Obfuscating the topical intention in enterprise text search

Hwee Hwa Pang, Xiaokui Xiao, Jialie Shen

Research output: Chapter in Book/Report/Conference proceedingConference Paper published in Proceedings

Abstract

The text search queries in an enterprise can reveal the users' topic of interest, and in turn confidential staff or business information. To safeguard the enterprise from consequences arising from a disclosure of the query traces, it is desirable to obfuscate the true user intention from the search engine, without requiring it to be re-engineered. In this paper, we advocate a unique approach to profile the topics that are relevant to the user intention. Based on this approach, we introduce an (ε 1, ε 2)-privacy model that allows a user to stipulate that topics relevant to her intention at ε 1 level should appear to any adversary to be innocuous at ε 2 level. We then present a Top Priv algorithm to achieve the customized (ε 1, ε 2)-privacy requirement of individual users through injecting automatically formulated fake queries. The advantages of Top Priv over existing techniques are confirmed through benchmark queries on a real corpus, with experiment settings fashioned after an enterprise search application.

Original languageEnglish
Title of host publication2012 IEEE 28th International Conference on Data Engineering
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages1168-1179
Number of pages12
ISBN (Electronic)978-0-7695-4747-3
ISBN (Print)978-1-4673-0042-1
DOIs
Publication statusPublished - 30 Jul 2012
Externally publishedYes
EventIEEE 28th International Conference on Data Engineering, ICDE 2012 - Arlington, VA, United States
Duration: 1 Apr 20125 Apr 2012

Conference

ConferenceIEEE 28th International Conference on Data Engineering, ICDE 2012
CountryUnited States
CityArlington, VA
Period1/04/125/04/12

Fingerprint Dive into the research topics of 'Obfuscating the topical intention in enterprise text search'. Together they form a unique fingerprint.

Cite this