基于多标记特征选择的英语词汇语义预测方法  

Semantic prediction of English vocabulary based on multi-label feature selection

作　　者：田烨 TIAN Ye(School of Foreign Languages,Xianyang Normal University,Xianyang 712000,Shaanxi Province,China)

机构地区：[1]咸阳师范学院外国语学院,陕西咸阳712000

出　　处：《信息技术》2023年第11期87-91,98,共6页Information Technology

基　　金：陕西省教育厅2020年度一般专项科学研究计划(20JK0432)。

摘　　要：为了精准预测大学生英语词汇语义掌握程度,提出基于多标记特征选择的英语词汇语义预测方法,通过多标记特征选择方法计算词汇样本相似度,构建样本相似矩阵,获取词汇样本特征,同时依靠特征参数对英语词汇的分布函数与几率密度函数进行计算,凭借计算结果优化极大似然算法,计算英语词汇语义预测的参数下线,以此确保预测的准确度,使用云计算方法对英语词汇之间的特征迭代计算。实验结果表明,所提方法能够精确预测出大学生所掌握的英语词汇语义,并且还能够分析大学生在不同语法的词汇语义掌握状况。In order to accurately predict college students’mastery of English vocabulary semantics,an English vocabulary semantic prediction method based on multi-label feature selection is proposed.The similarity of vocabulary samples is calculated through the multi-label feature selection method,the sample similarity matrix is constructed,and the vocabulary sample features are obtained.At the same time,the distribution function and probability density function of English vocabulary are calculated according to the feature parameters,and the maximum likelihood algorithm is optimized based on the calculation results to calculate the parameters of English vocabulary semantic prediction,so as to ensure the accuracy of prediction.Cloud computing method is used to iteratively calculate the features between English words.The experiment results show that the proposed method can accurately predict the English lexical semantics mastered by college students,and can also analyze the lexical semantics mastered by college students in different grammars.

关 键 词：多标记特征选择 英语词汇语义 几率密度后汉书 相似度 特征提取 

分 类 号：TP315[自动化与计算机技术—计算机软件与理论]