Topic Modeling on BanFakeNews
0.82CVGHTM
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| GHTMRT=231.152025.08 | 0.82 | 0.28 | 0.96 | 0.99 | |
| Top2vec (Doc2Vec)RT=1650.722025.08 | 0.8 | 0.12 | 0.99 | 1 | |
| NMF (TF-IDF)RT=15.862025.08 | 0.72 | 0.17 | 0.94 | 0.99 | |
| Combined TM (BOW+SBERT)RT=685.302025.08 | 0.71 | 0.14 | 1 | 1 | |
| NMF (BOW)RT=7.842025.08 | 0.67 | 0.13 | 0.84 | 0.97 | |
| BERTopic (SBERT+UMAP+KMeans)RT=306.822025.08 | 0.67 | 0.07 | 0.94 | 0.99 | |
| BERTopic (SBERT+UMAP+Spectral)RT=353.992025.08 | 0.66 | 0.07 | 0.92 | 0.98 | |
| LDA (TF-IDF)RT=29.882025.08 | 0.65 | 0.09 | 0.97 | 1 | |
| BERTopic (SBERT+UMAP+Agglomerative)RT=231.102025.08 | 0.65 | 0 | 0.97 | 0.99 | |
| LDA (BOW)RT=32.382025.08 | 0.64 | 0.12 | 0.86 | 0.97 | |
| BERTopic (SBERT+UMAP+HDBSCAN)RT=355.892025.08 | 0.62 | 0.03 | 0.93 | 0.99 | |
| BERTopic (SBERT+PCA+HDBSCAN)RT=238.232025.08 | 0.62 | 0 | 0.93 | 0.99 | |
| ZeroShot TM (BOW+SBERT)RT=352.852025.08 | 0.6 | 0.05 | 0.37 | 0.75 | |
| BERTopic (SBERT+UMAP+DBSCAN)RT=232.832025.08 | 0.59 | 0.01 | 0.94 | 0.99 | |
| BERTopic (SBERT+SVD+HDBSCAN)RT=284.712025.08 | 0.56 | 0.03 | 1 | 1 | |
| BERTopic (SBERT+HDBSCAN)RT=281.032025.08 | 0.53 | 0.02 | 0.97 | 0.99 | |
| ETM (FastText)RT=410.922025.08 | 0.49 | -0.12 | 0.08 | 0 | |
| LSA (TF-IDF)RT=20.782025.08 | 0.45 | -0.03 | 0.42 | 0.83 | |
| LSA (BOW)RT=11.812025.08 | 0.44 | -0.01 | 0.39 | 0.79 | |
| BERTopic (GloVe+UMAP+HDBSCAN)RT=277.122025.08 | 0.43 | -0.11 | 0.83 | 0.95 | |
| Neural LDART=2186.462025.08 | 0.42 | -0.22 | 0.96 | 0.99 | |
| BERTopic (Fast Text+UMAP+HDBSCAN)RT=166.282025.08 | 0.4 | -0.12 | 0.86 | 0.96 | |
| ProdLDART=2354.882025.08 | 0.39 | -0.19 | 0.89 | 0.98 | |
| BERTopic (Word2Vec+UMAP+HDBSCAN)RT=175.852025.08 | 0.39 | -0.12 | 0.85 | 0.95 | |
| BERTopic (Doc2Vec+UMAP+HDBSCAN)RT=156.382025.08 | 0.36 | -0.12 | 0.84 | 0.94 |