ATTRIBUTE-BASED UTILITY MEASUREMENT METHOD FOR DE-IDENTIFIED UNSTRUCTURED TEXT

摘要

The present invention relates to an attribute-based usability measurement technique for unstructured text in which de-identification has been performed, and more particularly, a usefulness measurement technique suitable for de-identification results by applying a protection model for unstructured text and measured values. The present invention proposes a method for intuitively determining the usefulness of the unstructured text. The present invention allows the measurement of a suitable usefulness value for the result of de-identification of an unstructured text and provides an intuitive value for de-identification. Even general data users without specialized knowledge can easily determine the usefulness of de-identified datasets.