Data preparation for KDD through automatic reasoning based on description logic

作者：

Highlights：

• The success of KDD largely depends on ad-hoc and costly tasks of data preparation.

• We proposed automatic data preparation mechanisms to support these tasks.

• The framework has been valuated statistically in several real KDD projects.

• It has demonstrated to save an average of 30% of the time in data preparation.

• It is capable of reducing time in every KDD projects of that uses XML data.

摘要

•The success of KDD largely depends on ad-hoc and costly tasks of data preparation.•We proposed automatic data preparation mechanisms to support these tasks.•The framework has been valuated statistically in several real KDD projects.•It has demonstrated to save an average of 30% of the time in data preparation.•It is capable of reducing time in every KDD projects of that uses XML data.

论文关键词：KDD,Data preparation,Data mining,Description logic,Automatic reasoning

论文评审过程：Received 14 February 2014, Revised 4 March 2014, Accepted 6 March 2014, Available online 19 March 2014.

论文官网地址：https://doi.org/10.1016/j.is.2014.03.002