Research on Improved Shallow Neural Network Big Data Processing Model based on Gaussian Function

The application of the current new generation communication technology is gradually diversified, and global Internet users are increasing, leading some large enterprises to increasingly rely on faster more efficient big data processing technology. In order solve shortcomings algorithms, such as slow computing speed, accuracy be improved, poor online real-time learning ability, this research combines incremental sliding window ideas design two improved radial basis function (RBF) neural network algorithms with Gaussian kernel function. Duffing equation example "Top 100 single products for Taobao search glasses sales" were used verify performance algorithm. experimental results show that when total sample 100000, mean square errors IOL, SWOL, SVM ResNet50 1.86e-07, 1.59e-07, 3.37e-07 2.67e-07 respectively. set SKUs Search Glasses Sales" number samples in test 800, root 0.0060, 0.0056, 0.0069 0.0073 This shows RBF algorithm designed study, which integrates windows, has a stronger comprehensive ability process data, certain value improving based commodity recommendation e-commerce other industries.