Exploration of the application of multimodal model in psychological analysis

HIGHLIGHTS

• What: Some parameters in the method (such as the sparse regularizer weights in the selection phase, the parameters of the Gaussian noise in the addition phase, etc.) have a significant impact on the model performance If the data of a certain modality has problems such as noise, missing values, or inaccurate annotations, it may affect the attention mechanism`s correct focus and integration of the modality information. Many benchmark models may overfit on the training set, but this model demonstrates better generalization ability . In this paper, the application of multimodal sentiment analysis models in mental illness is . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or E-mail:

  If you don't have an account, you can create one here.