Embedded Files
Conference Proceedings
(Published: 11 | Presented: 2)
Conference Proceedings
(Published: 11 | Presented: 2)
Advances in Machine Learning for Crop Yield Prediction: A Comprehensive Review of Techniques, Trends, and Challenges
Advances in Machine Learning for Crop Yield Prediction: A Comprehensive Review of Techniques, Trends, and Challenges
Authors: A.M Asik Ifthaker Hamim, ; Abdul Mohaimin ; Al Fahim Ishmum ; Rashedul Arefin Ifty ; M. Jamshed Alam Patwary
Presented in: 2025 4th International Conference on Electrical, Computer and Communication Engineering (ECCE)
DOI: 10.1109/ECCE64574.2025.11013031
Publisher: IEEE
Link: https://ieeexplore.ieee.org/abstract/document/11013031
Research keywords: Crop Yield Prediction, Machine Learning Algorithms, Agricultural Data Analysis, Predictive Modeling, Sustainable Agriculture
Abstract: Addressing food security amidst climate change and population growth requires precise crop yield prediction models that surpass the limitations of traditional approaches. Machine learning has emerged as a powerful tool in this domain, utilizing comprehensive data from soil metrics, climatic variables, and crop characteristics to enhance forecasting accuracy. This review examines the application of ML algorithms such as Random Forests, Gradient Boosting Decision Trees, and Artificial Neural Networks for crop yield prediction. Out of 200 articles initially reviewed from major databases, 47 were rigorously selected based on relevance and methodological rigor. The findings provide an in-depth analysis of ML’s efficacy in predicting yields across diverse agricultural environments, identifying trends, key challenges, and limitations related to data heterogeneity, regional specificity, and interpretability. This paper aims to guide researchers in advancing ML-driven yield prediction, particularly in integrating IoT technologies and improving model generalizability, thus supporting sustainable agriculture and precision farming.
Empowering Developing Nations: Sustainable Launch Approaches and Emerging Technologies for Small Satellites
Empowering Developing Nations: Sustainable Launch Approaches and Emerging Technologies for Small Satellites
Authors: A.M Asik Ifthaker Hamim, ; Abdullah Al Abrar Chowdhury ; Sazzad Hossen Chowdhury ; Amad Uddin Osama ; Anik Chowdhury ; Mohammed Moin Uddin ; Rashedul Arefin Ifty
Presented in: 2025 4th International Conference on Electrical, Computer and Communication Engineering (ECCE)
DOI: 10.1109/ECCE64574.2025.11013107
Publisher: IEEE
Link: https://ieeexplore.ieee.org/abstract/document/11013107
Research keywords: Small Satellite Constellations, Sustainable Space Technologies, Lunar Atmosphere Utilization, Space Ecosystem Development
Abstract: As humanity ventures deeper into the cosmos, the demand for innovative and sustainable solutions to address the complexities of space exploration continues to grow. This study investigates a wide range of strategies and technologies aimed at overcoming the challenges faced by small satellites, which are pivotal to this evolving landscape. Leveraging immersive mission-spanning visualization, the research identifies opportunities to optimize mission planning and readiness. Small satellite constellations are presented as an effective approach to enhance coverage, reliability, and cost efficiency. Furthermore, the potential of utilizing the lunar atmosphere is explored as a means to reduce launch mass and support sustainability. A philosophical perspective on the evolution of future space networks further enriches the discourse, providing insight into the long-term dynamics of space infrastructure. Amidst the rapid expansion of the small satellite industry, the study underscores the critical importance of collaboration and innovation in addressing these challenges, fostering a resilient and responsible space ecosystem, particularly for developing nations.
