Figure 1.
Framework of Graph Convolution Text Categorization Algorithm Based on LDA-THG
Figure 2.
The structure of LDA-THG
Figure 3.
The process diagram of topic and word distribution obtained by LDA model
TABLE I.
Statistical table of data sets used in experiments
| Dataset | Total_docs | Training_docs | Testing_docs | Total_words | Categories |
|---|---|---|---|---|---|
| 20NewsGroup | 18,846 | 13,192 | 5,654 | 42,739 | 20 |
| AGNews | 8,000 | 5,600 | 2,400 | 32,653 | 4 |
| CH_News | 1,080 | 756 | 324 | 13,324 | 5 |
TABLE II.
Comparison of experimental results
| DataSets | 20newsgroup | AGNews | Ch_News |
|---|---|---|---|
| Model | |||
| SVM+LDA | 0.6827 | 0.7241 | 0.6956 |
| CNN | 0.7160 | 0.8059 | 0.6893 |
| Text-GCN | 0.8634 | 0.9245 | 0.7402 |
| LDA-THG+GCN | 0.8796 | 0.9310 | 0.7594 |
Figure 4.
Influence of topic number on classification results of LDA-THG+GCN model on three data sets
Figure 5.
Influence of W on classification results of LDA-THG+GCN model on Ch_News data set