Jia Uddin is an Assistant Professor in Artificial Intelligence and Big Data Department, at Endicott College, Woosong University, South Korea, and an Associate Professor (On Leave), Computer Science and Engineering Department at Brac University, Dhaka, Bangladesh. He received Ph.D. in Computer Engineering from the University of Ulsan, Korea, in January 2015 and M.Sc. in Telecommunications from Blekinge Institute of Technology, Sweden in June 2010. He was an Assistant Professor in the CSE department at BRAC University and the CCE department at International Islamic University Chittagong, Bangladesh. He was invited as a visiting faculty member at the School of Computing, Staffordshire University, Stoke-on-Trent, United Kingdom funded by a European Union Grant in April 2017, and was invited to Professor at Telkom University, Indonesia in Summer 2021. Dr. Jia received the Best Research Faculty award in the 2016 academic year at BRACU for his outstanding research contributions in the area of multimedia signal processing. He is supervising several undergraduate and graduate thesis students and his research students’ papers won Best paper awards in several international conferences: ICEEICT-2016, ICCIT-2016, IEEE ICAEE-2017, ICERIE-2017, ICMIP2019, IHCI2020, and IVIC2021. Dr. Jia is the author of 3 books related to Data Science and Computer Vision published by Woosong publisher and has 47 SCI/Scopus indexed Journal publications.
Dr. Jia is involved with different research communities at home and abroad by serving as a member of the organizing committee, technical committee, technical Session Chair, and reviewers in different peer-reviewed journals: IEEE Access, Multimedia Tools and Applications (Springer), Journal of Supercomputing (Springer), Wireless Personal Communication, SAI Journal, Neural Computing and Applications (Springer), Journal of Information Processing Systems, etc. His research interests include fault diagnosis, computer vision, and multimedia signal processing.

Title: Smart diagnosis though artificial intelligence in various applications

Abstract:
In industry 4.0, artificial intelligence (AI) based smart devices are widely used in different applications such as smart healthcare, smart-factory, smart home, smart city, etc. Smart devices use different sensors to collect real-time data from the environment and the data are used in the AI models. Earlier AI models are mostly machine learning-based models where feature engineering plays a vital role in the diagnosis (detection and prediction). Optimal feature selection is a key issue in Machine learning (ML) models and is a challenging task as features change with the environments. To overcome the limitations of ML models, deep learning models are used nowadays, where deep features are automatically extracted for diagnosis. Computational complexity is a major concern in deep learning models and the availability of limited datasets in various applications. Nowadays, to deploy the deep learning-based diagnosis models in portable devices different key techniques like transfer learning, self-supervised learning, few shot-learning, etc. are playing a vital role in the smart diagnosis for various applications.