In facial expression recognition, the recognition rate will be disturbed due to inherent limitations including light, pose variations, noise, and occlusion. In this paper, a hybrid approach of facial expression based on sentiment analysis has been presented combining local and global features. Feature extraction is performed to fuse the histogram of oriented gradients (HOG) descriptor with the uniform local ternary pattern (U-LTP) descriptor. After the experimental analysis of U-LTP parameters, the most appropriate HOG parameter set is selected to improve the performance of the proposed technique for facial images containing noise points and occlusion. Features extracted by HOG and U-LTP are fused into a single feature vector, and the feature vector is sent to a multi-class support vector machine classifier for facial classification. Experiments in three public facial expression image databases show that the recognition rate of the proposed method is better than that of other facial expression recognition methods.