Table 3. The statistical description of the Urdu language corpora for fake news detection.
Corpus | Bend the Truth | Urdu Fake News | ||
---|---|---|---|---|
Labels | Fake | Legitimate | Fake | Legitimate |
Vocabulary | 1,84,023 | 1,20,394 | 11,47,547 | 9,54,254 |
Minimum Length Article | 57 | 59 | 25 | 25 |
Maximum Length Article | 2,159 | 1,153 | 7,045 | 6,068 |
Total Article | 400 | 500 | 968 | 1,032 |