| 统计调查表缺失数据插补效果的实证分析 |
| |
| 引用本文： | 余竞,钟涵宇,刘利,杨晋浩. 统计调查表缺失数据插补效果的实证分析[J]. 成都大学学报(自然科学版), 2010, 29(4): 307-310. DOI: 10.3969/j.issn.1004-5422.2010.04.009 |
| |
| 作者姓名： | 余竞 钟涵宇 刘利 杨晋浩 |
| |
| 作者单位： | 成都大学信息科学与技术学院,四川,成都,610106;成都大学信息科学与技术学院,四川,成都,610106;成都大学信息科学与技术学院,四川,成都,610106;成都大学信息科学与技术学院,四川,成都,610106 |
| |
| 摘 要： | 针对统计调查表的实际数据,对其缺失数据进行了常用插补方法的实证分析.首先,实证分析了一维模型的局限性及缺点;其次,分别对决策树模型、神经网络模型、关联规则模型算法,在对输入(预测)变量进行系统优化基础上,统计插补的准确率,比较优劣;最后,提出了提高插补准确率的一个值得进一步研究的方向.
|
| 关 键 词： | 缺失数据 插补 相关性 实证分析 一维模型 多维模型 |
Empirical Analysis on Interpolation Effect of Statistical Missing Data |
| |
| YU Jing,ZHONG Hanyu,LIU Li,LIU Li,YANG Jinhao. Empirical Analysis on Interpolation Effect of Statistical Missing Data[J]. Journal of Chengdu University (Natural Science), 2010, 29(4): 307-310. DOI: 10.3969/j.issn.1004-5422.2010.04.009 |
| |
| Authors: | YU Jing ZHONG Hanyu LIU Li LIU Li YANG Jinhao |
| |
| Affiliation: | (School of Information Science and Technology,Chengdu University,Chengdu 610106,China) |
| |
| Abstract: | Empirical analysis on interpolation of statistical missing data was made.The limitation and drawback of one dimensional model was analyzed first;then,accuracy rate of interpolation was counted based on system optimization of input variables with algorithm of tree model,neural networks model and association rule model respectively.Their advantage and disadvantage were compared.Finally improving the interpolation accuracy was a good aspect needing further research. |
| |
| Keywords: | missing data interpolation relationship empirical analysis one-dimensional model multidimensional model |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
