摘要
近年,各类网络平台大量投入高校"一站式"学生社区建设.主要探索机器学习算法在"一站式"学生社区网络平台事件言论分类中的效果.主要使用机器学习算法,包括K近邻、决策树、多项式朴素贝叶斯、逻辑回归、支持向量机、随机森林、轻量级梯度提升、极值梯度提升算法构建分类模型,对社区事件言论进行文本分类.结果显示,上述算法的准确率依次为0.62、0.74、0.89、0.88、0.85、0.87、0.80、0.82,结果表明分类具有较好效果.
Abstract
In recent years,various online platforms have been heavily invested in the construction of"one-stop"student com-munities in colleges and universities.This paper mainly explores the effect of machine learning algorithm in"one-stop"student community network platform event speech classification.This paper mainly uses machine learning algorithms,including K-nearest neighbor,decision tree,polynomial naive Bayes,logistic regression,support vector machine,random forest,lightweight gradient lift-ing,extreme gradient lifting algorithms to construct classification models for short text classification.The results show that the accu-racy of the above algorithm is 0.62,0.74,0.89,0.88,0.85,0.87,0.80,0.82,which shows that the classification has a good effect.
万方数据