In order to solve the problem of extracting product ideas and updating the ranking of ideas in real time in the mass collaboration mode, and to realize dynamic decision-making of product ideas, firstly, according to the needs of enterprises, the generation method of product ideas is proposed according to brainstorming method. Secondly, the participants and evaluation index system of involved in evaluation are classified, and the corresponding relationship between them is illustrated. Then, based on the Basic Belief Assignment (BBA) function, the extraction mechanism of product information is constructed, and the static integration process is constructed according to the rule of Evidence Reasoning (ER). On this basis, the method and steps of dynamic intelligent recommendation of product ideas in the mass collaboration mode are deduced. Finally, HUAWEI is taken as an example to simulate the specific operation process of the proposed method and verify the effectiveness of the proposed method. The innovation is to allow participants to infer only part of the indicators, and fully consider the impact of the weights of indicators and participants on the fusion results. It improves the innovation efficiency of products while maintaining the authenticity of the original data.