查询结果:   孙玲芳,徐会,王成文,祁军.基于动态约简的增量贝叶斯分类算法的研究[J].计算机应用与软件,2015,32(3):188 - 191.
中文标题
基于动态约简的增量贝叶斯分类算法的研究
发表栏目
人工智能与识别
摘要点击数
771
英文标题
ON INCREMENTAL Naïve BAYESIAN CLASSIFICATION ALGORITHM BASED ON DYNAMIC REDUCTION
作 者
孙玲芳 徐会 王成文 祁军 Sun Lingfang Xu Hui Wang Chengwen Qi Jun
作者单位
泰州学院 江苏 泰州 225300 江苏科技大学经济管理学院 江苏 镇江 212003    
英文单位
Taizhou University, Taizhou 225300, Jiangsu, China School of Economics and Management, Jiangsu University of Science and Technology, Zhenjiang 212003, Jiangsu, China    
关键词
粗糙集 动态约简 增量学习 朴素贝叶斯
Keywords
Rough set Dynamic reduction Incremental learning Naïve Bayesian
基金项目
教育部人文社会科学研究项目(10YJAZH069);江苏省第九批“六大人才高峰”高层次人才项目(XXRJ-013)
作者资料
孙玲芳,教授,主研领域:管理信息系统。徐会,硕士。王成文,硕士。祁军,硕士。 。
文章摘要
朴素贝叶斯由于条件独立性假设使其分类效果不明显,同时在处理海量数据时缺乏灵活性。针对以上情况,提出一种基于动态约简的增量贝叶斯分类算法。算法首先利用(F-λ)广义动态约简计算出数据集的核属性,然后根据训练集的先验信息构造分类器对测试实例进行分类,最后利用类置信度进行选择性增量学习,增强处理增量数据的能力。实验结果表明,该算法在处理属性少的小量数据时,分类效果有一定的改善,在处理多属性大量数据时,分类效果明显提高。
Abstract
The classification effect of Naïve Bayesian is not obvious because of conditional independence assumption, so does its lack of flexibility in dealing with massive data. In view of the above, we propose a dynamic reduction-based incremental nave Bayesian classification algorithm. This algorithm uses (F-λ) generalised dynamic reduction to calculate the core attributes of dataset first, and then constructs classifier according to priori information of training sets to classify the test cases. Finally, it uses class confidence to conduct selective incremental learning for enhancing the capability of incremental data processing. Experimental results show that the algorithm ameliorates the classification effect to a certain extent when dealing with few attributes and small amount of data, and when dealing with more attributes and large amount of data, the classification effect is obviously improved as well.
下载PDF全文