Investment risk assessment based on improved BP neural network

In general, the risk assessment and discussion in the actual construction process of the project mainly adopts expert scoring evaluation method, case study method, questionnaire survey method and fuzzy comprehensive evaluation method. These methods cannot provide good reference and portability for other projects. This paper proposes to use the improved BP neural network to explore the intelligent evaluation of highway investment risk. Taking the risk management data of highway engineering investment from the Second Bureau of China Communications as the research object, the differential evolution algorithm is used to improve the BP neural network model, and the existing highway investment risk data training model is used to realise the intelligent evaluation of the investment risk of new projects. The comparative experiment shows that the accuracy of the optimised evaluation model is improved by 10%, which can provide risk decision-making services for highway investment projects in China.