Ashraful Islam
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Ashraful Islam has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Sujan Chandra Roy
Institute of Information and Communication Technology, Chittagong University of Engineering and Technology, Chattogram 4349, Bangladesh
2 shared publications - A. K. M. Adham
Department of Irrigation and Water Management, Bangladesh Agricultural University, Mymensingh 2202, Bangladesh
1 shared publication - Muhammad Sajjadur Rahim
University of Rajshahi
1 shared publication - Nusrat Jahan
Department of Environmental Science, Bangladesh Agricultural University, Mymensingh 2202, Bangladesh
1 shared publication - Sayeed Iftekhar Yousuf
Department of Computer Science and Mathematics, Bangladesh Agricultural University, Mymensingh, Bangladesh
1 shared publication - Touhidul Islam
Department of Irrigation and Water Management, Bangladesh Agricultural University, Mymensingh 2202, Bangladesh
1 shared publication
Latest Publications
- Open Access
- 1 May 2026
- Published by IWA Publishing in Journal of Water and Climate Change
- Vol. 17 (5) , 1329-1357
River water level forecast sensitivity to hydrometeorological input variable selection: a comparison of machine and deep learning models in Bangladesh
Conceptual summary showing LSTM and GRU outperform ML models under meteorological-only inputs, while RFR and LGBMR dominate when lagged water level data are available, across five scenarios in the Old Brahmaputra River, Bangladesh. Accurate river water level forecasting is critical for flood management in Bangladesh's monsoon-dominated deltaic environment, yet optimal model selection under varying data availability remains unexplored. This study systematically compared six machine learning (ML) and deep learning (DL) algorithms, specifically linear regression, support vector regression, random forest regression (RFR), light gradient boosting machine regression (LGBMR), long short-term memory (LSTM), and gated recurrent unit (GRU), for water level prediction at Islampur station using 25 years of daily hydrometeorological data (1999–2024) across five input scenarios employing 5-day lag configurations. DL architectures dominated meteorological-only scenarios, with LSTM achieving R2 = 0.8809, explained variance score (EVS) = 0.8874, and symmetric mean absolute percentage error (SMAPE) = 4.83%, confirming unbiased symmetrical predictions. ML algorithms excelled with autoregressive inputs, where LGBMR achieved R2 = 0.9960 and RFR attained R2 = 0.9984; the latter's EVS = 0.9984 and SMAPE = 0.40% confirmed purely stochastic residuals. Principal component analysis (PCA)-based composite scoring of nine complementary metrics provided objective model rankings. Zero-shot spatial transferability at Sarishabari confirmed reproducible hierarchies without retraining, with violin plots and PCA composite scoring corroborating near-zero prediction bias across all transferred configurations. Results demonstrate that LSTM and GRU enable reliable forecasting under input-limited, meteorological-only conditions while ensemble ML provides superior accuracy when water level observations are available, offering actionable guidance for flood forecasting in data-constrained environments.
- 1 May 2026
- Published by IWA Publishing in Journal of Water and Climate Change
- Vol. 17 (5) , 1329-1357
Open AccessRiver water level forecast sensitivity to hydrometeorological input variable selection: a comparison of machine and deep learning models in Bangladesh
Conceptual summary showing LSTM and GRU outperform ML models under meteorological-only inputs, while RFR and LGBMR dominate when lagged water level data are available, across five scenarios in the Old Brahmaputra River, Bangladesh. Accurate river water level forecasting is critical for flood management in Bangladesh's monsoon-dominated deltaic environment, yet optimal model selection under varying data availability remains unexplored. This study systematically compared six machine learning (ML) and deep learning (DL) algorithms, specifically linear regression, support vector regression, random forest regression (RFR), light gradient boosting machine regression (LGBMR), long short-term memory (LSTM), and gated recurrent unit (GRU), for water level prediction at Islampur station using 25 years of daily hydrometeorological data (1999–2024) across five input scenarios employing 5-day lag configurations. DL architectures dominated meteorological-only scenarios, with LSTM achieving R2 = 0.8809, explained variance score (EVS) = 0.8874, and symmetric mean absolute percentage error (SMAPE) = 4.83%, confirming unbiased symmetrical predictions. ML algorithms excelled with autoregressive inputs, where LGBMR achieved R2 = 0.9960 and RFR attained R2 = 0.9984; the latter's EVS = 0.9984 and SMAPE = 0.40% confirmed purely stochastic residuals. Principal component analysis (PCA)-based composite scoring of nine complementary metrics provided objective model rankings. Zero-shot spatial transferability at Sarishabari confirmed reproducible hierarchies without retraining, with violin plots and PCA composite scoring corroborating near-zero prediction bias across all transferred configurations. Results demonstrate that LSTM and GRU enable reliable forecasting under input-limited, meteorological-only conditions while ensemble ML provides superior accuracy when water level observations are available, offering actionable guidance for flood forecasting in data-constrained environments.
- Open Access
- 1 March 2024
- Published by Institute of Advanced Engineering and Science in International Journal of Advances in Applied Sciences
- Vol. 13 (1) , 54-62
Performance assessment of routing protocols for campus area emergency delay-tolerant network
Handheld devices have recently become an indispensable part of our daily lives. These devices always require an end-to-end connection for proper message transmission. Message transmission will be halted if the connection fails. In this case, delay-tolerant networks (DTNs) are preferable. One of the most important applications of DTN is the transmission of emergency messages among rescue personnel in the aftermath of a disaster. Previously, we created a simulation map in the Rajshahi University (RU) campus area of Bangladesh and studied the performance of DTN routing protocols. This paper used our developed simulation map to evaluate the performance of DTN and social-based routing protocols for emergency message circulation in the RU campus area when traditional communication networks are unavailable. We conducted extensive experiments with the opportunistic network environment (ONE) simulator for evaluations. The performance analysis is based on the delivery ratio, average latency, transmission cost, and average hop count of each group when the message size and node density are changed. According to the simulation results, dLife outperforms all other routing protocols in terms of delivery ratio, while the Spray-and-Focus routing protocol outperforms all other performance metrics.
- 1 March 2024
- Published by Institute of Advanced Engineering and Science in International Journal of Advances in Applied Sciences
- Vol. 13 (1) , 54-62
Open AccessPerformance assessment of routing protocols for campus area emergency delay-tolerant network
Handheld devices have recently become an indispensable part of our daily lives. These devices always require an end-to-end connection for proper message transmission. Message transmission will be halted if the connection fails. In this case, delay-tolerant networks (DTNs) are preferable. One of the most important applications of DTN is the transmission of emergency messages among rescue personnel in the aftermath of a disaster. Previously, we created a simulation map in the Rajshahi University (RU) campus area of Bangladesh and studied the performance of DTN routing protocols. This paper used our developed simulation map to evaluate the performance of DTN and social-based routing protocols for emergency message circulation in the RU campus area when traditional communication networks are unavailable. We conducted extensive experiments with the opportunistic network environment (ONE) simulator for evaluations. The performance analysis is based on the delivery ratio, average latency, transmission cost, and average hop count of each group when the message size and node density are changed. According to the simulation results, dLife outperforms all other routing protocols in terms of delivery ratio, while the Spray-and-Focus routing protocol outperforms all other performance metrics.