← Back to Publications List

Leveraging Machine Learning and Data Analytics for Predictive Modeling in Smart Agriculture: A Case Study from Northern Bangladesh

Students & Supervisors

Student Authors
Naimul Islam Hridoy
Bachelor of Science in Computer Science & Engineering, FST
Supervisors
Md. Mortuza Ahmmed
Associate Professor, Faculty, FST

Abstract

Background: Agriculture in Bangladesh faces significant challenges due to unpredictable weather, resource constraints, and limited access to real-time data. Smart agricultural practices empowered by Machine Learning (ML) and Data Analysis have the potential to enhance productivity and sustainability. In Northern Bangladesh, particularly in the Dinajpur region, applying predictive analytics can support informed decision-making for farmers and policymakers. Methodology: This study begins with a trend analysis of key agricultural and climatic indicators using publicly available data from 2015 to 2021. Variables such as rice, wheat, and maize production, annual rainfall, average temperature, and cultivated land area were analyzed using Microsoft Excel and SPSS to identify temporal patterns and correlations. To enhance predictive capability, a hybrid machine learning framework combining Random Forest and Long Short-Term Memory (LSTM) networks is under development, targeting future crop yield forecasting using region-specific datasets. Results: The trend analysis from 2015 to 2021 reveals a consistent increase in rice, wheat, and maize production across Northern Bangladesh. Rice production rose from 12,000 tons to 14,200 tons, while maize experienced the highest growth among the three. The amount of cultivated land expanded steadily, indicating intensified agricultural practices. Simultaneously, average annual rainfall remained variable with a slight upward trend, and the average temperature increased gradually from 26.1°C to 26.8°C. These findings reflect a correlation between climatic changes and agricultural yield. Conclusion: The application of ML in data-driven agriculture holds promise for regions like Dinajpur, enabling precision farming through reliable forecasting and intelligent resource allocation. Ongoing work aims to enhance the model’s robustness by integrating IoT sensor data and extending its application to other agro climatic zones. This research highlights the importance of interdisciplinary collaboration in achieving digital transformation in agriculture. "

Keywords

Smart Agriculture Machine Learning Data Analytics Climate Change Crop Yield Prediction Bangladesh Trend Analysis Time Series Modeling

Publication Details

  • Type of Publication: Conference 
  • Conference Name: IEEE CSBDC Summer Symposium 2025
  • Date of Conference: 18/07/2025 - 18/07/2025
  • Venue: Hajee Mohammad Danesh Science and Technology University, Dinajpur
  • Organizer: IEEE Computer Society Bangladesh Chapter
View on Publisher