LSTM-RVFL-Attention: A Novel Hybrid Deep Learning Model for Accurate Sales Prediction

In the era of big data and digital transformation, accurate sales prediction is of great significance for enterprises. This study proposes a novel approach for sales prediction using the LSTM-RVFL-Attention hybrid neural network model and evaluates its performance on a comprehensive dataset. Sales prediction is influenced by various complex factors, and traditional methods have limitations, while neural networks show potential in handling such problems. This paper details the characteristics of the dataset used, elaborates on the architecture and functions of each layer (LSTM layer, RVFL layer, and attention mechanism) of the model. The experimental results show that the training and test loss curves of the model decrease steadily, the predicted values follow a similar trend to the actual values, but there are still deviations, and the error distribution is symmetric to some extent, leaving room for improvement. The research indicates that the model is effective in sales prediction, but further exploration is needed to enhance its performance, providing new ideas and methods for enterprise sales prediction.