INSIGHT EYE

  • Unique Paper ID: 170293
  • Volume: 11
  • Issue: 7
  • PageNo: 9-13
  • Abstract:
  • This project explores a machine learning approach to lie detection by analyzing eye movement patterns, including blink rate, fixation duration, pupil dilation, and saccades, as non-intrusive indicators of deception. Eye-tracking data is collected under controlled scenarios of truth and deception, creating a dataset for training models like Support Vector Machines, Decision Trees, and Neural Networks. The models are evaluated based on accuracy, precision, and F1-score, aiming to reliably distinguish truthful from deceptive behavior. The system’s interpretability is enhanced through feature importance analysis, providing insights into which eye movement metrics are most linked to deception. Additionally, the project investigates the adaptability of these models to various contexts, such as security and psychological assessments, to ensure broader applicability and robustness.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 7
  • PageNo: 9-13

INSIGHT EYE

Related Articles