Recently years, multiple attribute online evaluation information with a large data volume emerges from some websites, and online evaluation information often presents in the form of discrete random distributions. How to determine attribute weights and rank alternatives based on online evaluation information, it is a new noteworthy research issue. This paper proposes a method for determining attribute weights and ranking alternatives based on online evaluation information. In this method, online evaluation information with respect to each attribute is first described as the form of discrete probability distribution function, and the decision matrix about weighted cumulative distribution functions is constructed. Then, based on the decision matrix, optimization model is constructed to determine attribute weights by defining the ideal cumulative distribution vector and the distance between each alternative and the vector. Further, by solving the optimization model, the weight of each attribute can be obtained. On the basis of this, the ranking of alternatives can be determined through calculating the ranking value of each alternative. Finally, based on online evaluation information for automobile products from the website ‘autohome.com.cn’, a case analysis is given to illustrate the application of the proposed method.