Potato Crop Yield Prediction: A Data-Driven Federated Learning Approach
Potato Crop Yield Prediction: A Data-Driven Federated Learning Approach
Authors: Rashedul Arefin Ifty ; Afif Hossain Irfan ; Md. Ismail ; M. Jamshed Alam Patwary
Presented in: 2024 27th International Conference on Computer and Information Technology (ICCIT)
DOI: Will be published soon
Publisher: IEEE
Link: Will be published soon
Research keywords: Yield Prediction, Federated Learning, Machine Learning, Precision Agriculture
Abstract: Accurate prediction of potato yield is crucial for optimizing agricultural productivity and resource management in Bangladesh, where potatoes play a key role in the economy. This study explores the application of Federated Learning (FL) to predict potato yields across six major agricultural districts in Bangladesh: Chittagong, Dhaka, Dinajpur, Jessore, Mymensingh, and Rajshahi. Unlike traditional centralized models, FL enables decentralized training on local data, ensuring data privacy while capturing region-specific agricultural patterns. A diverse set of machine learning models was evaluated, including Linear Regression, Random Forest, Gradient Boosting, SVR, XGBoost, Extra Trees, AdaBoost, K-Nearest Neighbors, Decision Tree, and LightGBM, with extensive hyperparameter tuning conducted for each district to optimize performance. The local models were aggregated using the Federated Averaging (FedAvg) algorithm, enhancing predictive accuracy through knowledge integration from diverse districts. The evaluation was based on metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and the coefficient of determination (R²). The findings demonstrated that the federated global model achieved an overall MSE of 0.0078 for the validation set and an R² of 0.8794, with the lowest district-level MSE observed at 0.0052. The results demonstrate that Federated Learning significantly improves yield prediction while preserving data privacy. This approach provides a scalable, privacy-preserving solution for integrating advanced analytics into agricultural planning, benefiting policymakers and farmers alike. The study's findings also have broader implications for applying FL in precision agriculture globally.
A Novel Image-Based Machine Learning Approach for Feature-Driven Plant Disease Detection
A Novel Image-Based Machine Learning Approach for Feature-Driven Plant Disease Detection
Authors: Rashedul Arefin Ifty ; Afif Hossain Irfan ; Taivan Reza Dipta ; Md. Sorowar Mahabub Rabby ; Md. Ismail
Presented in: 2024 27th International Conference on Computer and Information Technology (ICCIT)
DOI: Will be published soon
Publisher: IEEE
Link: Will be published soon
Research keywords: Image Processing, Machine Learning, Ensemble Techniques, Image Classification, Agriculture.
Abstract: Plant disease detection significantly impacts precision agriculture by enabling early and accurate disease identification, as well as timely and targeted treatments, resulting in substantial reductions in crop losses and optimized resource utilization. Concerning plant disease detection, deep learning methods frequently encounter limitations stemming from their high computational costs and exacting technical requirements. Therefore, this study utilizes machine learning techniques over deep learning to reduce costs and effectively manage the technical complexities involved. This research introduces a novel Stacking Ensemble Classifier model for enhancing the accuracy of Apple leaf disease diagnosis. By integrating four base classifiers—Random Forest, Gradient Boosting, XGBoost, and LGBM—the Stacking Ensemble Classifier achieves notable improvements in prediction accuracy through robust image preprocessing and extensive hyperparameter tuning. The study leverages a dataset of over 5700 Apple leaf images sourced from PlantVillage. Results demonstrate that the proposed model achieved superior performance, surpassing individual existing machine learning classifiers with an outstanding accuracy of 98.44%. This approach maintains high detection rates while reducing computational overhead, outperforming other existing classifiers in accuracy as well as significant performance parameters comprising precision, recall, and F1-score.
Crop Prediction Using Machine Learning and IoT: A Comparative Analysis of Algorithms
Crop Prediction Using Machine Learning and IoT: A Comparative Analysis of Algorithms
Authors: Aishwarya Roy: Afif Hossain Irfan; Mukut Protim Memo ; Labib Bin Shahed ; Rashedul Arefin Ifty
Presented in: 2024 International Conference on Recent Progresses in Science, Engineering and Technology (ICRPSET)
DOI: 10.1109/ICRPSET64863.2024.10955896
Publisher: IEEE
Link: https://ieeexplore.ieee.org/document/10955896
Research keywords: IoT, Machine Learning, Precision Agriculture.
