Proceedings of International Conference on Computing, Circuits, Energy and Materials 2022 (ICCCEM 2022) 359 Artificial Intelligence and Machine Learning Driven Framework for Investigating the Usefulness of Physiological Indicators for User Trust in Forecasting Decision Making Sanjeev Kumar Thalari 1, a) , Afsana Anjum 2, b) , Mohammed Saleh Al Ansari 3, c) , R. Manikandan 4, d), , Prema.S 5, e) , Anuradha.S 6, f) , * A.Balakumar 7,g) 1 CMR Institute of Technology, Bengaluru, Karnataka, India 2 Dept.of Computer Science, Jazan University Jazan, KSA. 3 College of Engineering, Department of Chemical Engineering, University of Bahrain, Bahrain. 4 Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology, Avadi, Chennai, Tamil Nadu, India-600062. 5 Department of English, Panimalar Engineering College,Poonamallee, Chennai, Tamil Nadu-600123. 6 Department of English, Sri Sai Ram Engineering College, Sai Leo Nagar,West Tambaram Poonthandalam, Village, Chennai, Tamil Nadu 602109. 7 Department of Electronics and communication Engineering, K.Ramakrishnan College of Engineering, Trichy, Tamil Nadu, India, a) sanjeevkumar.t@cmrit.ac.in, b) aisrar@jazanu.edu.sa, c) malansari.uob@gmail.com d) rmani16806@gmail.com, rmanikandan@veltech.edu.in, e) premasubramanian08@gmail.com, f) anuradha.eng@sairam.edu.in, g) balakumar2712@gmail.com Abstract. Trust Machine Learning is one of the most crucial concerns of black-box Machine Learning because of its huge effect on practical applications. This research sheds light on the significance of demonstrating that user confidence is affected by the training of information sets on machine learning forecasts.A framework enabling fact-checking is presented in such a predictive scenario involved in decision-making to permitthe users to immediately evaluate the training statistics with varying effects on the prediction utilising a representation depending on paralleled dimensions. This study examines if physiological signals of consumer trustworthiness, such as that of the Galvanic Skin Response (GSR) and Blood Volume Pulse (BVP), could be utilized in predicting decision-making.A study found that displaying the effects of training information coordinates increased users ’ confidence in prognostications, but only if the training statistics had greater influencing values in rising design effectiveness scenarios, when users could support their decisions with much more evidence that was comparable to the evaluation statistic. GSR and BVP traits are connected with user trust under a variety of effect & method efficiency factors, according to this physiological signal study. The findings of the study suggest that physiological-indicators might be used in AI system user interfaces to accurately communicate changes in user trust while making prediction decisions. Keywords. Machine learning, Physiological feature, Trust, Black-box, Performance conditions INTRODUCTION Data sets are becoming more available in a variety of industries, including infrastructure, transportation, energy, health, telecommunications, finance and education. Getting insights out of these big-data &solutions which are data-analytics-driven are becoming increasingly in demand, thanks to substantial breakthroughs in Machine Learning (ML). While we are constantly seeing Machine Learning based AI systems which appear to perform or have done amazingly perfect in real-world circumstances.It's not obvious how these machine learning techniques made this prognosis or how trustworthy the outcome or choice made depending only on the predictions.These inquiries demonstrate that consumer adoption of machine learning in systems with a pragmatic focus is strongly influenced by consumer knowledge and assurance in the technology.The notion of