An Efficient Analysis Model for Growth Environment Information System using Multi Regression and Modified Kmeans Algorithms

Abstract

This paper set out to propose an analysis model to make various uses of Big Data to be measured through the expansion of IoT agricultural technology and crop cultivation. Certain elements of crop cultivation environments were extracted by designing a model for the extraction to combine principal component analysis and K-means algorithm and an analysis model to predict and assess crop quality and yield according to the extracted elements. The data of mushroom cultivation was used in the experiment. In the extraction of elements according to cultivation environments, elements were distinguished that had big impacts on the mushroom cultivation environment. The error range was within approximately 5% between the quality of mushrooms that had been harvested and the prediction value of mushroom quality based in the analysis model, which indicates that the analysis model produced similar results to the actual predictions.