Feature Selections Using Minimal Redundancy Maximal Relevance Algorithm for Human Activity Recognition in Smart Home Environments

Table 4

The information values of feature groups under the D-R criterion.

Feature
Iteration	1	2	3	4	5	6	7	8	9	10	11	12	13
0	1.188	0.688	0.047	1.173	0.863	1.490	1.200	1.321	1.508	1.535	1.168	1.183	0.530
1	0.476	0.213	−0.160	−0.009	0.441	0.774	0.310	0.146	−1.989	—	−0.032	0.607	0.134
2	0.376	0.230	−0.075	0.281	0.434	—	0.306	0.343	−0.594	—	0.261	0.363	0.151
3	0.520	0.288	−0.042	0.430	—	—	0.440	0.338	−0.041	—	0.453	0.520	0.195
4	0.425	0.268	−0.030	0.488	—	—	0.451	0.428	0.206	—	0.518	—	0.198
5	0.474	0.299	−0.023	0.452	—	—	0.456	0.458	0.238	—	—	—	0.216
6	—	0.173	−0.022	0.486	—	—	0.454	0.479	0.330	—	—	—	0.196
7	—	0.206	−0.019	—	—	—	0.484	0.482	0.326	—	—	—	0.207
8	—	0.207	−0.016	—	—	—	—	0.498	0.365	—	—	—	0.202
9	—	0.196	−0.025	—	—	—	—	—	0.364	—	—	—	0.188
10	—	0.196	−0.040	—	—	—	—	—	—	—	—	—	0.187
11	—	—	−0.040	—	—	—	—	—	—	—	—	—	0.190