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

Table 4

Recognition accuracies using three classifiers concerning dataset tulum2009.

Feature	SVM	SMO	RF
Frequency Feature (FF)	67.2%	79%	90.8
TF-IDF	87	85%	90.8
TF-IDF+Sigmod	87	88.4	90.5%
TF-IDF+Tanh	85.9%	87.2%	90.7%