Research on intelligent evaluation model of innovation and entrepreneurship projects based on big data analysis

With the surge in the number of innovation and entrepreneurship projects, traditional evaluation methods can no longer meet the needs of accurate evaluation. This paper constructs an intelligent evaluation model for innovation and entrepreneurship projects, and designs an improved hybrid attention graph convolutional network evaluation algorithm. 1,000 real project data are collected from public databases and industry platforms, and multi-dimensional information such as project teams and technological innovations are integrated to construct a comprehensive evaluation index system. The algorithm combines the multi-head attention mechanism with the self-attention mechanism to enhance the capture of key features; optimizes the graph convolutional network structure to improve the ability to mine project associations. The experiment uses accuracy, recall, F1 value and mean square error as indicators to compare with traditional BP neural networks and ordinary graph convolutional networks. The results show that the accuracy of the improved algorithm is 92.3%, which is 18.7% higher than that of the BP neural network; the recall rate is 89.6%, the F1 value is 90.9%, and the mean square error is reduced to 0.082. It is better than the comparison algorithm in all indicators, effectively improving the accuracy and reliability of the evaluation, and providing a scientific basis for investment decisions.