Intelligent Optimization Prediction Algorithm for Statistical Data Based on Recurrent Neural Network

Intelligent optimization forecasting is based on artificial intelligence and prediction science to analyze and process data, and select appropriate models and parameters to solve practical problems through artificial intelligence. Some time series are highly correlated with their own past states, while some time series are correlated with certain variables. For the dimensionality of time series input data, this study mainly considers two aspects from univariate and multivariate time series prediction. Therefore, this paper proposes a BP recurrent neural network model to complement the wolf pack algorithm. BP neural network has the feature of error back propagation seeking, and recurrent neural network (RNN) has the advantage of being able to fully consider the interconnection between signals and signals. The implementation of the intelligent data statistical platform to transform the enterprise data business process, to achieve instantaneous and accurate calculation of massive data and automatic analysis and aggregation, so that users get rid of the trouble of manual aggregation of paper reports, thus saving a lot of manpower and material resources and time, and the instantaneous response data information also provides a reliable basis for enterprise decision-making. Compared with BiGRU and GRU, DBiGRU does not have much improvement in prediction accuracy, but has a significant improvement in the speed and progress of handling unexpected events.