Mobile Information Systems / 2016 / Article / Tab 8

Research Article

A Mobile Application for Easy Design and Testing of Algorithms to Monitor Physical Activity in the Workplace

Table 8

Base classifiers accuracy by class.
(a) DT

Precision Recall -score Class

0.938 0.942 0.94 Walking
0.969 0.971 0.97 Jogging
0.72 0.705 0.712 Upstairs
0.7 0.699 0.7 Downstairs
0.984 0.972 0.978 Sitting
0.98 0.988 0.984 Standing

0.905 0.905 0.905 Weighted average

(b) -NN

Precision Recall-scoreClass

0.959 0.9930.976Walking
0.99 0.9920.991Jogging
0.911 0.8550.882Upstairs
0.915 0.8330.872Downstairs
0.995 0.9790.987Sitting
0.988 0.9910.99Standing

0.961 0.9620.961Weighted average

(c) NN

Precision Recall-scoreClass

0.981 0.9870.984Walking
0.991 0.9880.989Jogging
0.896 0.9130.904Upstairs
0.906 0.8690.887Downstairs
0.99 0.9820.986Sitting
0.985 0.9910.988Standing

0.968 0.9680.968Weighted average