Privacy Preserving Data Mining within Anonymous Credential Systems

Optimized authentication algorithm for privacy-preserving anonymous credentials using randomized aggregate signatures

In an anonymous credential system, a user can obtain credentials from the organization and can prove the possession of credentials to the service providers, without leaking the privacy. There are three kinds of parties in the scheme: organization, service providers and users. In the previous works, the anonymous credentials schemes cannot prevent the collusion between the organization and the service providers from finding the users' privacy. In this paper, a scheme of anti-collusion anonymous credentials is proposed, it can be based on any secure public key algorithm. In the scheme, there doesn't need a trusted third party to issue the public keys and the secret keys, the scheme take the advantage of the relation of the public key and the secret key, i.e., it is easy to generate a pair of public key and the corresponding secret key, but hard to deduce the secret key from a particular public key. Therefore, the participant only need to have the ability to generate a pair of public key and secret key.

The trends of data mining on healthcare data for improving medical services have increased because of the electronic healthcare record(EHR) system, which collects a massive amount of data on a daily basis. In the current scenario, hospital maintains its EHR system and stores the detailed information of patients. Data mining for healthcare improvement requires the data from all the EHR systems located at a different location to be stored at the central data mining server. Collection of healthcare data at some untrusted central data mining server raises privacy threats. Healthcare data contains patients’ private information and sharing this information for data mining creates privacy issues. Most of the previous research either focused on k-anonymity technique which causes information loss and decreases data mining accuracy or privacy preserving data mining which is focused on only specific data mining technique. We adopt source anonymous technique as privacy preserving scheme and present a novel scheme for healthcare data collection and mining in this paper. Our scheme collects data from all EHR systems without any information loss and stores at a single central data mining server, also ensuring privacy is preserved. Central data mining server helps to analyze the collected data with different data mining techniques (Association rule mining, Classification, Clustering, etc.) without the involvement of EHR systems. Our scheme is collusion resilient against central data mining server and EHR systems. Theoretical and experimental analysis show the efficiency of our scheme in terms of computation and communication cost. The experimental results using Heart disease dataset show the advantage to EHR systems using the proposed approach in terms of disease prediction accuracy.

Secure identity tokens such as Electronic Identity (eID) cards are emerging everywhere. At the same time user-centric identity management gains acceptance. Anonymous credential schemes are the optimal realization of user-centricity. However, on inexpensive hardware platforms, typically used for eID cards, these schemes could not be made to meet the necessary requirements such as future-proof key lengths and transaction times on the order of 10 seconds. The reasons for this is the need for the hardware platform to be standardized and certified. Therefore an implementation is only possible as a Java Card applet. This results in severe restrictions: little memory (transient and persistent), an 8-bit CPU, and access to hardware acceleration for cryptographic operations only by defined interfaces such as RSA encryption operations.Still, we present the first practical implementation of an anonymous credential system on a Java Card 2.2.1. We achieve transaction times that are orders of magnitudes faster than those of any prior attempt, while raising the bar in terms of key length and trust model. Our system is the first one to act completely autonomously on card and to maintain its properties in the face of an untrusted terminal. In addition, we provide a formal system specification and share our solution strategies and experiences gained and with the Java Card.

We present a technique that uses privacy enhancing technologies and biometrics to prevent the unauthorized lending of credentials. Current credential schemes suffer the weakness that issued credentials can be transferred between users. Our technique ensures the biometric identity of the individual executing the Issue and Show protocols of an existing credential system in a manner analogous to the enrollment and verification steps in traditional biometric systems. During Issue we create Pedersen commitments on biometrically derived keys obtained from fuzzy extractors. This issue-time commitment is sealed into the issued credential. During Show a verification-time commitment is generated. Correspondence of keys is verified using a zero-knowledge proof of knowledge. The proposed approach preserves the security of the underlying credential system, protects the privacy of the biometric, and generalizes to multiple biometric modalities. We illustrate the usage of our technique by showing how it can be incorporated into digital credentials and anonymous credentials.

