On the design and quantification of privacy preserving data mining algorithms

By Dakshi Agrawal, Charu C. Aggarwal. 2001

1,188 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Networks and Communications, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Networks and Communications

In Statistics and Probability

Citations in the corpus

Privacy Preserving Data Mining (2015, methodology): “Such techniques are: randomization method (Agrawal and Srikant, 2000; Agrawal and Aggarwal, 2002), k-anonymity model and l-diversity (Machanavajjhala et al ., 2007), distributed privacy preservation and downgrading application effectiveness. k-anonymity captures protection of released data against…”
Data Synthesis based on Generative Adversarial Networks (2018, background): “However, it is also very popular to study the removal of noise and recovery of the original data in many related fields [9].”
Privacy-Preserving Multivariate Statistical Analysis: Linear Regression and Classification (2004, methodology): “In general, there are two different approaches: one is to use data distortion approach, in which data are disguised by the added noises [1, 2, 13, 14, 22, 23].”
Recommender Systems (2016, neutral): “Rather, distributed protocols [133] or perturbation techniques [35, 38, 484, 485] are used to collect the data only in a perturbed way or in the aggregate.”
Utility-Privacy Tradeoffs in Databases: An Information-Theoretic Approach (2011, background): “However, a rigorous information-theoretic treatment of the utility-privacy (U-P) tradeoff problem remains open and the following questions are yet to be addressed: (i) the statistical assumptions on the data that allow information-theoretic analysis, (ii) the capability of revealing different…”
Mining Association Rules under Privacy Constraints (2008, background): “This approach was extended in [3] and [26] to address a variety of subtle privacy loopholes.”
Anonymity-preserving data collection (2005, methodology): “Difierent random perturbation methods have been proposed and applied in difierent data mining algorithms [2, 1 , 12, 28, 11, 9]. Random perturbation is very e‐cient, but in general it can induce a tradeofi between privacy of respondents and accuracy of the data mining result: the more privacy each ”
The rise of "big data" on cloud computing: Review and open research issues (2015, background): “Agarwal and Aggarwal [117] proposed a privacy measure based on differential entropy.”
Sharemind: A Framework for Fast Privacy-Preserving Computations (2008, background): “So far the focus has been on randomised response techniques [2,1,13].”
Deep Private-Feature Extraction (2018, neutral): “in [11].”
Privacy, accuracy, and consistency too: a holistic solution to contingency table release (2007, background): “… Cynthia Dwork Microsoft Research, SVC dwork@microsoft.com Kunal Talwar Microsoft Research, SVC kunal@microsoft.com ABSTRACT The contingency table is a work horse of o.cial statistics, the format of reported data for the US Census, Bureau of Labor Statistics, and the Internal Revenue…”
A survey on privacy in mobile participatory sensing applications (2011, background): “Other examples that make use of data perturbation, though not in the context of participatory sensing, can be found in [84, 85, 86].”
Practical privacy: the SuLQ framework (2005, background): “Subsequent work demonstrated that the privacy definition introduced in [2] is far too weak [1] and suggested alternative privacy definitions [4, 6, 7, 8] and perturbation techniques (see [7] for a short survey of the evolution of privacy definitions since [2]).”
On the complexity of optimal K-anonymity (2004, background): “In certain scenarios such as tracking epidemics and product marketing, a data miner requires access to large volumes of (possibly) personal information, in order to spot interesting trends or correlations.”
A General Survey of Privacy-Preserving Data Mining Models and Algorithms (2008, methodology): “In practice, one can combine the process of approximation of Z with subtraction of the distribution Y from Z by using a variety of iterative methods such as those discussed in [2, 5].”
The applicability of the perturbation based privacy preserving data mining for real-world data (2008, methodology): “Later, refinements of this approach tightened the bounds on what private information is disclosed, by showing that the ability to reconstruct the distribution can be used to tighten estimates of the original values based on the distorted data [2].”
Calibrating Noise to Sensitivity in Private Data Analysis (2006, background): “This was shown to be problematic, as the definition ignored what an adversary knowing the underlying probability distribution might infer about the data Agrawal and Aggarwal (2001). Evfimievsky et al.”