计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (11): 119-123.

• 数据库、数据挖掘、机器学习 • 上一篇    下一篇

海产品安全预警系统缺失数据填补方法

鄂  旭,林  爽,金璐璐   

  1. 渤海大学 信息科学与技术学院,辽宁 锦州 121000
  • 出版日期:2015-06-01 发布日期:2015-06-12

Method for filling missing data in seafood safety warning system

E Xu, LIN Shuang, JIN Lulu   

  1. School of Information Science and Technology, Bohai University, Jinzhou, Liaoning 121000, China
  • Online:2015-06-01 Published:2015-06-12

摘要: 针对海产品安全预警系统中数据缺失问题,提出了一种缺失数据填补方法,目前,使用粗糙集填补的方法很多,但很多方法并没有考虑到每个对象缺失属性个数。该方法将存在缺失数据的信息表分为完备和不完备两部分,并分别对其进行处理,对缺失数据填补时综合考虑属性重要性和缺失属性个数;不存在缺失数据的信息表则直接输出;实验结果表明方法能用于海产品安全预警系统中缺失数据填补。

关键词: 数据填补, 海产品安全, 相似矩阵

Abstract: In view of missing data problem in seafood safety warning system, a method for filling the missing data is proposed. Currently, there are many ways using rough set to fill missing data, but many methods do not take into account the number of missing attributes for each object. With the method information missing data table is divided into two parts, complete and incomplete information table, and it is processed separately, attribute importance and the number of missing attribute are taken into account when filling the missing data. No missing data information table directly outputs. Experimental results show that the method can be used for seafood safety warning system to fill missing data.

Key words: data filling, seafood safety, similar matrix