Table 5. Normalization results achieved by proposed rules and regular expressions.
Dataset | Correct changes | Incorrect changes | Remained changes |
---|---|---|---|
Biology | 1,922 | 12 | 0 |
Physics | 2,213 | 3 | 0 |
Chemistry | 1,798 | 22 | 0 |
Urdu literature | 2,079 | 27 | 0 |
Social studies | 1,548 | 40 | 0 |
Combined dataset | 9,560 | 104 | 0 |