Protecting Your Shopping Preference with Differential Privacy
|Project Title:||Protecting Your Shopping Preference with Differential Privacy.|
|Project Cost: (In Indian Rupees)||Rs.3000/|
|Project Buy Link:||Buy Link|
IEEE BASE PAPER ABSTRACT:
Online banks may disclose consumers’ shopping preferences due to various attacks. With differential privacy, each consumer can disturb his consumption amount locally before sending it to online banks. However, directly applying differential privacy in online banks will incur problems in reality because existing differential privacy schemes do not consider handling the noise boundary problem. In this paper, we propose an Optimized Differential prIvate Online tRansaction scheme (O-DIOR) for online banks to set boundaries of consumption amounts with added noises. We then revise O-DIOR to design a RO-DIOR scheme to select different boundaries while satisfying the differential privacy definition. Moreover, we provide in-depth theoretical analysis to prove that our schemes are capable to satisfy the differential privacy constraint. Finally, to evaluate the effectiveness, we have implemented our schemes in mobile payment experiments. Experimental results illustrate that the relevance between the consumption amount and online bank amount is reduced significantly, and the privacy losses are less than 0.5 in terms of mutual information.
PROJECT OUTPUT VIDEO:
- To protect consumers’ privacy, existing approaches mostly used cryptography. Cryptography schemes mainly utilized encryption technology and authentication technology, which could prevent illegitimate and unauthorized access. However, it is generally difficult for cryptography schemes to handle insider attacks effectively. Insider attackers can still misuse their authorized access to obtain credit statistics and shopping records.
- Zhang et al. proposed differential privacy-preserving schemes for smart meters, limiting the range of noise and capability of batteries.
- Hardt and Talwar gave polynomial time computable upper and lower boundaries on noise complexity and error. The work presented privacy buckets for computing upper and lower boundaries for approximate differential privacy after rfold composition. The paper preserved privacy of individual entries with constrained additive noise and its optimal probability density function could maximize the measure of privacy.
DISADVANTAGES OF EXISTING SYSTEM:
- Directly applying differential privacy in online banks incurs some problems.
- The consumption amount with added noise may be beyond the boundaries after transactions.
- The range of noise under differential privacy is from negative infinity to positive infinity, but in reality the consumption amount with added noise cannot exceed the balance in online bank account, otherwise in the online bank account there is no sufficient deposit to pay for bills. A straightforward method is to delete the noise beyond boundaries and regenerate the noise, but this method would not satisfy the standard definition of differential privacy, so the level of privacy guarantee cannot be controlled.
- Existing differential privacy approaches have not considered setting boundaries on data with added noise
- We propose an optimized differential private online transaction scheme (O-DIOR), in which we define a new noise probability density function. The fundamental strategy is to basically eliminate the probability that noise is generated beyond the boundaries. The scheme can satisfy the differential privacy definition because the noise can be any value in a valid range to avoid the case that the consumption amount and noise can be inferred. Considering the consumption amount may be great and there is not enough money to generate the noise, we propose a revised O-DIOR scheme (RO-DIOR) to select variable boundaries
- To implement the scheme, we design a security module for an online payment application to generate and eliminate the noise to guarantee the utility of consumption amounts. Here we take Apple Pay for example. In our scheme, a consumer uses Apple Pay to pay for his bill, obtaining money from his online bank account and Apple Pay account. Apple Pay does not store consumers’ card numbers and consumption records that can track consumers, so it cannot know consumers’ shopping preferences. Traditionally, Apple Pay directly withdraws money from online banks, our additional step is to use money from consumers’ own Apple Pay accounts, which may not incur more security and trust problems.
- The security module can compute the noise value and assign the consumption amount. For example, a consumer needs to pay $12 to a merchant. Without differential privacy, he needs to withdraw $12 from an online bank, so the actual consumption amount is exposed. With differential privacy, if the security module calculates the noise value as $5 and adds the noise to online bank account, so it needs to withdraw $17 from the online bank, which is not $12 as before. Hence, the personal consumption privacy can be protected. The security module then saves $5 in the Apple Pay to eliminate the added noise, so the actual consumption amount is $12 as before. The consumption record in the online bank is that Apple Pay has withdrawn the money $17 into the consumer’s online bank account, so attackers cannot infer the consumer’s payment amounts and shopping places in online banks.
ADVANTAGES OF PROPOSED SYSTEM:
- Our schemes can protect consumption privacy in online banks under differential privacy. The O-DIOR scheme is designed to limit the range of the consumption amount with added noise. O-DIOR is proven to satisfy the differential privacy constraint.
- The RO-DIOR scheme is further proposed to select variable upper and lower boundaries of consumption amount with added noise in online banks. The RO-DIOR scheme is also proven to satisfy differential privacy constraint.
- The privacy loss is less than 0.5. The performance of schemes have been demonstrated through experiments about different people.
- System : Pentium i3 Processor
- Hard Disk : 500 GB.
- Monitor : 15’’ LED
- Input Devices : Keyboard, Mouse
- Ram : 2 GB
- Operating system : Windows 10.
- Coding Language : Java.
- Tool : Netbeans 8.2
- Database : MYSQL
Jiaping Lin, Jianwei Niu yz Xuefeng Liu, Mohsen Guizani, “Protecting Your Shopping Preference with Differential Privacy”, IEEE Transactions on Mobile Computing, 2021.
PROJECT COST: Rs.3000/
PROJECT BUY LINK: Click here to buy this project in Online.