Skip to main content
. Author manuscript; available in PMC: 2017 Apr 7.
Published in final edited form as: Technol Health Care. 2017;25(2):251–264. doi: 10.3233/THC-161274

Table 9.

Part B percent accuracy (Acc) and Area Under the ROC Curve (AUC) for all older adult participants using three different classifiers.

Feature Subsets Naive Bayes J48 Tree SMO-SVM

Acc (SD) AUC Acc (SD) AUC Acc (SD) AUC

All 59.4 (18.6) % 0.58 61.5 (16.6)% 0.63 57.7 (16.9)% 0.57
All Digital 60.7 (18.5)% 0.58 61.3 (16.4)% 0.63 56.1 (17.2)% 0.56
Standard 64.6 (17.1)% 0.64 58.1 (14.0)% 0.57 56.4 (12.2)% 0.55
Timing 64.7 (17.3)% 0.63 62.1 (16.4)% 0.63 59.3 (17.0)% 0.59
Mobility 57.6 (16.9)% 0.56 46.4 (11.3)% 0.42 56.5 (17.0)% 0.55