Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{189705,
        author = {Diya S Thange and Monika M M and Bhanupriya V P and Ganavi R D and Rohan D Joel},
        title = {A Hybrid AI and Rule-Based Analytics Framework Using Pre-Trained NLP Sentiment Models for Web-Based Mental Well-Being Self-Assessment},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {8},
        pages = {1-9},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=189705},
        abstract = {Mental well-being issues such as stress, emotional distress, and mood imbalance are increasingly prevalent, yet many individuals lack access to simple, ethical, and transparent tools for early self-awareness and screening. Existing digital mental health solutions often rely on complex machine learning models, opaque decision-making, or clinical datasets, which may limit accessibility, interpretability, and responsible use for non-diagnostic purposes.
This paper presents a web-based hybrid mental well-being self-check system that combines rule-based analytics with sentiment analysis using pre-trained Natural Language Processing (NLP) models to provide a structured, interpretable, and user-friendly screening framework. The system employs a multi-domain questionnaire covering mood, anxiety, stress, sleep, energy, and social well-being, where responses are processed through a rule-based scoring mechanism to compute category-wise and overall difficulty indices. Additionally, free-text emotional input provided by the user is analyzed using a pre-trained transformer-based sentiment model to capture linguistic emotional cues. These two components are fused to generate a hybrid assessment that enhances contextual understanding while maintaining transparency and ethical boundaries.
The proposed system delivers visual analytics, including tabular summaries and graphical representations, to support user comprehension and reflection. Experimental demonstrations indicate consistent and interpretable outputs suitable for awareness-level screening. The framework is not intended for diagnosis but serves as an accessible tool for early self-reflection and mental health awareness. The proposed approach highlights the effectiveness of combining rule-based systems with pre-trained NLP models to create responsible, scalable, and explainable digital mental health applications.},
        keywords = {Artificial Intelligence, Digital Mental Health, Hybrid AI System, Mental Well-being Screening, Natural Language Processing, Rule-based Analytics, Sentiment Analysis, Web-based Self- Assessment},
        month = {December},
        }