A Hybrid Approach With GAN and DP for Privacy Preservation of IIoT Data

Hindistan, Yavuz Selim; Yetkin, Emrullah Fatih

doi:10.1109/access.2023.3235969

Cited by 14 publications

(3 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their research includes observations on trends and challenges involving the adoption of NLP techniques. Hindistan and Yetkin [35] studied differential privacy and generative adversarial network techniques for processing data in IoT use cases. Their research focused on processing of data with preservation of privacy.…”

Section: Related Workmentioning

confidence: 99%

A hybrid approach for generative process model with topic modelling towards efficient and dynamic document clustering

2023

IJATEE

View full text Add to dashboard Cite

Clustering has potential to generate and organize documents into some meaningful clusters helping in discovering latent structure in presences of unlabelled document corpora [1]. Latent Dirichlet allocation (LDA) is a widely used generative process model to deal with textual corpora in a systematic fashion. It is the model that helps in processing documents and underlying text for solving different problems [2]. Often LDA is used along with clustering techniques like K-means clustering. Another important aspect associated with processing documents is known as dimensionality reduction which improves processing performance.There are many existing methods that focused on document clustering problem. Most of the solutions,

show abstract

Section: Related Workmentioning

confidence: 99%

A hybrid approach for generative process model with topic modelling towards efficient and dynamic document clustering

2023

IJATEE

View full text Add to dashboard Cite

show abstract

“…As a result, other parties may be granted access rights to IIoT data that inappropriately depicts the manufacturing process throughout the data processing. IIoT data may have disguised essential elements, causing information leakage for businesses [77]. Due to these issues, companies cannot share their IIoT data with third parties.…”

Section: • Privacymentioning

confidence: 99%

The Management of IoT-Based Organizational and Industrial Digitalization Using Machine Learning Methods

Darbandi

Javaheri

et al. 2023

Sustainability

View full text Add to dashboard Cite

Recently, the widespread adoption of the Internet of Things (IoT) model has led to the development of intelligent and sustainable industries that support the economic security of modern societies. These industries can offer their participants a higher standard of living and working services via digitalization. The IoT also includes ubiquitous technology for extracting context information to deliver valuable services to customers. With the growth of connected things, the related designs often suffer from high latency and network overheads, resulting in unresponsiveness. The continuous transmission of enormous amounts of sensor data from IoT nodes is problematic because IoT-based sensor nodes are highly energy-constrained. Recently, the research community in the field of IoT and digitalization has labored to build efficient platforms using machine learning (ML) algorithms. ML models that run directly on edge devices are intensely interesting in the context of IoT applications. The use of intelligence ML algorithms in the IoT can automate training, learning, and problem-solving while enabling decision-making based on past data. Therefore, the primary aim of this research is to provide a systematic procedure to review the state-of-the-art on this scope and offer a roadmap for future studies; thus, a structure is introduced for industry sustainability, based on ML methods. The publications were reviewed using a systematic approach that divided the papers into four categories: reinforcement learning, semi-supervised learning, unsupervised learning, and supervised learning. The results showed that ML models could manage IoT-enabled industries efficiently and provide better results compared to other models, with significant differences in learning time and performance. The study findings are considered from a variety of angles concerning the industrial sector’s capacity management of the new elements of Industry 4.0 by combining the industry IoT and ML. Additionally, unique and relevant instructions are provided for the designers of expert intelligent production systems in industrial domains.

show abstract

“…GAN was first introduced to renewable scenario generation in (Chen et al, 2018), and has been used in load generation (Wang et al, 2021), reconstruction of high-temporal-resolution PV generation data (Zhang et al, 2021), etc. Besides, GAN has also been introduced to generating electroencephalographic data (Debie et al, 2020), spatial-temporal data (Qu et al, 2020), and sensitive data in IIoT operations (Hindistan and Yetkin, 2023), etc. which could realize data privacy protection through data generation.…”

Section: Introductionmentioning

confidence: 99%

A data-driven approach for generating load profiles based on InfoGAN and MKDE

Lan,

Zhou,

Guo

et al. 2023

Front. Energy Res.

View full text Add to dashboard Cite

High-quality demand-side management requires an abundance of load profiles to support decision-making processes. However, customer energy consumption data often contains sensitive personal information, and service providers face significant challenges in accessing a substantial amount of energy consumption data. To generate a large volume of customer data without compromising privacy, this study introduces a data-driven approach integrating Information Maximizing Generative Adversarial Networks (InfoGAN) with Multivariate Kernel Density Estimation (MKDE) for the generation of load profiles. InfoGAN is firstly trained based on existing customer load profiles, with the Q network disentangling the load into feature variables and the generator producing realistic profiles. Subsequently, MKDE is utilized to assess the distribution of these features, enabling the generation of new profiles by sampling new feature variables. The proposed method circumvents the need for intricate sampling or modeling processes and generates realistic data that represents the inherent uncertainties and fluctuations characterizing customers’ electricity consumption. The generated data could be used as the substitution for real electricity consumption data, thereby facilitating further applications without compromising privacy concerns.

show abstract

A Hybrid Approach With GAN and DP for Privacy Preservation of IIoT Data

Cited by 14 publications

References 49 publications

A hybrid approach for generative process model with topic modelling towards efficient and dynamic document clustering

A hybrid approach for generative process model with topic modelling towards efficient and dynamic document clustering

The Management of IoT-Based Organizational and Industrial Digitalization Using Machine Learning Methods

A data-driven approach for generating load profiles based on InfoGAN and MKDE

Contact Info

Product

Resources

About