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

Table 6

Recognition precisions using three classifiers concerning dataset tulum2009.

Feature	SVM	SMO	RF
FF	64.3%	71.5%	88%
TF-IDF	82.7	78.5%	88.1
TF-IDF+Sigmod	81.9%	85.3	87.5%
TF-IDF+Tanh	80.7%	82.1%	87.9%