User Behavior Assessment Towards Biometric Facial Recognition System: A SEM-Neural Network Approach

TL;DR

This study investigates user acceptance of biometric facial recognition in smart homes, employing SEM and neural networks to analyze behavioral factors influencing system adoption.

Contribution

It introduces a combined SEM-ANN approach to model user behavior towards facial recognition technology in smart homes, incorporating TAM, PSQ, and SI.

Findings

Perceived System Quality significantly influences usability.

All variables in the framework affect behavioral intention.

Multi-analytical approach enhances understanding of user acceptance.

Abstract

A smart home is grounded on the sensors that endure automation, safety, and structural integration. The security mechanism in digital setup possesses vibrant prominence and the biometric facial recognition system is novel addition to accrue the smart home features. Understanding the implementation of such technology is the outcome of user behavior modeling. However, there is the paucity of empirical research that explains the role of cognitive, functional, and social aspects of end-users acceptance behavior towards biometric facial recognition systems at homes. Therefore, a causal research survey was conducted to comprehend the behavioral intention towards the use of a biometric facial recognition system. Technology Acceptance Model (TAM)was implied with Perceived System Quality (PSQ) and Social Influence (SI)to hypothesize the conceptual framework. Data was collected from 475respondents through online questionnaires. Structural Equation Modeling(SEM) and Artificial Neural Network (ANN) were employed to analyze the surveyed data. The results showed that all the variables of the proposed framework significantly affected the behavioral intention to use the system. The PSQ appeared as the noteworthy predictor towards biometric facial recognition system usability through regression and sensitivity analyses. A multi-analytical approach towards understanding the technology user behavior will support the efficient decision-making process in Human-centric computing.