We investigate privacy-preserving Public-Key Infrastructures (PKIs) with additional security, privacy, and performance requirements meant to support vehicular networks. While conventional PKIs are a cornerstone for identity management in certain security applications, they do not offer the privacy properties that vehicular networks demand. Over the last few years, there have been efforts to generalize conventional PKIs to meet these new privacy and security demands. One particular example is the class of multiple certificate PKIs where each vehicle is given multiple short-lived and anonymous certificates that can function as credentials until the vehicle is due for, say, the next inspection or dealership visit. While such PKIs offer attractive security and privacy properties, achieving acceptable performance has remained an open problem thus questioning their practicality.In this paper, we formally define and investigate the efficient certificate revocation list (CRL) search problem for multiple certificate PKIs, which asks whether it is possible to efficiently test CRL membership of a given certificate in the same order of computational complexity as in conventional PKIs (that is, by a simple and fast binary search over the set of revoked PKI members). Any solution to this problem is of particular interest when it does not introduce unacceptable compromises on other important properties such as security and privacy. We start by formally defining vehicular networks and a number of security, privacy and performance requirements for them. Then we demonstrate that for PKIs based on general classes of efficient algorithms, containing binary search over totally ordered sets and approximate nearest neighbor search over metric spaces, a solution to the efficient CRL search problem can be used to violate privacy. Finally, we present and analyze new solutions that simultaneously solve the efficient cRL search problem and satisfy privacy requirements by allowing all vehicles to locally (i.e., with minimal help from the CA) and consistently (i.e., with the same results across all vehicles) manage their CRLs. Using these ideas, we can maintain security and privacy properties while reducing the CRL membership testing time to logarithmic in the number of revoked vehicles, which is essentially the same performance that one obtains with conventional PKIs where privacy is not required.

This paper introduces and formalizes homomorphic proofs that allow ‘adding’ proofs and proof statements to get a new proof of the ‘sum’ statement. Additionally, we introduce a construction of homomorphic proofs, and show an accumulator scheme with delegatable non-membership proofs (ADNMP) as one of its applications with provable security. Finally, the proposed accumulator method extends the BCCKLS scheme [1] to create a new provably secure revocable delegatable anonymous credential (RDAC) system. Intuitively, the new accumulator’s delegatable non-membership (NM) proofs enable user A, without revealing her identity, to delegate to user B the ability to prove that A’s identity is not included in a blacklist that can later be updated. The delegation is redelegatable, unlinkable, and verifiable.

We consider the problem of constructing anonymous credentials for use in a setting where the issuer of credentials is also the verifier, or more generally where the issuer and verifier have a shared key. In this setting we can use message authentication codes (MACs) instead of public key signatures as the basis for the credential system. To this end, we construct two algebraic MACs in prime-order groups, along with efficient protocols for issuing credentials, asserting possession of a credential, and proving statements about hidden attributes (e.g., the age of the credential owner). We prove the security of the first scheme in the generic group model, and prove the security of the second scheme\dash using a dual-system-based approach\dash under decisional Diffie-Hellman (DDH). Our MACs are of independent interest, as they are the only uf-cmva-secure MACs with efficient proofs of knowledge. Finally, we compare the efficiency of our new systems to two existing constructions of anonymous credentials: U-Prove and Idemix. We show that the performance of the new schemes is competitive with U-Prove (which does not have multi-show unlinkability), and many times faster than Idemix.

We extend the Camenisch-Lysyanskaya anonymous credential system such that selective disclosure of attributes becomes highly efficient. The resulting system significantly improves upon existing approaches, which suffer from a linear number of modular exponentiations in the total number of attributes. This limitation makes them unfit for many practical applications, such as electronic identity cards. Our novel approach can incorporate a large number of binary and finite-set attributes without significant performance impact. It compresses all such attributes into a single attribute base and, thus, boosts the efficiency of all proofs of possession. The core idea is to encode discrete binary and finite-set values as prime numbers. We then use the divisibility property for efficient proofs of their presence or absence. In addition, we contribute efficient methods for conjunctions and disjunctions. The system builds on the strong RSA assumption. We demonstrate the aptness of our method in realistic application scenarios, notably electronic identity cards, and show its advantages for small devices, such as smartcards and cell phones.

