The prediction of early adopters has been a challenging research problem for agricultural economists and rural sociologists. This study identifies the characteristics of pig farmers in Korea who have adopted MIS(management information systems) systems relative to non-adopters. Along with a most commonly used statistical method(logistic regression) for agricultural technology adoption. We used popular machine learning algorithm(decision trees) to develop the advanced prediction method. The performances of two methods are compared for accuracy, sensitivity, specificity, precision, and response rate. The results indicate that two methods identify similar factors for early adopters. As farm size, business complexity and business specialization increase, so does the probability of MIS adoption. Older farmers are less likely to adopt MIS, and education level is positively associated with MIS adoption. Finally, logistic regression produces higher prediction performance than decision tree.