According to the competition data set automatic processing system and method based on multiplexing. Through the technical scheme, original data sets can be automatically arranged and divided and uploaded to a competition website server. The method comprises the steps that an original data set is automatically sorted, segmented and uploaded, and a competition website data set is generated, so thatthe problems that in the prior art, codes cannot be reused, sorting efficiency is low, and segmented sub-data set samples do not accord with probability density distribution of the whole data set samples due to manual sorting and segmentation of the data set are solved. According to the method, stratified sampling can be performed on the data with the annotated attributes as categories, the annotated attributes as numerical values and multi-attribute annotations, so that the sampling accuracy is improved, and the data distribution in the segmented sub-data set better conforms to the data distribution of the original data set.