An Improved XGBoost Classifier for Micro Expression Recognition using Hybrid Optimization Algorithm

Facial Expression Recognition is an interesting and emerging research field that concentrates on the enhancement of developing. In areas like education, gaming, healthcare, and safety, technology plays a big role in various applications. Many researchers are studying facial expressions to create methods that can understand and extract them. In deep learning, CNN is very important. To make XGBOOST work better, this new method uses genetic algorithms and a mix of optimization strategies. By combining XGBoost with genetic algorithms (GAs) and particle swarm optimization (PSO), this method aims to solve this problem. This paper introduces the new PSO-GA algorithm and discusses its advantages, including how it outperforms other approaches already in use and produces reliable, cutting-edge results across a range of difficult datasets. In this paper we used the XGBoost classifier. Compared to other methodologies, CNN is able to discern emotions and facial expressions quite effectively to the hybrid PSO-GA algorithm.