Crop Recommendation System Using LightGBM

Purpose:The present study is an attempt to study the farmers’ challenge’s like selecting the best crop for their agricultural site, and measure the impact of precision agricultureas a solution for farmers and most suited to their environment through historical data about soil type and nutrient levels. Design/Methodology/Approach: To test the research framework and data set, a machine-learning-based ensemble method that recommends crops to grow on an agricultural site using the light gradient boosting machine(LightGBM) algorithm has beenconsidered to achieve higher accuracy and efficiency in recommending a crop at the site. Findings:Crop recommendationindeed plays a vital role in agriculture for the farmers. The study also found that the LightGBM ensemble machine learning algorithm which produces a series of hypotheses that are then compiled into a final output that maximizes the predictive accuracy of the classification. This study also compared the accuracy, execution time of LightGBM with other algorithms like Adaboost, Gradientboost, Xgboost, Catboost, and found LightGBM far better than the others. Research Limitations:The study has several limitations. For instance, it engaged the theory of an ensemble machine learning technique to recom mend the crop to the farmers. Future, there is a need to expand the techni ques byintegrating them with pretrained and LLM models. Furthermore, the sample dataset was selected from an open source of 22 crops with 2200 records. Managerial Implications: Practically, it brings a focus on the ensemble techniques and theiravailable algorithms with sample dataset implications. The study, thus, showcases the implementation and comparison of an en semble technique for experimental purposes, knowledge, and competencies to increase with some real-time data in future. Originality/Value: The study highlighted the importance of ensemble techniques of machine learning supportedby the LightGBM algorithm in the agriculture domain