Table 5 Influence of the generated claims on UTDRM. The highest scores for each dataset and metric are in bold
From: UTDRM: unsupervised method for training debunked-narrative retrieval models
Datasets | Metrics | Generated Claims | |||
|---|---|---|---|---|---|
N = 2 | N = 6 | N = 10 | N = 20 | ||
Snopes | MAP@1 | 0.821 | 0.831 | 0.830 | 0.829 |
MRR | 0.881 | 0.890 | 0.890 | 0.889 | |
CLEF 22 2A | MAP@1 | 0.914 | 0.933 | 0.933 | 0.933 |
MRR | 0.934 | 0.948 | 0.948 | 0.949 | |
CLEF 21 2A | MAP@1 | 0.906 | 0.906 | 0.901 | 0.896 |
MRR | 0.936 | 0.936 | 0.932 | 0.931 | |
CLEF 20 2A | MAP@1 | 0.935 | 0.945 | 0.945 | 0.945 |
MRR | 0.957 | 0.961 | 0.963 | 0.964 | |
Average | MAP@1 | 0.894 | 0.904 | 0.902 | 0.901 |
Twitter-based | MRR | 0.927 | 0.934 | 0.933 | 0.933 |
Politifact | MAP@1 | 0.508 | 0.516 | 0.500 | 0.496 |
MRR | 0.616 | 0.627 | 0.619 | 0.615 | |
CLEF 22 2B | MAP@1 | 0.362 | 0.392 | 0.400 | 0.392 |
MRR | 0.441 | 0.467 | 0.473 | 0.468 | |
CLEF 21 2B | MAP@1 | 0.323 | 0.348 | 0.354 | 0.335 |
MRR | 0.394 | 0.422 | 0.424 | 0.416 | |
Average | MAP@1 | 0.397 | 0.419 | 0.418 | 0.408 |
Political-based | MRR | 0.484 | 0.505 | 0.505 | 0.499 |