Privacy-Preserving Locally Weighted Linear Regression Over Encrypted Millions of Data

Abstract

The emerging development of cloud computing makes a trend that the cloud becomes a outsourced agglomeration for storing big data that generally contains numerous information. To mine the rich value involved in big data, the machine learning methodology is widespread employed due to its ability to adapt to data changes. However, the data mining process may involve the privacy issues of the users, hence they are reluctant to share their information. This is the reason why the outsourced data need to be dealt with securely, where data encryption is considered to be the most straightforward method to keep the privacy of data, but machine learning on the data in ciphertext domain is more complicated than the plaintext, since the relationship structure between data is no longer maintained, in such a way that we focus on the machine learning over encrypted big data. In this work, we study locally weighted linear regression (LWLR), a widely used classic machine learning algorithm in real-world, such as predict and find the best-fit curve through numerous data points. To tackle the privacy concerns in utilizing the LWLR algorithm, we present a system for privacy-preserving locally weighted linear regression, where the system not only protects the privacy of users but also encrypts the best-fit curve. Therefore, we use Paillier homomorphic encryption as the building modular to encrypt data and then apply the stochastic gradient descent in encrypted domain. After given a security analysis, we study how to let Paillier encryption deal with real numbers and implement the system in Python language with a couple of experiments on real-world data sets to evaluate the effectiveness, and show that it outperforms the state-of-the-art and occurs negligible errors compared with performing locally weighted linear regression in the clear. INDEX TERMS Locally weighted linear regression, privacy-preserving, paillier homomorphic encryption, stochastic gradient descent. HAIFENG QIAN received the B.S. and master's degrees in algebraic geometry from the

Moiré‐Driven Topological Transitions and Extreme Anisotropy in Elastic Metasurfaces

“…Pillar‐type resonators have been employed in the design of metamaterials and metasurfaces, notably in the context of bandgaps, [ 50–54 ] cloaking, [ 55 ] and for seismic mitigation. [ 56 ] Nonsymmetric embedded resonators have previously been used to implement elastic hyperbolic metamaterials, [ 57–61 ] whereby large asymmetric couplings within the plate can generate anisotropic effective properties, which can be exploited in the context of waveguiding and subdiffraction imaging. Rather than breaking the resonator symmetry, here we induce strong anisotropy through lattice effects, by spatially modulating the resonant features of the array with a wavelength λ = Na .…”

Preserving Data Confidentiality in Association Rule Mining Using Data Share Allocator Algorithm

“…Their approach draws on a group of sophisticated bilateral reliable computation algorithms. Jıe Chen et al [10] endeavor the privacy anxiety by utilizing the LWLR process. They process a Privacy-preserving privately distorted linear regression scheme system, where their approach encrypts the best-fit curve and protects users' privacy.…”

Enhanced Computation Architecture for Big Data Analytics Using Encryption Algorithm

“…Due to the complexity of cloud computing's ecosystem, there is a growing concern about the usefulness of the Privacy of the cost and data. State-of-the-art encryption methods, but to protect the user's Privacy, exclude the meaningful calculation in the encrypted data [13]. Searchable Encryption (SE) such as encrypted search technology, because it is designed for use in data types of the same type, such as text or digital, a plurality of technologies to a Database Management System (DBMS) 14].…”