From: Multimodal machine learning for language and speech markers identification in mental health
Set | Characteristic | Marker presence |
---|---|---|
1 | Starting setup | 45% |
2 | Sensitive | 75% |
3 | Balanced Sensitivity | 86% |
4 | Moderately Strict | 39% |
5 | Increased Sensitivity | 96% |
6 | Increased Specificity | 22% |
7 | Refined based on statistics | 43% |
8 | Refined based on statistics and distribution plots | 29.6% |
9 | Re-refined based on statistics | 34% |
10 | Moderately Inclusive | 49% |