Dream Emotion Prediction Using Random Forest Algorithm on EEG Signal Data: An Age wise Analysis

Abstract

Dreams serve as a reflection of both conscious and subconscious emotional states, often influenced by age, psychological development, and cognitive function. This study presents an agewise analysis of dream emotion classification using EEG (Electroencephalogram) signals and a Random Forest algorithm. EEG data collected during REM sleep was used to classify emotions into three categories: Positive, Negative, and Neutral. The study evaluates emotional trends across four distinct age groups 0–16, 17–30, 31–45, and 46–100 years through machine learning and visual analytics, including bar charts, confusion matrices, heatmaps, ROC curves, and pie charts [1]. The Random Forest model achieved a high overall accuracy of 99%, with an AUC score of 0.95 for the 17–30 age groups, indicating superior predictive performance. Results reveal significant variations in emotional expression during dreams across age brackets, with negative emotions predominantly observed in adults. These findings demonstrate the potential of EEG-based emotion prediction for advancing applications in sleep therapy, mental health monitoring, and affective computing.