Abstract: Agriculture is a cornerstone of Bangladesh’s economy, yet it faces significant challenges due to weather fluctuations, including changes in temperature, humidity, and precipitation. These environmental variables critically affect farming activities, demanding constant adaptation. To mitigate these challenges, we propose an IoT-based model designed to enhance agricultural data collection and provide valuable insights to farmers. Our model integrates various sensors, such as the ESP8266-01 microcontroller, NPK sensor, DHT22 humidity sensor, DS18B20 temperature sensor, pH sensor, and water level sensor, all housed in a protective enclosure and powered by a DC source. This system collects real-time data on critical environmental parameters, forming the foundation of predictive models. We employ multiple machine learning algorithms, including Logistic Regression, Naive Bayes, Support Vector Machine, K-Nearest Neighbours, Decision Tree, Random Forest, Bagging, AdaBoost, Gradient Boosting, and Extra Trees. Using a dataset containing parameters like nitrogen, phosphorus, potassium, temperature, humidity, pH, and rainfall, we split the data into training and testing sets. Models are evaluated based on accuracy, precision, recall, F1 score, and undergo 10-fold cross-validation to assess performance comprehensively. The integration of IoT-derived data with machine learning models helps identify the most suitable crops based on soil and climate conditions. By offering real-time, actionable insights, our system enables farmers to make better crop selection decisions. This approach aims to enhance agricultural productivity in Bangladesh by optimizing weather-dependent farming strategies and ensuring more informed agricultural practices.
Project AgriSage Tech: Federated Learning-Driven Agricultural Innovations for Precision Agriculture
Project AgriSage Tech: Federated Learning-Driven Agricultural Innovations for Precision Agriculture
Authors: Rashedul Arefin Ifty
Presented in: 2024 IEEE 12th Region 10 Humanitarian Technology Conference (R10-HTC), Kuala Lumpur, Malaysia
DOI: 10.1109/R10-HTC59322.2024.10778905
Publisher: IEEE
Link: https://ieeexplore.ieee.org/document/10778905
Research keywords: Federated Learning, Crop Yield Prediction, Soil Moisture Estimation, Agricultural Disease Detection
Abstract: Agriculture in Bangladesh faces numerous challenges, including suboptimal crop selection, disease susceptibility, frequent natural disasters, fluctuating market prices, budgetary mismanagement, and inadequate soil classification. To tackle these issues, a mobile application has been developed under this project, designed to revolutionize traditional farming practices through the use of advanced artificial intelligence (AI) technologies. This application integrates decades of agricultural data with Federated Learning to improve accuracy in crop prediction and recommendations. It features real-time disease and soil detection capabilities, similar to Google Lens, allowing for quick identification and resolution of agricultural problems. Additionally, the application includes a proprietary budget distribution algorithm that offers customized financial planning based on crop type, geographical area, and location. Weather forecasts and SMS-based natural disaster alerts help mitigate the effects of environmental disruptions, ensuring farmers are better prepared. To address the digital divide in rural areas, the application offers limited functionalities via a USSD interface, allowing access for non-smartphone users. This comprehensive AI-powered solution aims to enhance agricultural productivity, promote sustainability, and support farmers in making informed decisions, aligning with the goals of sustainable development.
