Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Table 1.

CMU-MOSI dataset information.

More »

Table 1 Expand

Table 2.

CMU-MOSEI dataset information.

More »

Table 2 Expand

Fig 1.

Structure of AFR-BERT multimodal sentiment analysis model.

AFR-BERT is divided into four network modules, which correspond to data input, data fusion, data analysis, and data output.

More »

Fig 1 Expand

Fig 2.

BiLSTM model structure.

(Forward) means forward propagation of the model. (Backward) means model backward propagation.

More »

Fig 2 Expand

Fig 3.

Cross-modal fusion attention mechanism structure.

(Kt) represents text feature data. (Ka) represents audio feature data. (Relu, Row Softmax, softmax, concat) are all function calculations. (Mask) is a matrix.

More »

Fig 3 Expand

Fig 4.

Scaled dot product attention structure.

(Q) means the query matrix. (K) means the key matrix. (V) means the value matrix. (Mask) represents matrix operations for processing non-fixed-length sequences. () is the scale factor for scaling.

More »

Fig 4 Expand

Table 3.

The optimal parameter settings report.

More »

Table 3 Expand

Table 4.

Comparative experiments of multimodal sentiment analysis models on the dataset CMU-MOSI.

More »

Table 4 Expand

Fig 5.

Cross-sectional histograms of Corr metrics for each model on the CMU-MOSI.

More »

Fig 5 Expand

Table 5.

Comparative experiments of multimodal sentiment analysis models on the dataset CMU-MOSEI.

More »

Table 5 Expand

Fig 6.

Cross-sectional histograms of Corr metrics for each model on the CMU-MOSEI dataset.

More »

Fig 6 Expand

Table 6.

Ablation experiments on the CMU-MOSI dataset.

More »

Table 6 Expand

Table 7.

Sample analysis.

More »

Table 7 Expand