In this paper, we present our solutions to the two sub-challenges of Affective Behavior Analysis in the wild (ABAW) 2023: the Emotional Reaction Intensity (ERI) Estimation Challenge and Expression (Expr) Classification Challenge. ABAW 2023 focuses on the problem of affective behavior analysis in the wild, with the goal of creating machines and robots that have the ability to understand human feelings, emotions and behaviors, which can effectively contribute to the advent of a more intelligent future. In our work, we use different models and tools for the Hume-Reaction dataset to extract features of various aspects, such as audio features, video features, etc. By analyzing, combining, and studying these multimodal features, we effectively improve the accuracy of the model for multimodal sentiment prediction. For the Emotional Reaction Intensity (ERI) Estimation Challenge, our method shows excellent results with a Pearson coefficient on the validation dataset, exceeding the baseline method by 84 percent.