Improving Relational Database Quality Based on Adaptive Learning Method for Estimating Null Value
Ching‐Hsue Cheng, Liang‐Ying Wei, Tzu-Cheng Lin · 2007
The poor data quality can cause a host of negative results, including lost information, operational inefficiencies in a database system and much more. Therefore, for many applications in the area of data analysis processing, data preprocessing is an essential step. In addition, the handling of null values is the major task of the data quality. In this paper, we present a new method for estimating null values in relational database systems based on adaptive learning techniques. The proposed method utilizes clustering algorithms to cluster data, and calculate coefficient values between different attributes by generating minimum average error. To verify our method, this paper utilizes two database ( Waugh 's database and Employee relational database), and Mean of Absolute Error Rate (MAER) as evaluation criterion to compare with the listing methods; it is shown that our proposed method is better than the listing methods for estimating null values in relational database systems.