Effective Prediction System for Affective Computing on Emotional Psychology with Artificial Neural Network

EasyChair Preprint no. 8801, version 2

Abstract

Human emotions reveal mental health. Understanding people's emotions help make vital decisions. With recent improvements in AI and machine learning, affective computing has become an interesting topic of research that adapts human emotional behavior and increases learning outcomes connected to behavioral psychiatry. Machine learning algorithm evaluation improves prediction quality, yet problems arise with connected decisions. The suggested research predicts true emotion using neurophysiological data. Emotional changes trigger physiological responses. The suggested system uses Gaussian mixture models to develop a novel prediction algorithm utilizing the AMIGOS dataset. The dataset included ECG, EEG, and GSR (GSR). The findings affect the statistical response following data processing, measurable emotion labeling, and training samples. The provided system is compared to state-of-the-art statistical measures like standard deviation, population mean, etc. The system can compare an interpreter to validate emotion labeling parameters. A unique Emotional detecting artificial Neural Network (EMONN) system is improved by using deep learning models to find covariate values that help identify participant personality traits. Novel Emotional detecting artificial Neural Network (EMONN) achieves 93% accuracy with reduced computing time. A new Emotional detecting artificial Neural Network (EMONN) system is researched and developed by analyzing deep learning models to detect covariate values in the data set.

Keyphrases: Affective Computing, Amigos, emotion prediction, Emotional Psychology, Mental health matters, Personality Detection

@Booklet{EasyChair:8801,
  author = {V.S Bakkialakshmi and T Sudalaimuthu and Stephen Winkler},
  title = {Effective Prediction System for Affective Computing on Emotional Psychology with Artificial Neural Network},
  howpublished = {EasyChair Preprint no. 8801},

  year = {EasyChair, 2022}}