Research on Prediction of Port Cargo Throughput based on PCA-BP Neural Network Combination Model

Effective prediction can help people make reasonable and accurate judgments about the future development level of things, and then guide and regulate production management activities. With the development of big data technology, data prediction technology is no longer limited to the traditional time series prediction and simple causal prediction, but more biased towards machine learning, AI technology and so on. However, there are some limitations in using big data for prediction, such as data size and threshold problem. In this paper, the combination model of Principal Component Analysis (PCA) method and BP Neural Network algorithm is applied to the prediction. Firstly, the dimension of a large number of index data is reduced through PCA, and the effective information is retained while the amount of data is reduced. Then the BP Neural network model is used to predict. This paper chooses Dalian port cargo throughput as an example to verify the effectiveness of the model, the results show that the model has higher accuracy and efficiency.