Table 2 Results when applying rule set in table i to a corpus of 24 millions tweets. The last rows indicates the numbers of tweets extracted with given effects (insomnia, stress and headache)
From: Extracting health-related causality from twitter messages using natural language processing
Matched rule # | Insomnia (of 3827) | Stress (of 29,705) | Headache (of 11,252) |
|---|---|---|---|
1 | 58 | 381 | 78 |
2 | 4 | 12 | 3 |
3 | 0 | 4 | 1 |
4 | 1 | 21 | 2 |
5 | 0 | 32 | 0 |
6 | 9 | 51 | 10 |
Total | 72 | 501 | 94 |
# extracted causalities | 41 | 98 | 42 |