Hybrid Capsule-CNN with Puzzle Aquila Optimization for Energy-Efficient and Reliable Mental Health Analytics
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The rising prevalence of mental illnesses like depression and anxiety has strained the need to have automated, reliable and scalable diagnostic tools. Older tools that depended on clinical surveys and manual reviews tend to be time-intensive and prone to bias, hence they are not useful in large populations. This paper proposes a Hybrid Capsule Convolutional Neural Network (Capsule-CNN) with Puzzle Aquila Optimization (PAO) to provide energy-efficient and reliable mental health analytics to overcome these limitations. The structure takes advantage of the local characteristic of CNNs and spatial awareness of Capsule Networks and the hierarchical and spatial characteristics of the former and PAO is utilized to optimize the hyperparameters and the model weights to enhance robustness and minimize computational cost. The suggested system is tested on the review datasets of the real world where sentiment indicators of depression are present. Experimental analysis reveals that there is a high performance with respect to classification accuracy, detection latency, false positive rate, robustness index, and energy efficiency compared to the traditional CNN, RNN-LSTM, Capsule-CNN and hybrid BERT-based models. In this method, the use of deep learning in conjunction with advanced optimization techniques is noted as potentially capable of producing an intelligent and scalable mental health diagnostic system, applicable to real-time surveillance in IoT-based health care facilities.
Fields of Study & Subjects
Mental Health via Writing
Sentiment Analysis and Opinion Mining
Emotion and Mood Recognition
Publication details
- Journal / source
- International Conference on Intelligent Systems for Communication, IoT and Security (ICISCoIS)
- Pages
- 1–6
- Publication date
- 8 January 2026
- Language
- EN