The Intelligence Decision Model Through Improved Differential Privacy

In recent years, with the vigorous promotion and support of national policies, the science and technology innovation enterprises are booming. They have actively responded to the country’s call and constantly pursued new technologies, business formats, and models, creating a good atmosphere of "mass entrepreneurship and innovation". The development of science and technology innovation enterprises still faces endless problems, such as irrational allocation of funds and lack of innovative talents. Reasonable strategic intelligence decisions are crucial for science and technology innovation enterprises. However, due to science and technology innovation enterprise data’s professionalism, privacy, and sensitivity, it is difficult to share the data publicly, resulting in data decision model analysis and stagnant development. Therefore, to solve the data sharing problem, this article proposes an improved differential privacy based on information entropy and constructs a data query model. To assess the effectiveness of improved differential privacy, experiments are conducted using the Kaggle dataset and the personal information data of the employees, and error metrics are selected for evaluation, the results prove that the proposed method is strongly competitive.