Multimodal Emotion Recognition Algorithm for Artificial Intelligence Information System

Emotional expression is an essential and important link in human social activities daily communication activities. Emotion recognition a technology that uses computer to recognize the emotion of recognized objects. When was born, research on mainly focused single-modal recognition. Single-modal has been most commonly used method. However, recent years, with continuous development science improvement people’s awareness defects recognition, multimodal increasingly emerging. This paper aimed at analyzing studying algorithm artificial intelligence information system. Based this topic, designs comparison simulation experiment for four emotional characteristics happiness, sadness, shock, anger extracted from systems. The concluded comprehensive accuracy rate system reached 93%. Compared single-mode improved by 7%.