An Analysis of Audio Features to Develop a Human Activity Recognition Model Using Genetic Algorithms, Random Forests, and Neural Networks

Table 9

Confusion matrix of the NN-based HAR model with all features.

⁢	Brew coffee	Cooking	Use microwave oven	No activity	Take a bath	Dish washing	Hand washing	Teeth brushing	Type II error
Brew coffee	47	10	1	0	2	2	1	1	0.266
Cooking	5	13	2	1	9	2	1	4	0.649
Use microwave oven	2	2	6	0	4	3	0	0	0.647
Take a bath	0	4	0	1	92	6	0	1	0.115
Dish washing	3	1	1	0	6	20	1	2	0.412
Hand washing	0	0	0	0	7	2	11	0	0.450
Teeth brushing	2	3	2	0	7	1	1	16	0.500
No activity	1	2	1	1	3	1	1	1	0.909
Type I error	0.217	0.629	0.538	0.667	0.292	0.459	0.313	0.360