At present, the number of children with autism spectrum disorders in China is increasing year by year. In order to improve the training effect of ASD children's intervention in APP and the efficiency of rational decision-making in interface design, a method of ASD children's needs analysis based on Kano-entropy weight method is proposed. First of all, based on the Kano model, the needs of ASD children's intervention in APP are collected and divided into three categories (basic type, expectant type and excitable type) according to relative satisfaction coefficient. Secondly, the objective weight value of each demand index is calculated by entropy method. Then, combined with the correlation between weight value and ASD child intervention training, case design practice is carried out, and human-computer interaction test is used to verify whether the intervention APP based on ASD needs analysis can effectively play the role of intervention in APP. The interactive design of ASD children's intervention APP integrating Kano-entropy weight method divides the demand indicators into three dimensions:content, operability and vision, and the content demand accounts for the largest proportion. On this basis, the human-computer interaction test is used to verify the intervention effect of APP. It is found that the improved APP can enrich the interface content, improve the intervention operation, and enhance the visual attraction. The intervention APP interaction design based on ASD child needs analysis can enhance indicative performance, extend training duration and enhance the effectiveness of intervention training by optimizing content, operation and visual needs, so as to provide reference and basis for subsequent related design.