基于Apriori关联规则的大学语文阅读材料体裁分类方法  

Genre Classification of College Chinese Reading Materials Based on Apriori Association Rules

作　　者：采国润 肖宏飞[2] CAI Guo-run;XIAO Hong-fei(Basic Teaching Department,Chuzhou Vocational and Technical College,Chuzhou,Anhui 239000,China;Information Engineering College,Chuzhou Vocational and Technical College,Chuzhou,Anhui 239000,China)

机构地区：[1]滁州职业技术学院基础教学部,安徽滁州239000 [2]滁州职业技术学院信息工程学院,安徽滁州239000

出　　处：《河北北方学院学报（自然科学版）》2023年第3期15-21,共7页Journal of Hebei North University:Natural Science Edition

基　　金：安徽省职业与成人教育学会2021年度教育教学研究规划课题“高职院校大学语文课程思政的育人价值及其实践路径研究”(Azcj2021176)。

摘　　要：提出基于Apriori关联规则的大学语文阅读材料体裁分类方法,以便于大学语文阅读材料的检索。从大学语文阅读材料中初步提取符号、词性、词汇特征,通过分析各类特征的关联度、差异度,准确选择阅读材料特征,经极差正规化无量纲处理后,构建阅读材料体裁分类的样本数据,通过Predictive Apriori算法挖掘分类样本数据中特征与体裁类别间的强关联规则,并根据影响度指标值筛选强关联规则,选择其中全部正关联规则构建阅读材料体裁分类器,将待分类大学语文阅读材料特征作为体裁分类器的输入,通过关联规则匹配确定分类精度最大的关联规则,该规则对应类别即为大学语文阅读材料体裁的分类结果。实验结果表明:大学语文阅读材料的符号、词性、词汇特征可反映其体裁类别特点;该方法可实现大学语文阅读材料体裁分类,分类误差小。A genre classification method of college Chinese reading materials based on Apriori association rules is proposed to facilitate the retrieval of college Chinese reading materials.The characteristics of symbols,parts of speech and vocabulary were initially extracted from college Chinese reading materials.By analyzing the correlation and difference of various features,the features of reading materials were accurately selected.After the range normalization and dimensionless processing,the sample data of genre classification of reading materials were constructed.The strong association rules between features and genre categories in the classified sample data were mined through the Predictive Apriori algorithm,and the strong association rules were filtered according to the influence index value.All the positive association rules were selected to construct a genre classifier for reading materials,and the features of college Chinese reading materials were classified as the input of the genre classifier.The association rule with the highest classification accuracy were determined through association rule matching.The corresponding category of the rule was the classification result of college Chinese reading materials.The experimental results show that the characteristics of signs,parts of speech and vocabulary of college Chinese reading materials can reflect their genre characteristics.This method can be used to realize genre classification of college Chinese reading materials with small classification error.

关 键 词：Apriori关联规则 大学语文 体裁分类 特征差异 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]