Applying machine learning approach to study the willingness to pay for irrigation water

Abstract

   In order to study the water user behaviors, most of the previous studies have been using the traditional statistical approach, however, the development of model analytical techniques like machine learning approach which will help analysts to get more results. In this article, both the traditional statistical analysis and machine learning (ML) approach is used to study the behavior of irrigation water consumption with a logistic regression model. 235 households in Nam Dinh, Thai Nguyen, and Phu Tho province were surveyed. The research results show that the regression coefficients between the two approaches are quite similar. However, by the machine learning approach, more specific research results are analyzed such as assessing the probability of water users for increasing the water charge, and the accuracy of the models. In other words, the machine learning approach offers a prognosis better compared to the traditional statistical approach.