线上平台英语听说能力训练模式自动匹配方法  

Automatic Matching Method of English Listening and Speaking Ability Training Mode on Online Platform

作　　者：罗曼林 杨高云[1] LUO Man-lin;YANG Gao-yun(International Education Department,Yueyang Vocational and Technical College,Yueyang 414000 China)

机构地区：[1]岳阳职业技术学院国际教育学院,湖南岳阳414000

出　　处：《自动化技术与应用》2022年第1期44-47,74,共5页Techniques of Automation and Applications

基　　金：湖南省社会科学成果评审委员会一般自筹项目(XSP20YBC254)。

摘　　要：为确保训练模式更加符合学生发展需求,提出线上平台英语听说能力训练模式自动匹配方法。利用改进的最大信息系数方法判断线性与非线性变量之间关联程度,获取影响学生听说能力的有关因素;采用朴素贝叶斯算法明确用户特征属性选取最大属性类别作为用户能力水平预测结果;通过特征抽取器获取用户特征矢量,经过神经网络训练,减少误差,计算特征矢量与各模式之间相似度,获得最高值的元素即为候选映射,得到最佳匹配模式。仿真实验表明,所提方法运算速度快,用时低于2s,匹配模式更加符合用户需求,以为训练提供有效指导,推动平台发展。In order to ensure that the training mode is more in line with the development needs of students, an automatic matching method of English listening and speaking ability training mode on online platform is proposed. This paper uses the improved maximum information coefficient method to judge the correlation between linear and nonlinear variables to obtain the relevant factors affecting students’ listening and speaking ability;uses naive Bayes algorithm to define the user’s characteristic attributes, selects the largest attribute category as the user’s ability level prediction. The user feature vector is obtained by the feature extractor, and the error is reduced by neural network training, and the similarity between the feature vector and each pattern is calculated. The element with the highest value is the candidate mapping, and the best matching training mode is obtained. The simulation results show that the proposed method has fast operation speed, the matching time is less than 2 s, and the matching pattern is more in line with the needs of users, to provide effective guidance for training and promote the development of the platform.

关 键 词：线上平台 英语听说能力 训练模式 自动匹配 神经网络 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]