Abstract:
Deep learning has emerged as a transformative paradigm in the development of classification systems across diverse domains, owing to its remarkable ability to automatically learn hierarchical feature representations from raw data. This paper provides a comprehensive overview of deep learning-based classification methods, highlighting popular architectures such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), transformers, and hybrid models. We examine their effectiveness in a wide range of applications including computer vision, natural language processing, healthcare diagnostics, cybersecurity, and smart city infrastructures. Despite their success, deep learning-based classifiers face significant challenges related to interpretability, computational cost, data dependency, and vulnerability to adversarial attacks. We further discuss current advancements in model optimization, transfer learning, and explainable AI aimed at overcoming these limitations. Finally, open research directions and future opportunities are identified to guide the development of more efficient, robust, and trustworthy classification systems.
Page(s):
27-27
DOI:
DOI not available
Published:
Journal: 4th International Conference of Sciences “Revamped Scientific Outlook of 21st Century, 2025” , November 12,2025, Volume: 1, Issue: 1, Year: 2025
Keywords:
natural language processing
,
Convolutional neural networks
,
Recurrent Neural Networks
,
Hybrid Models