Table 2 Larger training set sizes (negative, positive) modestly improve algorithm performance
Negative Papers | Positive Papers | Paper Score |
|---|---|---|
10 | 100 | 0.60 |
20 | 100 | 0.63 |
30 | 100 | 0.63 |
40 | 100 | 0.64 |
10 | 200 | 0.56 |
20 | 200 | 0.59 |
30 | 200 | 0.58 |
40 | 200 | 0.60 |
10 | 300 | 0.60 |
20 | 300 | 0.63 |
30 | 300 | 0.64 |
40 | 300 | 0.66 |
10 | 400 | 0.61 |
20 | 400 | 0.65 |
30 | 400 | 0.66 |
40 | 400 | 0.66 |