Table 8. Text embeddings on Cora for link prediction (micro-F1, metric lies between (0,1) and higher value means better results).
| % Train edges | 5% | 10% | 30% | 50% |
|---|---|---|---|---|
| BoW | 0.69 ± 0.01 | 0.71 ± 0.00 | 0.75 ± 0.01 | 0.76 ± 0.00 |
| TF-IDF | 0.67 ± 0.01 | 0.69 ± 0.01 | 0.72 ± 0.01 | 0.74 ± 0.01 |
| LDA | 0.68 ± 0.01 | 0.69 ± 0.01 | 0.71 ± 0.01 | 071 ± 0.01 |
| SBERT pretrained | 0.69 ± 0.00 | 0.71 ± 0.00 | 0.74 ± 0.01 | 0.76 ± 0.01 |
| Word2Vec pretrained | 0.60 ± 0.02 | 0.62 ± 0.00 | 0.63 ± 0.00 | 0.64 ± 0.01 |
| Word2Vec (d = 300) | 0.68 ± 0.00 | 0.70 ± 0.00 | 0.72 ± 0.00 | 0.73 ± 0.01 |
| Word2Vec (d = 64) | 0.70 ± 0.00 | 0.70 ± 0.00 | 0.72 ± 0.01 | 0.73 ± 0.01 |
| Doc2Vec pretrained | 0.63 ± 0.02 | 0.66 ± 0.00 | 0.70 ± 0.00 | 0.70 ± 0.00 |
| Doc2Vec (d = 300) | 0.67 ± 0.01 | 0.70 ± 0.00 | 0.73± 0.00 | 0.74 ± 0.00 |
| Doc2Vec (d = 64) | 0.66 ± 0.01 | 0.68 ± 0.00 | 0.69 ± 0.00 | 0.69 ± 0.00 |
| Sent2Vec pretrained | 0.66 ± 0.01 | 0.69 ± 0.00 | 0.73 ± 0.00 | 0.75 ± 0.00 |
| Sent2Vec (d = 600) | 0.71 ± 0.00 | 0.72 ± 0.01 | 0.75 ± 0.00 | 0.76 ± 0.01 |
| Sent2Vec (d = 64) | 0.70 ± 0.01 | 0.71 ± 0.00 | 0.73 ± 0.00 | 0.74 ± 0.00 |
| Ernie pretrained | 0.56 ± 0.01 | 0.58 ± 0.01 | 0.62 ± 0.01 | 0.63 ± 0.01 |
Note:
The best values with respect to confidence intervals are highlighted in bold.