A new method for filling missing values by gray relational analysis

In Data Mining and Machine Learning, the missing attribute will have a negative impact on the learning results. The filling of missing values is a very challenging work. In this paper, a new algorithm based on gray relational analysis is presented, which takes the differences of the relationships between the properties into account. When calculating the gray relational grade, the weights of attributes will be considered. The experimental results demonstrate that this method performs well when filling the discrete missing values.