| Prediction ratesa (%) |
---|
Without heart rate data | With heart rate data |
---|
Overall | 79 (62–88) | 79 (57–88) |
Overall transport | 74 (41–89) | 74 (41–89) |
Activity place | 79 (61–89) | 80 (55–89) |
Bike | 90 (77–100) | 92 (76–100) |
Private motorized | 69 (1–89) | 69 (0–87) |
Public transport | 62 (0–91) | 62 (0–90) |
Walking | 80 (23–95) | 80 (28–95) |
- aPrediction rates presented as median prediction rates from 126 iteration of RF models with 2.5th and 97.5th percentiles in the parentheses. The overall prediction rate is from a model that is unweighted by category size, while the mode-specific prediction rates are from a corrected model applying a weight. Correction for category size entails modifying the cut-offs for prediction to the observed proportions of the categories, at the forest prediction step (when aggregating information from all trees). For the same tree predictions, a higher “proportion of votes” is reached for rarer categories in the weighted vs. unweighted model