Forecasting Geomagnetic Storms for Earth Using LSTM with Solar Wind and IMF Features
Forecasting Geomagnetic Storms for Earth Using LSTM with Solar Wind and IMF Features
Authors: Nusrat Jahan Shammi ; Taqia Ashraf ; Rashedul Arefin Ifty ; Md. Ziaur Rahman
Presented in: International Conference on Innovations in Science, Engineering and Technology (ICISET 2024)
DOI: 10.1109/ICISET62123.2024.10939246
Publisher: IEEE
Link: https://ieeexplore.ieee.org/document/10939246
Research keywords: Geomagnetic Storms, Interplanetary Magnetic Field, Plasma, Disturbance Storm-Time Index, Solar Wind, ACE, DSCOVR.
Abstract: When the solar wind interacts with Earth's field, it can cause geomagnetic storms, which pose serious risks to vital infrastructure such as satellite communication, GPS systems, and electric power transmission. We presented a novel method for forecasting the Disturbance Storm-Time Index (Dst) using a long-short-term memory (LSTM) neural network in order to meet the urgent requirement for precise predictions. Our model employs sensing data from NASA's ACE and NOAA's DSCOVR satellites to analyze the complex relationship between interplanetary magnetic field \& plasma from solar wind \& sunspot activity. These factors are crucial to understanding geomagnetic disturbances. The LSTM model accumulates complex patterns and temporal relationships within the space weather data by including real-time data assimilation, resulting in enhanced weather prediction capabilities. Evaluation metrics like the root mean square error (RMSE) and coefficient of determination (\( R^2 \)) are used to assess the model's performance and reliability. The outcomes indicate that the LSTM model is highly effective in accurately predicting geomagnetic storms. The predicted values offer vital information for operators of satellites, power grids and magnetic navigation systems, enabling them to implement preemptive actions to mitigate the effects of any disruptions caused by geomagnetic storms. The research we accomplished helps to improve the approaches employed for forecasting space weather. This provides decision-makers with a useful and timely tool to strengthen vital systems and prepare for geomagnetic disturbances.
NobleMeta: A Noble Technique to Predict Potato Crop Yields in Bangladesh
NobleMeta: A Noble Technique to Predict Potato Crop Yields in Bangladesh
Authors: Rashedul Arefin Ifty ; A.M Asik Ifthaker Hamim ; Abdul Mohaimin ; Al Fahim Ishmum ; M. Jamshed Alam Patwary
Presented in: ACM 3rd International Conference on Computing Advancements (ICCA 2024)
DOI: 10.1145/3723178.3723255
Publisher: ACM
Link: https://dl.acm.org/doi/10.1145/3723178.3723255
Research keywords: Yield Prediction, Machine Learning, Ensemble Method
Abstract: Agriculture in Bangladesh frequently encounters hurdles due to reliance on traditional practices and limited access to cutting-edge technologies. Conventional predictive models may struggle to accurately forecast crop yields due to the complex interplay of various factors such as weather patterns, soil conditions, also agricultural practices. To address the aforementioned problems, the NobleMeta algorithm for predicting potato crop yield is presented. Leveraging meticulously curated datasets sourced from historical weather data and agricultural records obtained from authoritative sources, NobleMeta integrates five state-of-the-art machine learning models, including CatBoost Regressor, XGBoost Regressor, LightGBM Regressor, Gradient Boosting Regressor and Bayesian Ridge Regressor. This fusion results in unparalleled accuracy and robustness in crop yield prediction. Through rigorous experimentation and evaluation, NobleMeta demonstrates exceptional accuracy, with an average R-squared score of over 95%, surpassing conventional models. The study further emphasizes the significance of dataset curation process in guaranteeing the accuracy and dependability of the input data. NobleMeta's superior performance represents a noteworthy breakthrough in agricultural predictive modelling, providing insightful information for Bangladesh's food security, productivity, and crop production optimization.
Fault-Tolerant Space Weather Prediction: Leveraging Raw DSCOVR Data with Long Short-Term Memory in Machine Learning
Fault-Tolerant Space Weather Prediction: Leveraging Raw DSCOVR Data with Long Short-Term Memory in Machine Learning
Authors: Sara Karim ; Rashedul Arefin Ifty ; Takia Sultana Nova ; Anika Rehnum Ema ; Shahriar Rahman ; Khondker Omar Anwar
Presented in: 2024 IEEE International Conference on Space Mission Challenges for Information Technology (SMC-IT), Mountain View, California, USA.
