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

Fig 1.

The architecture of adaptive high-order transformer network.

More »

Fig 1 Expand

Table 1.

Comparison on the IEMOCAP dataset.

More »

Table 1 Expand

Table 2.

Comparison on the CMU-MOSEI dataset.

More »

Table 2 Expand

Fig 2.

Ablation study.

More »

Fig 2 Expand

Fig 3.

The influence of number of atoms and dimensions on the IEMOCAP dataset.

More »

Fig 3 Expand

Fig 4.

The influence of number of atoms and dimensions on the CMU-MOSEI dataset.

More »

Fig 4 Expand

Table 3.

The analysis of model efficiency.

More »

Table 3 Expand

Fig 5.

The confusion matrix of the IEMOCAP dataset.

More »

Fig 5 Expand

Fig 6.

t-SNE visualization on the CMU-MOSEI dataset.

P, WP, HP, WN, N, HN, and NT represent Positive, Weakly Positive, Highly Positive, Weakly Negative, Negative, Highly Negative, and Neutral, respectively.

More »

Fig 6 Expand