Fake News Detection On Fnc 1
Métriques
Per-class Accuracy (Agree)
Per-class Accuracy (Disagree)
Per-class Accuracy (Discuss)
Per-class Accuracy (Unrelated)
Weighted Accuracy
Résultats
Résultats de performance de divers modèles sur ce benchmark
Nom du modèle | Per-class Accuracy (Agree) | Per-class Accuracy (Disagree) | Per-class Accuracy (Discuss) | Per-class Accuracy (Unrelated) | Weighted Accuracy | Paper Title | Repository |
|---|---|---|---|---|---|---|---|
| Sepúlveda-Torres R., Vicente M., Saquete E., Lloret E., Palomar M. (2021) | 75.03 | 63.41 | 85.97 | 99.36 | 90.73 | Exploring Summarization to Enhance Headline Stance Detection | |
| Baseline based on skip-thought embeddings (Bhatt et al., 2017) | 31.80 | 0.00 | 81.20 | 91.18 | 76.18 | On the Benefit of Combining Neural, Statistical and External Features for Fake News Identification | - |
| Neural baseline based on bi-directional LSTMs (Bhatt et al., 2017) | 38.04 | 4.59 | 58.132 | 78.27 | 63.11 | On the Benefit of Combining Neural, Statistical and External Features for Fake News Identification | - |
| Baseline based on word2vec + hand-crafted features (Bhatt et al., 2017) | 50.70 | 9.61 | 53.38 | 96.05 | 72.78 | On the Benefit of Combining Neural, Statistical and External Features for Fake News Identification | - |
| Neural method from Mohtarami et al. + TF-IDF (Mohtarami et al., 2018) | - | - | - | - | 81.23 | Automatic Stance Detection Using End-to-End Memory Networks | - |
| Neural method from Mohtarami et al. (Mohtarami et al., 2018) | - | - | - | - | 78.97 | Automatic Stance Detection Using End-to-End Memory Networks | - |
| Bi-LSTM (max-pooling, attention) | 51.34 | 10.33 | 81.52 | 96.74 | 82.23 | Combining Similarity Features and Deep Representation Learning for Stance Detection in the Context of Checking Fake News | - |
| ZAINAB A. JAWAD, AHMED J. OBAID (CNN and DNN with SCM, 2022) | 88.47 | 96.00 | 87.70 | 95.04 | 84.60 | Combination Of Convolution Neural Networks And Deep Neural Networks For Fake News Detection | - |
| 3rd place at FNC-1 - Team UCL Machine Reading (Riedel et al., 2017) | 44.04 | 6.60 | 81.38 | 97.90 | 81.72 | A simple but tough-to-beat baseline for the Fake News Challenge stance detection task | - |
| Bhatt et al. | 43.82 | 6.31 | 85.68 | 98.04 | 83.08 | On the Benefit of Combining Neural, Statistical and External Features for Fake News Identification | - |
0 of 10 row(s) selected.