Research on Improving Personalized Recommendation Accuracy Based on NLP Semantic Analysis

Abstract

In order to improve the accuracy of English machine translation under the deep learning mode, an improved personalized recommendation model for English machine translation based on NLP(natural language processing) semantic analysis is proposed. According to the semantic mapping structure characteristics of the source language words, the NLP semantic analysis model of English machine translation under the deep learning mode is established. The parameters of the training model with large-scale corpus are taken as the initial parameters of the scarce corpus model. The neural network parameters of the deep learning are used to learn multiple English machine translation tasks at the same time, and the meta learning is introduced into machine translation. The NLP semantic learning task of English machine translation is solved by using a small amount of computational gradient iteration. The NLP semantic analysis information is introduced into the translation model to expand the existing corpus resources and improve the personalized recommendation accuracy based on NLP semantic analysis. The test results show that the personalized recommendation learning ability of English machine translation using this method is strong, which can effectively use the contribution of different parameters to the loss function to compare whether the parameters are beneficial to the training process, and improve the efficiency of recommendation and the vocabulary matching ability in translation.