YouthMate : Evaluating the Functional Performance and Holistic Impact of a Multi-Domain AI Assistant for Young Adults

This paper presented a comprehensive technical and functional evaluation of YouthMate, a multi-domain AI assistant for youth wellbeing, examining the internal working, algorithmic design, and performance benchmarks of all six AI-powered modules: Mental Health, Study and Career, Finance and Budget, Fitness and Diet, Digital Wellbeing, and Resume and Job Assistance. The evaluation produced four principal findings.   First, all six modules achieved strong individual performance: the BERT-based Mental Health module reached 89.4% emotion classification accuracy; the Study and Career module achieved 86.4% course recommendation accuracy; the Finance regression model demonstrated an MAE of 0.07; and the Resume module achieved 84.9% job-role match accuracy — all consistent with or exceeding comparable single-domain systems in the literature.   Second, system-level performance was validated with average API response latencies of 1.3–1.8 seconds under standard and concurrent load conditions, supported by a 34.6% throughput improvement from the asynchronous Node.js architecture. Peak load testing identified horizontal autoscaling as the primary infrastructure investment required for production-scale deployment.   Third, the System Usability Scale evaluation yielded an overall mean score of 80.4 (Excellent), confirming that the platform's complexity is effectively abstracted by a well-designed conversational and dashboard interface.   Fourth, and most significantly, the cross-module integration analysis demonstrated that YouthMate's holistic architecture produces emergent intelligence — coordinated multi-domain responses to single user signals — that cannot be replicated by any combination of single-purpose applications. This emergent value is the defining characteristic of YouthMate's contribution to the field of AI-powered youth support.   With planned enhancements including wearable sensor integration, multilingual NLP, federated learning, and production-scale autoscaling, YouthMate is well-positioned to evolve from a validated prototype into a deployable, privacy-preserving, and globally accessible AI ecosystem for holistic student wellbeing.

[1] R. Bobade, A. Rajarshi, S. Khot, and P. Sathe, "YouthMate: AI Assistant for Young Minds," International Journal of Scientific Research in Engineering and Management (IJSREM), vol. 10, no. 02, Feb. 2026. [2] L. Chen, P. Chen, and Z. Lin, "Artificial intelligence in education: A review," IEEE Access, vol. 8, pp. 75264–75278, 2020. [3] E. Kochmar et al., "Automated personalized feedback improves learning gains in an intelligent tutoring system," in Proc. Int. Conf. on Artificial Intelligence in Education (AIED), 2021, pp. 140–146. [4] J. Brooke, "SUS: A quick and dirty usability scale," in Usability Evaluation in Industry, P. W. Jordan et al., Eds. London: Taylor and Francis, 1996, pp. 189–194. [5] M. R. Chowdhury, T. Qu, A. S. Aung, S. A. Kaseb, and K. N. Truong, "Predicting and Understanding College Student Mental Health with Interpretable Machine Learning," in Proc. ACM/IEEE Int. Conf. on Connected Health (CHASE), 2025. [6] S. G. Hart and L. E. Staveland, "Development of NASA-TLX (Task Load Index): Results of empirical and theoretical research," Advances in Psychology, vol. 52, pp. 139–183, 1988. [7] D. Dicheva, C. Dichev, G. Agre, and G. Angelova, "Gamification in education: A systematic mapping study," Journal of Educational Technology and Society, vol. 18, no. 3, pp. 75–88, 2015. [8] S. K. Jaiswal and P. Tiwari, "AI-Based Emotional Health Monitoring for Students Using Natural Language Processing," Int. Journal of Artificial Intelligence and Applications, vol. 14, no. 2, pp. 67–78, 2024. [9] M. A. Bari, S. S. Islam, and M. S. Rahman, "AI-Powered Digital Wellbeing Applications: A Review of Machine Learning Techniques and Challenges," IEEE Access, vol. 11, pp. 11642–11657, 2023. [10] J. Devlin, M. Chang, K. Lee, and K. Toutanova, "BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding," in Proc. NAACL, pp. 4171–4186, 2019.