Table 1 Kappa statistic on the training sets built by manual classification
From: Redocumenting APIs with crowd knowledge: a coverage analysis based on question types
Observed | Kappa | Strength | |
|---|---|---|---|
agreement (%) | |||
Android | 78.75 | 0.675 | Substantial |
Swing | 84.75 | 0.750 | Substantial |
Both | 81.75 | 0.712 | Substantial |