Table 10 Feature impact analysis of the machine learning model without post-processing on the development set
From: Ambiguity and variability of database and software names in bioinformatics
Feature group | Recall (%) | Precision (%) | F-score (%) |
|---|---|---|---|
All features | 54 (49) | 78 (70) | 62 (57) |
No lexical features | 46 (43) | 68 (62) | 54 (50) |
No syntactic features | 53 (48) | 77 (69) | 61 (55) |
No orthographic features | 48 (43) | 70 (62) | 55 (50) |
No dictionary features | 49 (44) | 70 (62) | 57 (51) |