機械学習による遊休農地所有者の意向予測 一千葉県勝浦市と茨城県水戸市の農地情報から一

When agricultural policymakers promote the accumulation of idle farmland to leading farmers, it is more important to predict the intentions of landowners (i.e., intentions to sell, rent, or cultivate) with high accuracy than to elucidate structural factors. In this study, we attempted to solve this problem by specifying a support vector machine (SVM) based on agricultural land information systems in two regions of the Tokyo metropolitan area. As a result, the SVM showed high prediction accuracy on the training data set and the test data set. [ABSTRACT FROM AUTHOR]