Activity Feature Solving Based on TF-IDF for Activity Recognition in Smart Homes

Table 7

Recognition precisions using three classifiers concerning dataset cairo.

Feature	SVM	SMO	RF
FF	52.7%	79.8%	85%
TF-IDF	82	85	86.3
TF-IDF+Sigmod	80.5%	83.2%	85.4%
TF-IDF+Tanh	73.3%	85	86.1%