Abstract: There are two distinguishing characteristics for modern decision making problems in comparison with the traditional situation: one is the availability of large amount of original data emerging with the development of system automation technology; the other is the complexity and uncertainty underlying the real-life decision problems, which make it infeasible to establish precise models. Traditional model-based decision making methodologies are inefficient under this circumstance. Therefore, a number of research works have been conducted on data-based decision making methodologies. This paper reviews the prevalent data-based decision making methodologies from three aspects based on the characteristics of the considered decision problems: classification methodology, decision analysis methodology, and optimization methodology. The characteristics, development history, and perspective are summarized for each specific methodology.