Open AccessIdentification of Violent Response with Stretch Sensor Data from a Smart-Jacket using Naïve Bayes Algorithm
Author(s) -
Princy Randhawa,
Vijay Shanthagiri,
Ajay Kumar
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.a9244.119119
Subject(s) - naive bayes classifier , bayes' theorem , outlier , computer science , identification (biology) , artificial intelligence , set (abstract data type) , property (philosophy) , pattern recognition (psychology) , data mining , data set , machine learning , algorithm , bayesian probability , support vector machine , philosophy , botany , epistemology , biology , programming language
In this paper, a smart-jacket using stretch sensors, pressure sensors was built for purpose of generating body-movements data and in order to record different kinds of signals and the distribution of the same on the jacket. Every degree of motion, when exercised, generates voltage changes in the stretch sensors as it is its property to do so. This data is collected in a flora chip set, which is Arduino based. The collected data is processed, pruned and filtered for outliers. This paper concerns with a supervised learning algorithm called Naive Bayes, which is applied over independent datasets, meaning one set of observation has no direct relations to each other. The placement of sensor are on the shoulders and elbows and the responses from each are independent of each other. Using Naive Bayes, the date has been classified for the violent response and the normal action.