基于近邻噪声处理的KNN缺失数据填补算法
文献类型：期刊论文
| 作者 | 郝胜轩 ; 周晓锋; 宋宏
|
| 刊名 | 计算机仿真
 |
| 出版日期 | 2014 |
| 卷号 | 31期号:7页码:264-268 |
| 关键词 | 缺失数据填补 近邻 噪声最近邻 |
| ISSN号 | 1006-9348 |
| 其他题名 | Predicting Missing Values with KNN Based on the Elimination of Neighbor Noise |
| 产权排序 | 1 |
| 中文摘要 | 在优化算法的研究中,针对KNN算法对缺失数据的填补效果会因为原始数据中存在噪声而受到严重影响的问题,根据待填补缺失数据最近邻的近邻关系,提出了一种新的缺失数据填补算法——ENN-KNN(Eliminate Neighbor Noise k-Nearest Neighbor)。通过比较待填补缺失数据每个最近邻的真实近邻程度能够有效地识别潜在的噪声最近邻。最后使用所有非噪声最近邻对待填补缺失数据进行填补,从而消除了噪声最近邻对填补结果的影响。通过观察四组UCI数据集的仿真结果,可知ENN-KNN算法的填补准确性总体上要优于KNN算法。 |
| 英文摘要 | Traditional KNN imputation method for dealing with missing data is severely affected by the noise in the original data. This paper presents a novel imputation method for dealing with missing data，which is based on the relationship of nearest neighbors of missing data-ENN-KNN( Eliminate Neighbor Noise k － Nearest Neighbor) . ENN － KNN imputation method can effectively identify potential noise nearest neighbor by comparing each real nearest degree of nearest neighbor of missing data． It uses all nearest neighbors which are not noise nearest neighbor to deal with missing data，for this reason it can eliminate the effect of noise nearest neighbor for dealing with missing data． The experiment results of four groups of UCI data sets show that the ENN － KNN imputation method is overall superior to KNN imputation method on the performance of prediction accuracy. |
| 收录类别 | CSCD |
| 资助信息 | 北京市自然科学基金(7110001) |
| 语种 | 中文 |
| CSCD记录号 | CSCD:5206116 |
| 公开日期 | 2014-11-03 |
| 源URL | [http://ir.sia.ac.cn/handle/173321/15198]  |
| 专题 | 沈阳自动化研究所_数字工厂研究室 |
| 推荐引用方式 GB/T 7714 | 郝胜轩,周晓锋,宋宏. 基于近邻噪声处理的KNN缺失数据填补算法[J]. 计算机仿真,2014,31(7):264-268. |
| APA | 郝胜轩,周晓锋,&宋宏.(2014).基于近邻噪声处理的KNN缺失数据填补算法.计算机仿真,31(7),264-268. |
| MLA | 郝胜轩,et al."基于近邻噪声处理的KNN缺失数据填补算法".计算机仿真 31.7(2014):264-268. |
入库方式： OAI收割
来源：沈阳自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。