By contrasting the Logistic Regression method with the Random Forest algorithm, a Novel Stream Sensor Based Human Activity Recognition Method Using Deep Learning Techniques

Abstract

The main fundamental target of this research study is to further devolop accuracy for Human Activity Recognition using Deep Learning Techniques.Material And Methods:  Human activity recognition in computer vision is performed by Logistic regression with sample size 10 and Random forest algorithm with sample size 10. It has been iterated at different times to predict the accuracy percentage of human activity. Results: The Human Activity recognition utilizing Logistic regression 95.52% and with Random forest 89.3%. Logistic regression seems to perform essentially better compared toRandom forest (p=0.90) (p<0.05). Conclusion: Within this human activity recognition Logistic regression has more precision than Random forest