Pingshui Wang

Bio: Pingshui Wang is an academic researcher from Anhui University of Finance and Economics. The author has contributed to research in topics: Information privacy & Association rule learning. The author has an hindex of 1, co-authored 1 publications receiving 10 citations.

Research on privacy preserving association rule mining a survey

Abstract: Association rule mining is one of the hottest research areas that investigate the automatic extraction of previously unknown patterns or rules from large amounts of data. Recently, there has been growing concern over the privacy implications of association rule mining. This paper described the basic concepts related to association rule mining, and analyzed and summarized the general principles and methods of privacy preserving association rule mining, and pointed out the drawback of the these methods. It also introduced an effective metrics for measuring side-effects resulted from privacy preserving process. Finally the present problems and directions for future research are discussed.

Privacy preserving association rule mining in vertically partitioned data

Abstract: A privacy preserving association rule mining algorithm was introducedThis algorithm preserved privacy of individual values by computing scalar productMeanwhile the algorithm of computing scalar product was given and the security was analyzed

Context-Aware Verifiable Cloud Computing

Abstract: Internet of Things (IoTs) has emerged to motivate various intelligent applications based on the data collected by various “things.” Cloud computing plays an important role for big data processing by providing data computing and processing services. However, cloud service providers may invade data privacy and provide inaccurate data processing results to users, and thus cannot be fully trusted. On the other hand, limited by computation resources and capabilities, cloud users mostly cannot independently process big data and perform verification on the correctness of data processing. This raises a special challenge on cloud computing verification, especially when user data are stored at the cloud in an encrypted form and processed for satisfying the requests raised in different contexts. But the current literature still lacks serious studies on this research issue. In this paper, we propose a context-aware verifiable computing scheme based on full homomorphic encryption by deploying an auditing protocol to verify the correctness of the encrypted data processing result. We design four optional auditing protocols to satisfy different security requirements. Their performance is evaluated and compared through performance analysis, algorithm implementation, and system simulation. The results show the effectiveness and efficiency of our designs. The pros and cons of all protocols are also analyzed and discussed based on rigorous comparison.

Method and apparatus for verifying processed data

Abstract: A method, comprising: obtaining, at a trustworthy party, a data processing result from a requesting party or a processing party in response to receiving a request for verifying correctness of the data processing result from the requesting party, wherein the data processing result is obtained by the requesting party from the processing party; obtaining, at a trustworthy party, the data used to getting the data processing result and a corresponding algorithm from the processing party, wherein the processing party uses the corresponding algorithm to process the data and gets the data processing result; processing, at the trustworthy party, the obtained data with the corresponding algorithm and comparing the processed result with the received data processing result, and if the two results are the same, the data processing result verified by the trustworthy party is correct.

Protecting Sensitive knowledge in association patterns mining

Abstract: Mining association rules from huge amounts of data is an important issue in data mining, with the discovered information often being commercially valuable. Moreover, companies that conduct similar business are often willing to collaborate with each other by mining significant knowledge patterns from the collaborative datasets to gain the mutual benefit. However, in a cooperative project, some of these companies may want certain strategic or private data called sensitive patterns not to be published in the database. Therefore, before the database is released for sharing, some sensitive patterns have to be hidden in the database because of privacy or security concerns. To solve this problem, sensitive-knowledge-hiding (association rules hiding) problem has been discussed in the research community working on security and knowledge discovery. The aim of these algorithms is to extract as much as nonsensitive knowledge from the collaborative databases as possible while protecting sensitive information. Sensitive-knowledge-hiding problem was proven to be a nondeterministic polynomial-time hard problem. After that, a lot of research has been completed to solve the problem. In this article, we will introduce and discuss the major categories of sensitive-knowledge-protecting methodologies. © 2011 Wiley Periodicals, Inc.