In an anonymous credential system a user can prove anonymously the possession of credentials to a service provider. Multi-show and non-transferability are two important properties of such systems. More precisely, in a multi-show system the same credential can be used more than once without threatening anonymity, moreover, lending of non-transferable credentials is inconvenient. In this paper we give a construction for multi-show non-transferable credentials for which the owner can prove that the credentials satisfy access control policies expressed by means of linear boolean formulae.

Privacy preservation of online user behavior data is important aspect in data mining, as of late, the study of privacy preserving data mining or networking has been studied mainly due to the extensive proliferation of confidential data on the internet. As online social networking sites attract thousands of million people to use it every day for different purpose, due to this lots of data is being generated every day on internet. Online user behavior analysis and security is becoming increasingly important, as it offers valuable information which can be useful for research purpose or analyst to develop better e-commerce strategies. However, it raises significant privacy concern of users data. Data aggregation scheme is generally used for privacy concern. In this paper, for privacy preservation of multiple online users behavior data, we have studied and analyzed multiple techniques. The main objective behind this survey paper is to percepting the existing privacy preservation and analysis technique of user behavior data and to find a better techniques and algorithms as a way to offer excessive efficiency and scalability. Subsequently, several methods for the privacy preservation and analysis have been suggested to attain the highest relevance.

Privacy preserving rule-based classifier using modified artificial bee colony algorithm

A number of privacy preserving techniques have been proposed recently in data mining. In this paper, we provide a review of the state-of-the-art methods for privacy preserving data mining. and discuss methods for randomization, secure multipart computation, and so on. We also make a classification for the privacy preserving data mining technologies, and analyze some works in this field, such as data distortion method for achieving privacy preserving association rule mining. Detailed evaluation criteria of privacy preserving algorithm were illustrated, which include algorithm performance, data utility, and privacy protection degree. Finally, the development of privacy preserving data mining for further directions is given.

In view of present advancements in computing, with the development of distributed environment, many problems have to deal with distributed input data where individual data privacy is the most important issue to be addressed, for the concern of data owner by extending the privacy preserving notion to the original learning algorithms. Privacy Preserving Data Mining has become an active research area in addressing various privacy issues while bringing out solutions for them. There has been lot of progress in developing secure algorithms and models, able to preserve privacy using various data mining techniques like association, classification and clustering, where as importance of privacy preserving techniques applied for learning algorithms related to neural networks for mining problems are still in infancy. Our work in this paper focused on preserving privacy of an individual, using self organizing map (SOM) adopted for collaborative clustering of distributed data between multiple parties. We present Privacy Preserving Collaborative Clustering method using SOM (PPCSOM) for Horizontal Data Distribution, which allows multiple parties perform clustering in a collaborative approach using SOM neural network, without revealing their data directly to each other, in order to preserve privacy of all parties. Our simulation results shows that implementation of PPCSOM method achieves comparable accuracy and assured Privacy than the original non privacy preserving SOM algorithm.

Data Mining has wide applications in many areas such as banking, medicine, scientific research and among government agencies. Classification is one of the commonly used tasks in data mining applications. For the past decade, due to the rise of various privacy issues, many theoretical and practical solutions to the classification problem have been proposed under different security models. However, with the recent popularity of cloud computing, users now have the opportunity to outsource their data, in encrypted form, as well as the data mining tasks to the cloud. Since the data on the cloud is in encrypted form, existing privacy-preserving classification techniques are not applicable. In this paper, we focus on solving the classification problem over encrypted data. In particular, we propose a secure k-NN classifier over encrypted data in the cloud. The proposed protocol protects the confidentiality of data, privacy of user's input query, and hides the data access patterns. To the best of our knowledge, our work is the first to develop a secure k-NN classifier over encrypted data under the semi-honest model. Also, we empirically analyze the efficiency of our proposed protocol using a real-world dataset under different parameter settings.