ENHANCED MODEL TO HANDLE DATA SENSITIVITY FOR CROP YIELD PREDICTION USING MACHINE LEARNING TECHNIQUES

Abstract

The purpose of this study is to gather and analyze data on temperature, rainfall, soil, seed, crop productivity, humidity, and wind speed (in select regions) in order to assist farmers in increasing agricultural yield. Prior to using the proposed architecture, evaluates and processes the massive volume of data, pre-process the data in a Python environment. Second, Dense Region k-Means clustering (DRk-M) is applied to the outcomes of Proposed, yielding an average accuracy result for the data. Additionally, the harvests have been predicted by a self-created recommender system and shown on a graphical user interface created in a Flask environment. The recommended crops of additional states can be found in a similar way in the future thanks to the scalable system design.