A Method for Classification of Missing Values using Data Mining Techniques
Bhavani Sankar Panda, Rajesh Kumar Adhikari · 2020 International Conference on Computer Science, Engineering and Applications (ICCSEA) · 2020
This paper is intended for research in the area of Data Mining. The classification accuracy is improved by attributing the missing values of data. The effective attributing method model based Missing value attribution using Correlation is proposed. It represents the missing value using the correlation among the attributes. It works adequately for both categorical and numeric data. It handles the missing values for classification. It increases the accuracy of classification. The proposed imputation method handles the missing values during classification is contributed because it will be suitable for real-time applications. In this article moreover directs a proper review of the missing value approach. A catalog of missing techniques which are normally recognized by the society under which missing data can happen - MCAR (Missing Completely at Random), MAR (Missing at Random), NMAR (Not Missing at Random). This paper discusses several techniques for treating missing data. It discusses numerous strategies for management of missing values there in the dataset.