Auditing Keyword Queries Over Text Documents
Lists
Only you see this
Abstract
Designing a robust data management system, requires securing access to all the three \ntypes of data, namely structured, semi-structured and unstructured data. In this paper, \nwe present an auditing model to secure text document access. Given a sensitive docu- \nment, we compute the candidate suspicious keyword queries that may have accessed the \nsensitive document. Our auditing model allows users to specify either the full document \nor some specific portion of the document as sensitive. All queries that have accessed a \nsensitive document may not lead to disclosure of the sensitive document, some of them \nmight be very regular accesses. We present an outlier mining based algorithm to find \ntop-k anomalous queries from candidate suspicious queries. \nData security and privacy is an issue of growing importance in healthcare domain. Query \nauditing is often used in healthcare domain to detect privacy violation. However, as \nunstructured healthcare data, such as medical reports, etc., are not easily available \nfor public research. In this paper, we show how one can use the publically available \nDBLP data to create an equivalent healthcare data, which can be used for experimental \nevaluation.
BibTeX
@inproceedings{appareddy2017auditing,
author = {Bharath Kumar Reddy Appareddy and Manish Singh},
title = {{Auditing Keyword Queries Over Text Documents}},
booktitle = {Research},
year = {2017},
}Loading related content...