Deep Learning: A Comprehensive Overview on Techniques, Taxonomy, Applications and Research Directions

Deep Learning: A Comprehensive Overview on Techniques, Taxonomy, Applications and Research Directions

Received: 29 May 2021 / Accepted: 7 August 2021 / Published online: 18 August 2021 | Iqbal H. Sarker
This article provides a comprehensive overview of deep learning (DL), a core technology in the Fourth Industrial Revolution (4IR or Industry 4.0). DL, rooted in artificial neural networks (ANN), has become a hot topic in computing due to its ability to learn from data. The article presents a structured taxonomy of DL techniques, categorizing them into supervised or discriminative learning, unsupervised or generative learning, and hybrid learning. It highlights the challenges in building appropriate DL models, such as the dynamic nature of real-world problems and the lack of transparency in black-box models. The article also discusses the real-world applications of DL, including healthcare, visual recognition, text analytics, and cybersecurity. Finally, it identifies ten potential research directions for future DL modeling, emphasizing the importance of effective data representation, new algorithm design, and model optimization. The overall goal is to serve as a reference guide for both academic and industry professionals in the field of DL.This article provides a comprehensive overview of deep learning (DL), a core technology in the Fourth Industrial Revolution (4IR or Industry 4.0). DL, rooted in artificial neural networks (ANN), has become a hot topic in computing due to its ability to learn from data. The article presents a structured taxonomy of DL techniques, categorizing them into supervised or discriminative learning, unsupervised or generative learning, and hybrid learning. It highlights the challenges in building appropriate DL models, such as the dynamic nature of real-world problems and the lack of transparency in black-box models. The article also discusses the real-world applications of DL, including healthcare, visual recognition, text analytics, and cybersecurity. Finally, it identifies ten potential research directions for future DL modeling, emphasizing the importance of effective data representation, new algorithm design, and model optimization. The overall goal is to serve as a reference guide for both academic and industry professionals in the field of DL.
Reach us at info@study.space
Understanding Deep Learning%3A A Comprehensive Overview on Techniques%2C Taxonomy%2C Applications and Research Directions