关键词: 缺失数据, 缺失分类, 填充方法, 方法比较, 效果评价

Abstract:

In the era of big data, with the explosive growth of data, data acquisition has become easier and data missing has become more common. The lack of data greatly reduces the usefulness of the data and the handling of data missing has become a hot research topic in big data processing. The article first introduces the research significance of the problem of missing data and the current research status at home and abroad, then analyzes the reasons for the missing data systematically, classifies the problem of missing data, and reviews the methods of processing missing data at home and abroad in recent years, summarizes their respective advantages and disadvantages, scope of application, and effect evaluation indicators which focus on the regression filling, cluster filling and other filling methods. Finally, it summarizes and looks forward to the field of missing data processing methods.

Key words: missing data, missing classification, filling method, method comparison, effect evaluation