Cognitive-Aware Emotion Detection: A Novel Deep Reinforcement Learning Framework for Adaptive Interaction Systems
Dandao Xuebao/Journal of Ballistics · 29.05.2026 · 38 (2) · 77-89
DOI: 10.52783/dxjb.v38.312 ↗Kurz gefasst
Fachartikel in der Zeitschrift Dandao Xuebao/Journal of Ballistics, online erschienen am 29.05.2026. Verfasst von Prashant P Baste.
Abstract (Originalsprache)
A cognitive-aware emotion detection framework is necessary to enable interactive systems to recognize nuanced human affect and modify responses with greater contextual sensitivity. While context-aware decision-making is the primary emphasis of current cognitive interaction techniques, they frequently lack mechanisms that combine emotional inputs with adaptive reinforcement-driven behaviors. The present research aims to design an emotion-centred adaptive mechanism powered by a deep reinforcement learning (DRL) architecture optimized for real-time decision refinement. Data is gathered from publicly available Multimodal EEG Emotion Interaction Dataset on Kaggle, comprising 1,500 recordings with EEG signals, cognitive load, task difficulty, and multimodal stimuli for emotion analysis. Pre-processing involves noise reduction, Min Max normalization, and temporal alignment using a suitable signal-cleaning technique to ensure coherent input representation. Feature extraction employs a convolutional Auto encoder (CAE) to capture spatial and temporal affective indicators. The proposed framework follows a structured path where raw inputs are collected, pre-processed, converted into discriminative emotional embeddings, and fed into the reinforcement module for adaptive output generation. The core contribution is the Honey Badger Algorithm-tuned Dueling Double Deep Q-Learning (HBA-D3QN), where the HBA mechanism performs hyperparameter tuning, D3QN stabilizes action selection, and the duelling structure distinguishes value estimation from advantage computation, enabling more precise emotion-driven decisions. This mechanism enhances sensitivity to emotional fluctuations, ensuring adaptive system reactions. Experimental evaluation indicates improvements in recognition accuracy, Precision, Recall, F1 Score and reward progression across emotional tasks. The framework demonstrates strong potential for next-generation adaptive interaction systems capable of aligning cognitive understanding with emotion-aware behavioural intelligence.
Zitieren
APA
Baste, P. P. (2026). Cognitive-Aware Emotion Detection: A Novel Deep Reinforcement Learning Framework for Adaptive Interaction Systems. Dandao Xuebao/Journal of Ballistics, 38(2), 77-89. https://doi.org/10.52783/dxjb.v38.312
Vancouver
Baste PP. Cognitive-Aware Emotion Detection: A Novel Deep Reinforcement Learning Framework for Adaptive Interaction Systems. Dandao Xuebao/Journal of Ballistics. 2026;38(2):77-89. doi:10.52783/dxjb.v38.312
BibTeX
@article{Baste2026,
author = {Baste, Prashant P},
title = {Cognitive-Aware Emotion Detection: A Novel Deep Reinforcement Learning Framework for Adaptive Interaction Systems},
journal = {Dandao Xuebao/Journal of Ballistics},
year = {2026},
volume = {38},
number = {2},
pages = {77--89},
doi = {10.52783/dxjb.v38.312}
}
EUVIMED