CAJ | 학술논문

提出一种基于概率的加权朴素贝叶斯分类算法。通过对每个属性做朴素贝叶斯分类，得到该属性分类正确的概率，把该概率作为相应的权重，分别加在条件属性上，得到加权后的朴素贝叶斯分类器。以weka自带的数据集和uci数据集进行分类测试，得到了较好的结果。
제출일충기우개솔적가권박소패협사분류산법。통과대매개속성주박소패협사분류，득도해속성분류정학적개솔，파해개솔작위상응적권중，분별가재조건속성상，득도가권후적박소패협사분류기。이weka자대적수거집화uci수거집진행분류측시，득도료교호적결과。
This paper presents a probability-based weighted naive Bayes classification algorithm. Doing naive Bayesian classifier for each attribute, then it can get the probability of correct classificationof the attribute. The result as the corresponding weights were added to the condition attributes, and it get the weighted naive Bayes classifier. Weka data sets and UCI data sets are tested to get better results.