Abstract | ||
---|---|---|
Addressing privacy regulation such as GDPR requires organizations to find and classify sensitive and personal data in their datastores. First, data discovery tools are applied to identify the data. Then, data classification tools are applied on the data that was discovered. Organizations must classify the data into concrete categories to manage data appropriately. In this paper we focus on multi-value classification, where the classifier provides a category to set of values all from the same category. Traditional classifiers usually apply single-value classification methods to a multi-value data set. However, in many cases this resulting an incorrect classification when, for example, domain categories overlap. In this paper, we address this scenario and provide two methods to overcome this problem. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1007/978-3-030-32213-7_16 | Communications in Computer and Information Science |
Keywords | Field | DocType |
Classification,Data discovery,Relational database | Computer science,Artificial intelligence,Natural language processing | Conference |
Volume | ISSN | Citations |
1085 | 1865-0929 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Sigal Assaf | 1 | 0 | 0.34 |
Ariel Farkash | 2 | 5 | 3.47 |
Micha Moffie | 3 | 0 | 0.34 |