Frequent Itemset Mining with Hadamard Response Under Local Differential Privacy

Liu, Haijiang; Bai, Xiangyu; Ma, Xuebin; Cui, Lianwei

doi:10.1109/iceiec49280.2020.9152248

Cited by 6 publications

(13 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Differential privacy [2], [16], [17], [19], as a rigorous privacy protection model, has attracted a lot of attention, since it can provide theoretical privacy guarantee against adversaries with arbitrary background information. Before introducing the definition of differential privacy, we first give the definition of neighboring database in the following.…”

Section: Related Workmentioning

confidence: 99%

“…Definition 2 (ε-DP [20]): Given a randomized algorithm A, A satisfies ε-DP if for any two neighboring databases D 1 and D 2 , and any output O of A, there exists [17], [19] have been devoted to publishing statistics for high-dimensional data. Existing works can be classified into three categories.…”

Section: Related Workmentioning

confidence: 99%

“…Because DPsense is closely related with our work and has a superiority to GS, as validated in [1], we will give more details about it in Section 3.3. [16], [17], [19], [23] There are a lot of works [16], [17], [19], [23] focusing on the top-k frequent itemsets (column patterns) mining. Since our problem top-k frequent columns mining is just one fundamental step of frequent itemsets mining, we just pay attention to the techniques in existing works [16], [17], [19], [23] designed to optimize this fundamental step.…”

Section: B Publishing Noisy Column Frequencies [1]-[3]mentioning

confidence: 99%

“…However, the noises in frequencies make the columns with frequencies close to the one of the kth frequent column (called near-k-fluctuation-column) be misjudged, leading to inaccurate results. The other line [16], [17] invokes exponential mechanism to sample one frequent column with probability inversely proportional to k. This is performed k times to get published results. However, it does not work well for a large k in high-dimensional applications.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Differentially Private Top-$k$ Frequent Columns Publication for High-Dimensional Data

Wang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Differential privacy (DP) is a promising scheme for releasing the results of statistical queries on sensitive data. This paper focuses on top-k frequent columns publication on sensitive data, with high result utility under differential privacy. Existing works directly select frequent columns from all columns (called one-phase scheme), which is far from ideal due to the large privacy consumption or misjudgments for columns with frequencies close to the frequency of the kth frequent column (called near-k-fluctuationcolumn). This paper presents a new solution Two Phase Selection (TPS) to carefully choose frequent columns in two phases. The main idea is to classify columns into two distinct categories based on whether it is one near-k-fluctuation-column or not. Frequent columns are chosen from the two categories using different techniques, which is totally different from existing solutions without classifying. Furthermore, by analyzing the distribution of near-k-fluctuation-columns, we introduce a block-centric column-choosing method privacy-free-mechanism (PFM). By partitioning columns into blocks, PFM makes the privacy consumption proportional to the number of blocks, instead of frequent columns. Extensive experiments on real datasets show that our proposals outperform the state-of-the-art techniques for top-k column publication.INDEX TERMS Differential privacy, privacy protection, top-k, high dimensional dataset, privacy consumption, block.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: B Publishing Noisy Column Frequencies [1]-[3]mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Differentially Private Top-$k$ Frequent Columns Publication for High-Dimensional Data

Wang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…So far, many traditional data mining algorithms have been extended to satisfy differential privacy, e.g. k-means [11], k-nearest neighbour classification [12], random forest [13], frequent itemset mining [14,15] and so on. Nevertheless, when focusing on spectral clustering, there are few works with regard to differentially private spectral clustering algorithms.…”

Section: Introductionmentioning

confidence: 99%

Privacy‐preserving constrained spectral clustering algorithm for large‐scale data sets

Wei

et al. 2020

IET Information Security

View full text Add to dashboard Cite

Locally Differentially Private Frequent Itemset Mining

Wang

Jha

2018

2018 IEEE Symposium on Security and Privacy (SP)

172

157

View full text Add to dashboard Cite

The notion of Local Differential Privacy (LDP) enables users to respond to sensitive questions while preserving their privacy. The basic LDP frequent oracle (FO) protocol enables an aggregator to estimate the frequency of any value. But when each user has a set of values, one needs an additional padding and sampling step to find the frequent values and estimate their frequencies. In this paper, we formally define such padding and sample based frequency oracles (PSFO). We further identify the privacy amplification property in PSFO. As a result, we propose SVIM, a protocol for finding frequent items in the set-valued LDP setting. Experiments show that under the same privacy guarantee and computational cost, SVIM significantly improves over existing methods. With SVIM to find frequent items, we propose SVSM to effectively find frequent itemsets, which to our knowledge has not been done before in the LDP setting.

show abstract

Frequent Itemset Mining with Hadamard Response Under Local Differential Privacy

Cited by 6 publications

References 9 publications

Differentially Private Top-$k$ Frequent Columns Publication for High-Dimensional Data

Differentially Private Top-$k$ Frequent Columns Publication for High-Dimensional Data

Privacy‐preserving constrained spectral clustering algorithm for large‐scale data sets

Locally Differentially Private Frequent Itemset Mining

Contact Info

Product

Resources

About