vioft2nntf2t|tblJournal|Abstract_paper|0xf4ffdf402c000000ecd60b0001000300
In situations where vast amounts of data are being gathered and published, machine learning is a potential solution. In order to estimate large-scale agricultural yields, we applied machine learning techniques to agronomic principles. A workflow that emphasises consistency, modularity, and reusability serve as a baseline for the project. To ensure accuracy, we worked to create predictors or traits that could be explained and then used machine learning without leaking any information. MCYFS data from the weather, remote sensing, and soil sensors was used to generate new functionalities. Smaller configuration adjustments allow us to handle many different crops and countries with our modular and reusable work flow design. Standard input data and the methodology can be utilised for repeatable tests with repeatable outcomes. It is from these findings that we may go on to refine our algorithms even further.