Functions that Preserve Privacy but Permit Analysis of Text Paper 246
نویسنده
چکیده
Cryptographically strong functions can be used to preserve privacy of text content. For example, one way functions have been construed as random functions on their inputs. Given this, it is reasonable to ask if a one way function can still preserve some “property” of its inputs. Specifically, is it possible to perform some measurement on the image on a one way function that is correlated with the same measurement on the pre-image of the function? In this paper we show that this is indeed possible. If the measurement function “throws away” enough entropy it will still be possible to perform the correlated measurement. Thus it is possible to analyze properties of text while still providing security and privacy for its content.
منابع مشابه
Author Attribute Anonymity by Adversarial Training of Neural Machine Translation
Text-based analysis methods allow to reveal privacy relevant author attributes such as gender, age and identify of the text’s author. Such methods can compromise the privacy of an anonymous author even when the author tries to remove privacy sensitive content. In this paper, we propose an automatic method, called Adversarial Author Attribute Anonymity Neural Translation (ANT), to combat such te...
متن کاملA4NT: Author Attribute Anonymity by Adversarial Training of Neural Machine Translation
Text-based analysis methods allow to reveal privacy relevant author attributes such as gender, age and identify of the text’s author. Such methods can compromise the privacy of an anonymous author even when the author tries to remove privacy sensitive content. In this paper, we propose an automatic method, called Adversarial Author Attribute Anonymity Neural Translation (ANT), to combat such te...
متن کاملUnderstanding the Inconsistencies between Text Descriptions and the Use of Privacy-sensitive Resources of Mobile Apps
Permission warnings and privacy policy enforcement are widely used to inform mobile app users of privacy threats. These mechanisms disclose information about use of privacy-sensitive resources such as user location or contact list. However, it has been reported that very few users pay attention to these mechanisms during installation. Instead, a user may focus on a more user-friendly source of ...
متن کاملA Machine Learning Based System for Semi-Automatically Redacting Documents
Redacting text documents has traditionally been a mostly manual activity, making it expensive and prone to disclosure risks. This paper describes a semi-automated system to ensure a specified level of privacy in text data sets. Recent work has attempted to quantify the likelihood of privacy breaches for text data. We build on these notions to provide a means of obstructing such breaches by fram...
متن کاملA Heuristic Approach to Preserve Privacy in Stream Data with Classification
Data stream Mining is new era in data mining field. Numerous algorithms are used to extract knowledge and classify stream data. Data stream mining gives birth to a problem threat of data privacy. Traditional algorithms are not appropriate for stream data due to large scale. To build classification model for large scale also required some time constraints which is not fulfilled by traditional al...
متن کامل