Deep Learning-Driven Assessment of Adolescent Emotional Intelligence and Psychological Well-Being Using Optimized Feature Selection
International Journal of Medical Toxicology and Forensic Medicine,
Vol. 16 (2026),
1 Dey 2026
,
Page 1-7
https://doi.org/10.22037/ijmtfm.v16.53413
Abstract
Background: Emotional Intelligence (EI) is an important factor that affects adolescents' academic performance, social behavior, psychological health, and life resilience. Identifying the major emotional intelligence factors that affect adolescent behavior in colleges can help teachers and psychologists design effective intervention strategies. However, the EI datasets contain many emotional and behavioral features that require feature selection and classification methods to improve prediction accuracy. This research recommends a Hybrid Feature Selection-based Emotional Deep Learning (HFS-EDL) Framework for estimating adolescents' emotional intelligence at the undergraduate and postgraduate levels in Tamil Nadu.
Methods: The survey was carried out using a questionnaire with 3510 students across five districts of Tamil Nadu. The dataset includes 25 attributes; age group is the target variable. After data preprocessing, the most significant EI factors were selected using two techniques: Chi-Square and Mutual Information (MI). Three deep learning models, namely Artificial Neural Network (ANN), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM), were then used to classify the selected features. Model performance was measured using accuracy, recall, precision, specificity, and error rate.
Results: The experimental results showed that the feature selection method performs better than the full feature set. Mutual Information is most effective with ANN and RNN models, while Chi-Square feature selection is most effective with LSTM.
Conclusion: Overall, the proposed HFS-EDL framework demonstrates the capability to successfully recognize important emotional intelligence factors and improve classification accuracy in predicting an adolescent's emotional intelligence.
- Emotional Intelligence, Deep learning, Artificial neural network, Recurrent neural network, Long short-term memory network
How to Cite
References
[1] Aziz I, Arshad S, Hussan MA. An adaptive machine learning framework integrating large language models to assess and enhance emotional intelligence in adolescents. Asian J Psychiatry. 2026;104851. [DOI: 10.1016/j.ajp.2026.104851]
[2] Irfan H, et al. Enhancing group-based emotional intelligence in children through an adaptive activity system driven by artificial intelligence. PeerJ Comput Sci. 2026;12:e3364. [DOI: 10.7717/peerj-cs.3364]
[3] Aye SM. An Analytical Study of Emotional Intelligence and Risk-taking Behavior of Adolescents. Int J Transdiscipl Res Perspect. 2026;2(8):13-8. [DOI: 10.65138/ijtrp.2026.v2i8.70]
[4] Doghonadze N, Jikia N, Gkatsa T. Developing Social-Emotional Intelligence in the Process of Teaching. In: Promoting Positive Behaviors and Emotional/Social Wellbeing in School. IGI Global Scientific Publishing; 2026. p. 1-26. [DOI: 10.4018/979-8-3373-3351-9.ch001]
[5] Chen HL, Tedla YG, Chen SY. Enhancing students’ emotional intelligence through nonviolent communication-based emotional scaffolding-infused digital storytelling. J Educ Comput Res. 2026;64(2):371-402. [DOI: 10.1177/07356331251396347]
[6] Li X. Machine learning-based risk warning for adolescents’ prosocial behavior. Appl Psychol Health Well-Being. 2026;18(3):e70156. [DOI: 10.1111/aphw.70156]
[7] Jiang C, Feng H. Bullying Victimization and Delinquent Behavior During Adolescence: A Double Machine Learning Approach. Child Fam Soc Work. 2026;1-11. [DOI: 10.1111/cfs.70212]
[8] Wang W, Li Q, Zhang B. Classroom Situational Learning Behavior Recognition Based on YOLOv5 and ShuffleNet V2 Algorithms Optimization. J Inst Eng (India) Ser C. 2025;106(6):1763-76. [DOI: 10.1007/s40032-025-01252-2]
[9] Piombo MA, La Grutta S, Epifanio MS, Di Napoli G, Novara C. Emotional intelligence and adolescents’ use of artificial intelligence: A parent-adolescent study. Behav Sci. 2025;15(8):1142. [DOI: 10.3390/bs15081142]
[10] Vanhoffelen G, Vandenbosch L, Schreurs L. Teens, Tech, and Talk: Adolescents’ Use of and Emotional Reactions to Snapchat’s My AI Chatbot. Behav Sci. 2025;15(8):1037. [DOI: 10.3390/bs15081037]
[11] Zhang S, Huang Y, Zhong Y, Huang Q, Huang X, Meng J. Unraveling the Dynamics of Adolescent Loneliness: A Longitudinal Study of its Reciprocal Links with Perceived Control and Emotional Intelligence. J Youth Adolesc. 2026;55(2):536-49. [DOI: 10.1007/s10964-025-02294-6]
[12] Hussain L, Alabdulkreem E, Lone KJ, Al-Wesabi FN, Nour MK, Hilal AM, et al. Feature ranking chi-square method to improve epileptic seizure prediction using machine learning algorithms. Waves Random Complex Media. 2026;36(4):6456-82. [DOI: 10.1080/17455030.2023.2226246]
[13] Gu X, Li X, Ke W, Wang Z, Lv Y. Feature selection algorithm based on ant colony optimization and mutual information. Cluster Comput. 2026;29(3):180. [DOI: 10.1007/s10586-026-06035-7]
[14] Shobana R, Kumar R, Jaint B. A recurrent neural network-based identification of complex nonlinear dynamical systems: a novel structure, stability analysis, and a comparative study. Soft Comput. 2026;30(2):1245-61. [DOI: 10.1007/s00500-023-09390-4]
- Abstract Viewed: 35 times
- 53413 Downloaded: 4 times