DOI: 10.1109/SMC-IT61443.2024.00012
Publisher: IEEE
Link: https://ieeexplore.ieee.org/document/10795050
Research keywords: Space Weather Forecasting, Geomegnatic Strom, Semi-Supervised Machine Learning
Abstract: This paper introduces the groundbreaking Helios-LSTM algorithm for precision forecasting of solar wind activity, addressing the urgent need for accurate predictions. Leveraging data from NASA's Solar Wind, Solar Radiation (CME), and Geomagnetic Storm APIs, the model achieves an unprecedented 94% accuracy rate by meticulously integrating information. Through sophisticated data preprocessing techniques and imputation strategies for missing values, the algorithm utilizes hourly features from solar wind data to capture nuanced temporal dependencies. Its architecture includes Bidirectional LSTM and GRU layers for comprehensive feature transformation, outperforming existing methods in predicting solar wind patterns over varying time intervals. Moreover, the model forecasts the Disturbance Storm-Time Index (Dst) by analyzing interplanetary magnetic fields, solar wind plasma, and sunspot activity, demonstrating efficacy in predicting geomagnetic storms. Evaluation metrics confirm the model's effectiveness, offering invaluable insights for satellite operators, power grid managers, and navigation systems, while also laying the groundwork for safeguarding Earth against disruptive geomagnetic impacts. This research signifies a significant advancement in space weather forecasting, providing decision-makers with a robust tool to mitigate risks posed by geomagnetic disturbances.
Enhancing Precision Agriculture with Machine Learning & IoT: SVR and DTR Comparative Analysis
Enhancing Precision Agriculture with Machine Learning & IoT: SVR and DTR Comparative Analysis
Authors: Rashedul Arefin Ifty; Afif Hossain Irfan; MD. Ismail ; Mohammad Abu Huraira Saim.
Presented in: 2024 3rd International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE)
DOI: 10.1109/ICAEEE62219.2024.10561658
Publisher: IEEE
Link: https://ieeexplore.ieee.org/document/10561658
Research keywords: Precision Agriculture, Regression, Crop Yield Prediction, IoT
Abstract: The paper explores the declining significance of agricultural revenue and employment in nations like Bangladesh, attributing it to poor crop selection due to a lack of accurate agricultural information and manual data reliance. Employing multiple regression methods, it investigates the efficacy of agricultural production prediction, particularly focusing on integrating machine learning with IoT for better estimation. By analyzing data spanning 43 years from the Bangladesh Bureau of Statistics and the Bangladesh Meteorological Department, with a focus on Aman rice, the study identifies seven environmental characteristics crucial for crop selection. It proposes an IoT model for data collection and utilizes Decision Tree Regression for crop selection and yield prediction, aiming to enhance predictability, mitigate food issues, and empower farmers with informed choices.
