Date of Award
8-2024
Document Type
Project
Degree Name
Master of Science in Information Systems and Technology
Department
Information and Decision Sciences
First Reader/Committee Chair
Dr. Conrad Shayo
Abstract
To meet the increasing demand, for food while also reducing impact this study introduces an innovative "Integrated Crop Recommendation System" that combines advanced machine learning with sustainable farming methods. The goal of this system is to transform how crops are chosen by considering factors like soil quality, local climate and habitats for pollinators thereby enhancing the precision and effectiveness of crop suggestions. In contrast to agricultural decision support systems that often neglect the interconnectedness of soil health, weather conditions and biodiversity, this new approach aims to improve food security and sustainability. The primary research focus is on optimizing practices that support pollinators in environments. The research aims to provide farmers with enhanced guidance and deeper insights into the relationships among soil quality, weather patterns and ecological sustainability offering a solution for modern farming practices. The study encompasses a literature review, methodology development, data analysis, and discussion of findings. Outlines research directions. Research Questions are: Q1) How can incorporating pollinator-related data into machine learning models enhance the accuracy and efficiency of agricultural decision support systems for optimal crop recommendations? Q2) What kind of effect does incorporating practices to support pollinators have on the overall strength and durability of crop recommendations produced by the integrated machine learning model?
The Findings and Discussions for the questions are: Q1) In our research we. Evaluated a system, for suggesting crops based on machine learning. This system considers factors such as soil quality, weather conditions and reliance on pollinators by studying data sets related to crop recommendations and pollination. Our analysis of the data showed connections like the relationship between rainfall and crop production. Additionally, our decision tree model performed better than the SVM model, in predicting crop yields. Q2) The research shows that including methods to support pollinators in crop recommendations based on machine learning can improve their performance and durability. It stresses the importance of factoring in pollination aspects when making decisions. By grouping crops based on their reliance on pollinators it underscores the need for customized conservation approaches. Proves that taking types of pollinators into account greatly enhances the precision of predicting crop yields. Conclusions for each question are: Q1) Our study shows that using machine learning to examine the connections, among soil makeup, weather conditions and reliance on pollinators improves decision making tools, for agriculture. This in turn boosts the accuracy of crop recommendations. Helps ensure food security. Q2) By including actions that support pollinators in crop suggestions generated by machine learning it boosts their dependability and strength. This underscores the importance of efforts on conserving pollinators to enhance the resilience of crops and the overall health of ecosystems. Areas of further studies for each question are: Q1) The success of crop recommendation systems powered by machine learning models relies heavily on tuning hyperparameters and structural components than just sticking to preset or manually configured settings. Q2) Incorporating a variety of factors and crop specific characteristics into crop recommendation systems provides a grasp of growth elements essential for improving the accuracy of recommendations.
Recommended Citation
Saddikuti, Arun Kumar Reddy, "INTEGRATED CROP RECOMMENDATION SYSTEM: HARNESSING MACHINE LEARNING" (2024). Electronic Theses, Projects, and Dissertations. 2018.
https://scholarworks.lib.csusb.edu/etd/2018