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.

Comparative analysis of work related to multimodal sentiment analysis.

More »

Table 1 Expand

Fig 1.

The proposed multimodal sentiment analysis model.

More »

Fig 1 Expand

Fig 2.

Tourism reviews data collection and preprocessing.

More »

Fig 2 Expand

Fig 3.

The process of constructing the text graph.

More »

Fig 3 Expand

Fig 4.

The process of constructing the image graph.

More »

Fig 4 Expand

Fig 5.

The process of constructing the fusion graph [37].

More »

Fig 5 Expand

Fig 6.

The principle of the GraphSAGE algorithm.

More »

Fig 6 Expand

Table 2.

Parameter settings for BERT, ResNet, and GraphSAGE in the proposed model.

More »

Table 2 Expand

Table 3.

Results of text single-modal comparison of the proposed model with each of the six SOTA models on the Yelp dataset, TripAdvisor dataset and Ctrip dataset.

More »

Table 3 Expand

Table 4.

Results of multi-modal comparison of the proposed model with each of the six SOTA models on the Yelp dataset, TripAdvisor dataset and Ctrip dataset.

More »

Table 4 Expand

Fig 7.

The heat map of the experimental results of the model proposed in this study on the Yelp dataset.

More »

Fig 7 Expand

Table 5.

Performance comparison of the proposed model with different loss function weight settings (α, β) across four datasets (TripAdvisor, Amazon, Ctrip, MM-CPC) in text-only modality.

More »

Table 5 Expand

Table 6.

Results of comparison experiment on each of the other four datasets in multi-modal mode.

More »

Table 6 Expand

Table 7.

Computational efficiency comparison between the proposed model and TETFN in terms of training time, inference time, and memory usage.

More »

Table 7 Expand

Fig 8.

Accuracy and F1-score comparison between the proposed model and baseline methods (TFN, LMF, MulT) on the TripAdvisor and Yelp datasets across three sentiment classes.

More »

Fig 8 Expand

Fig 9.

Performance impact of removing individual components (BERT text encoder, ResNet visual encoder, or GraphSAGE aggregator) measured by relative change in accuracy (Δ Acc) and macro-F1 score (Δ F1) across all test datasets.

More »

Fig 9 Expand

Table 8.

Ablation study results comparing BERT and Word2Vec for text feature extraction (Accuracy, Precision, Recall, F1-score, AUC-ROC).

More »

Table 8 Expand

Table 9.

Ablation study results comparing ResNet and plain CNN for image feature extraction (Accuracy, Precision, Recall, F1-score, AUC-ROC).

More »

Table 9 Expand

Table 10.

Ablation study results comparing GraphSAGE and standard GCN for multimodal graph aggregation (Accuracy, Precision, Recall, F1-score, AUC-ROC).

More »

Table 10 Expand