Quality Data for Data Mining and Data Mining for Quality Data: A Constraint Based Approach in XML

Abstract

As quality data is important for data mining, reversely data mining is necessary to measure the quality of data. Specifically, in XML, the issue of quality data for mining purposes and also using data mining techniques for quality measures is becoming more necessary as a massive amount of data is being stored and represented over the Web. We propose two important interrelated issues: how quality XML data is useful for data mining in XML and how data mining in XML is used to measure the quality data for XML. When we address both issues, we consider XML constraints because constraints in XML can be used for quality measurement in XML data and also for finding some important patterns and association rules in XML data mining. We note that XML constraints can play an important role for data quality and data mining in XML. We address the theoretical framework rather than solutions. Our research framework is towards the broader task of data mining and data quality for XML data integrations.