An Improved BPNN Prediction Method Based on Multi-Strategy Sparrow Search Algorithm

Data prediction can improve the science of decision-making by making predictions about what happens in daily life based on natural law trends. Back propagation (BP) neural network is a widely used method. To reduce its probability falling into local optimum and accuracy, we propose an improved BP method multi-strategy sparrow search algorithm (MSSA). The weights thresholds are optimized using (SSA). Three strategies designed to SSA enhance optimization-seeking ability, leading MSSA-BP model. MSSA was tested with nine different types benchmark functions verify optimization performance algorithm. Two datasets were selected for comparison experiments three groups models. Under same conditions, mean absolute error (MAE), root square (RMSE), percentage (MAPE) results significantly reduced, convergence speed improved. effectively accuracy has certain application value.