Simulation Research on Artificial Intelligence Design Decision Model Based on Deep Convolution Neural Network

At the initial stage of the decision-making process of AI(Artificial Intelligence) design, there are only a small number of training samples, and the decision information that can be extracted is very small. After that, the labeling data gradually increases, and finally forms an incremental process of data and decision information increase. In this paper, the artificial intelligence design decision model is simulated based on DCNN (Deep Revolution Neural Network). The intelligent design theory and decision theory in artificial intelligence are applied to the management and decision work of the actual construction site with the quantitative analysis of the decision support tools. The change trend of classification accuracy increases with the increase of decision steps. This shows that the performance of the decision model is gradually improving with the sequential process and the increase of marked images, which shows that the sequential incremental decision method is effective. However, contrary to the decision cost and total cost, the accuracy of GA (Genetic Algorithm) decision is lower than that of CF (Collaborative Filtering) decision. However, the algorithm in this paper has the best performance among them, and its accuracy is higher than the above two algorithms.