Arjan Ghosh is a Senior Lecturer in the Department of Computer Science and Engineering at Northern University of Business and Technology, Khulna, Bangladesh. He completed his M.Sc. in Computer Science and Engineering from Khulna University, Bangladesh, with Highest Distinction, following his B.Sc. in Computer Science and Engineering from BRAC University, Dhaka, Bangladesh. His academic and research interests include Artificial Intelligence, Machine Learning, Natural Language Processing, Internet of Things, communications, software development, and data-driven computing systems. He has authored and co-authored approximately 13 scholarly publications, including journal articles, conference papers, and book chapters. His teaching experience spans courses such as Object-Oriented Programming, Computer Networks, Computing, and Software Development. He also serves as a reviewer for international conferences and is actively engaged in student mentoring, research supervision, and interdisciplinary academic collaboration.

Title: AI-Driven Sea-Level Rise Prediction and Coastal Risk Assessment in Bangladesh Using Machine Learning and DEM Data -  Abstract:  Bangladesh is a very climate vulnerable country which has been identified to have great deal of threat imposed by sea level rise (SLR) to its coastal communities, ecosystems, and infrastructure. In this work we present an integrated machine learning and Digital Elevation Model (DEM) based framework which we put forth for the prediction of sea level rise and assessment of coastal flood risk in Bangladesh. We used historical tide gauge records from chosen coastal stations (which include Charchanga, Chittagong, Cox's Bazar, Khepupara, and Hiron Point) to develop and compare many forecasting models which are Linear Regression, ARIMA, Long Short Term Memory (LSTM) and a hybrid ensemble model. We then put together the predictive outputs which we obtained from high resolution DEM data to present flood risk maps which we did so under different sea level rise scenarios. Also we developed a Streamlit based web platform which we made to present projected inundation zones and which also gives AI supported information for decision support. We found out that the LSTM model did very well in terms of prediction of non linear temporal trends, also the hybrid ensemble model did very well in terms of stable and reliable performance across stations. DEM based simulations we did identified very vulnerable low level coastal areas which we present the value of putting together temporal forecasting with geospatial analysis. We validated our results using satellite imagery and historical data which we did to support the use of the put forth framework for coastal planning and climate change adaptation. We put forth a scalable and user friendly tool which policy makers, researchers and local stakeholders can use to support data based coastal risk management in Bangladesh.