基于神经网络词嵌入模型的单词联想记忆系统设计

Design of word association memory system based on neural network word embedding model

针对传统单词记忆软件因缺乏语境与单词关联造成的学习效率低下问题,文章采用Pytorch深度学习框架,基于神经网络构建词嵌入模型进行单词联想记忆系统设计。在待学习的英语语料上训练模型并生成词嵌入向量,计算英文单词之间的相似度,通过相似度进行单词关联分析。使用SpringBoot构建后端系统,使用VUE技术构建前端页面,设计并实现单词联想记忆系统。系统依据学习进度,将关联单词及其相应语境图片进行可视化展示,达到单词联想记忆的效果,从而提升单词的记忆效率。

To address the inefficiency of traditional word learning software due to the lack of context and word association,the Pytorch deep learning framework is used to build a word embedding model based on neural networks,train the model on the English corpus to be learned and generate a word embedding vector,use the word embedding vector to calculate the similarity between English words,and perform word association analysis through the similarity.The back-end system is built using SpringBoot and the front-end page is built using VUE technology to design and implement the word association memory system.The system visualizes the associated words and their corresponding contextual pictures according to the learning progress,so as to achieve the effect of word association memory and thus improve the efficiency of word memory.