP' in MSE, 2024.
  • Published 'TEZARNet: TEmporal Zero-Shot Activity Recognition Network' at ICoNIP, 2023.
Background
  • Currently a Senior Lecturer in the Department of Computer Science and Engineering at the University of Moratuwa, Sri Lanka.
  • Research interests focus on Interpretable Machine Learning, particularly designing accurate and interpretable Deep Neural Networks (DNNs).
  • Aims to explain the actual reasons behind DNN decisions in ways easily understandable to humans.
  • Works on both post-hoc explanation methods for trained DNNs and inherently interpretable DNN architectures.
  • Believes that improved understanding and performance of AI systems will increase public trust and real-world adoption.
  • Also develops state-of-the-art DNN-based models for real-world applications including human activity recognition, automatic program repair, biometric authentication, scam detection, and sign language recognition.