Decision-Making Techniques and Methods for Sustainable Technological Innovation

10Efficient Human Threat Recognition Using Novel Logistic Regression Compared Over Linear Regression with Improved Accuracy

P. Sai Sateesh and Vijaya Bhaskar K.*

Department of Computer Science Engineering, Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences, Saveetha University, Chennai, India

Abstract

The objective of this study is to develop an efficient human threat recognition system using logistic regression and compare its performance with a system based on linear regression. The primary goal is to enhance accuracy in threat detection by leveraging the capabilities of logistic regression. The research utilizes two datasets, each comprising 100 tuples of data. The first dataset employs a novel neural network technique termed novel logistic regression, while the second dataset is analyzed using linear regression. The computations were conducted using G*Power 3.1 with an α level of 0.05 and a confidence interval of 95% to ensure statistical rigor. The analysis reveals that the novel logistic regression algorithm achieves an accuracy of 96.40%, outperforming the linear regression model, which recorded an accuracy of 95.35%.

Keywords: G*Power, human threat recognition, industry, K-nearest neighbor, linear regression, machine learning, novel logistic regression

10.1 Introduction

A wide range of real-time detecting applications has become increasingly accessible, especially in the domains of life logging and wellness tracking. These applications ...