A BP Neural Network Wordle Result Prediction Model Optimized Based on Genetic Algorithm

This paper presents a BP neural network model that has been enhanced through the application of a genetic algorithm, referred to as GA-BP, which leads to improved prediction accuracy. Traditional BP neural networks often suffer from problems such as slow convergence and a propensity to converge on local minima. To surmount these limitations, the genetic algorithm is employed to optimize the weights and thresholds of the BP neural network. The performance of the GA-BP model is evaluated using metrics like Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). Experimental results demonstrate that the GA-BP model significantly outperforms the conventional BP neural network, achieving RMSE, MAE, and MAPE values of 0.92, 0.74, and 37.8%, respectively, compared to the BP model's 1.34, 0.96, and 44.3%. The study concludes that the GA-BP model is highly effective for tasks requiring high-precision predictions and offers a promising direction for future research in neural network optimization.