Peer-reviewed research in artificial intelligence, robotics, and machine learning — published in IEEE and Springer conferences.
This paper presents a deep learning approach to obstacle avoidance in mobile robots using RGB-D images, semantic segmentation, and neural networks — enabling robots to navigate complex environments in real time without manual programming. The system uses depth information combined with visual recognition to identify and avoid obstacles dynamically.
This paper introduces a high-performance hybrid deep learning model for detecting political fake news, combining multiple neural network architectures to achieve superior detection accuracy. The hybrid model outperforms single-architecture approaches by leveraging complementary strengths of different deep learning methods for natural language understanding and classification.
I am open to research discussions, academic collaborations, and speaking opportunities.