Authors: Arfanul Islam; Rashedul Arefin Ifty; Mohammad Abu Huraira Saim; Junaid Al Mahin; Md. Fahim Nizamee; Khaled Eabne Delowar; Muhammed J. A. Patwary
Published in: 2023 26th International Conference on Computer and Information Technology (ICCIT)
DOI: 10.1109/ICCIT60459.2023.10441217
Publisher: IEEE
Link: https://ieeexplore.ieee.org/abstract/document/10441217
Research keywords: Precision Agriculture, Crop Yield Prediction, Machine Learning, IoT Integration, Environmental Factors
Abstract: The countries of the Indian subcontinent are indeed very reliant on agriculture for their daily necessities. Among that country, the majority of agricultural production in Bangladesh is categorized as traditional subsistence farming. Rice, wheat, corn, legumes, fruits, vegetables, meat, fish, seafood, and dairy products are among the agricultural products manufactured in Bangladesh. The main staple food of Bangladesh is rice. The scarcity of arable land and natural resources emphasizes the significance of creating innovative agricultural technologies to improve productivity and satisfy future demand. Precision agriculture is a difficult task to execute. In this study, we used meteorological data obtained from the Bangladesh Bureau of Statistics (BBS) and Bangladesh Meteorological Department from more than 8 districts in Bangladesh over 45 years. The primary objective of this study is to create a one-of-a-kind machine learning model utilizing seven environmental factors. Our model also improves forecast accuracy of the best qualities for overcoming hunger challenges. Based on our collected Aus rice dataset, we utilized voting regression (VR) with a novel combination of MLA. The VR’s concept is to integrate MLA and return the average anticipated values. Compared with other MLA, our proposed algorithm achieved the highest R^2 of 0.8928 in a study on Aus rice yield prediction. In addition, we proposed a self-designed IoT device to automatically gather data from agricultural fields and thereby increase crop production forecasts by incorporating the data into our proposed model. The proposed system will indeed be immensely helpful to a country’s agro-economic advancement.
Authors: Arfanul Islam; Imranul Khair; Sakawat Hossain; Rashedul Arefin Ifty; Muhammed Nazmul Arefin; Muhammed J. A. Patwary
Published in: 2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)
DOI: 10.1109/ECCE57851.2023.10101585
Publisher: IEEE
Link: https://ieeexplore.ieee.org/document/10101585
Research keywords: EMLA, Crop selection, Yield prediction
Abstract: This research presents an Ensemble Machine Learning Approach (EMLA) for crop selection and yield prediction in Bangladesh. With decreasing importance of agriculture, it is important to have accurate predictions for crop output. The study uses three major crop data and seven weather parametrized data from the last 43 years to create a comprehensive dataset for analysis. The EMLA is developed using Catboost Regressor and XGBoost Regressor with their novel combination of Machine Learning Algorithms. It is then compared with eight well-known machine learning algorithms to establish its superiority. The proposed EMLA achieved high accuracy with R-squared scores of 88.084%, 91.776%, and 90% respectively for Aus rice, Aman rice, and Potato. The primary goal of this research is to improve predictability for overcoming food difficulties and create an intelligent information prediction analysis on farming in Bangladesh for efficient and profitable farming decisions. The results show that the EMLA technique improves the output and prediction by relying on the strong performance of another model.
Authors: Arfanul Islam; Imranul Khair; Rashedul Arefin Ifty; Sakawat Hossain; Muhammed Nazmul Arefin; Faisal Bin Al Abid
Published in: 2022 International Conference on Recent Progresses in Science, Engineering, and Technology (ICRPSET)
DOI: 10.1109/ICRPSET57982.2022.10188538
Publisher: IEEE
Link: https://ieeexplore.ieee.org/document/10188538
Research key words: CatBoost Regressor, Crop Prediction, Regression Methods.
Abstract: A recent study in Bangladesh has investigated the use of machine learning algorithms to improve crop prediction accuracy for farmers. The study compared the performance of various regression algorithms, including Catboost Regressor, Gradient Boosting Regression, Kernel Ridge Regression, Elastic Net Regression, Bayesian Ridge Regression, Linear Regression, LGBM Regressor, XGBoost Regressor, Stochastic Gradient Descent Regression, and Support Vector Machine. The study focused on three major crops in Bangladesh, Aus rice, Aman rice, and potato, and considered seven environmental factors such as temperature, rainfall, relative humidity, wind speed, cloud cover, and bright sunshine. The results showed that the Catboost Regressor Algorithm was the most effective for crop prediction. With the use of these algorithms, farmers in Bangladesh could potentially improve their precision farming skills, increase crop yield, and reduce the growing of undesirable crops.
Google Sites
Report abuse