Stereoscopic Gated Graph Scalable Quantum Capsule Convolutional Neural Networks with Artificial Hummingbird Algorithm for Social Media Content Classification and Community Detection
A. Elumalai, S. Rukmani Devi, D. Rosy Salomi Victoria, A. Jagadeeswaran, Aparajita Mukherjee, K. Vinoth · 2025
Social Media Content Classification and Community Detection (SMCCCD) classify content and identify communities through deep learning, and NLP. Traditional models are weak in scalability and interpretability. To solve all these problems, this paper introduces a Stereoscopic Gated Graph Scalable Quantum Capsule Convolutional Neural Networks with Artificial Hummingbird Algorithm (SGGSQCCN2Nets+AHA) for social content classification and community detection. Data inputs are accessed from the Twitter dataset and are initially preprocessed by the Natural Language (NL) method. After preprocessing, content classification and community detection are done using the Stereoscopic Gated Graph Scalable Quantum Capsule Convolutional Neural Networks (SGGSQCCN2Nets) and optimization is then performed with the Artificial Hummingbird Algorithm (AHA). The performance of the new SGGSQCCN2Nets+AHA model is tested on a Twitter dataset with a high accuracy rate of 99.9% and specificity of 99.8%. The proposed model is developed by utilizing Python programming language. The outcome of the proposed SGGSQCCN2Nets+AHA model enhances the performance of content classification and improves user community detection accuracy. It is effective in addressing scalability issues, obtaining complex features, and offering more interpretability. The model is also resilient against noisy data dynamically adapts to evolving social media trends and outperforms traditional deep learning techniques in real-time analysis.