摘要
大数据时代产生了大量的文本数据,由于数据源的不同以及产生数据过程的复杂性,造成了大量的数据具有异质结构.将几种最新的机器学习方法与分层技术结合,针对具有异质结构的文本数据的算法.提高文本分类的精准率.实证数据分析表明我们提出的算法在提高分类精确度上有很显著的效果.
The age of big data has produced a lot of text data.Due to the different data sources and the complexity of the process of generating data,a large amount of data with heterogeneous structures is produced.This paper combines several new machine learning methods with layering techniques,an algorithm for text data with heterogeneous structure is designed to improve the accuracy of text classification.Empirical data analysis shows that the proposed algorithm has a significant effect on improving the classification accuracy.
作者
李俊华
程婉玉
LI Jun-hua;CHEN wan-yu(School of Maths and Computer Science,Hanjiang Normal University,Shiyan 442000,China;School of Maths and Statistics,Wuhan University,Wuhan 430072,China)
出处
《汉江师范学院学报》
2021年第6期80-86,共7页
Journal of Hanjiang Normal University
基金
湖北省教育厅科学研究基金项目“基于KwMO扩展的统计分布研究”(项目编号:B2021286)
湖北省教育厅科学技术研究计划青年人才项目“超网络的建模及重构研究”(项目编号:Q20203103)
关键词
文本数据
分类
机器学习
分层抽样
text data
classify
machine learning
stratified sample
