Privacy preservation for machine learning training and classification based on homomorphic encryption schemes

Respecting Partial Privacy of Unstructured Data via Spectrum-Based Encoder

Since the popularity of Machine Learning as a Service (MLaaS) has been increasing significantly, users are facing the risk of exposing sensitive information that is not task-related. The reason is that the data uploaded by users may include some information that is not useful for inference but can lead to privacy leakage. One straightforward approach to mitigate this issue is to filter out task-independent information to protect user privacy. However, this method is feasible for structured data with naturally independent entries, but it is challenging for unstructured data. Therefore, we propose a novel framework, which employs a spectrum-based encoder to transform unstructured data into the latent space and a task-specific model to identify the essential information for the target task. Our system has been comprehensively evaluated on three benchmark visual datasets and compared to previous works. The results demonstrate that our framework offers superior protection for task-independent information and maintains the usefulness of task-related information.

Privacy-Preserving and Explainable AI in Industrial Applications

The industrial environment has gone through the fourth revolution, also called “Industry 4.0”, where the main aspect is digitalization. Each device employed in an industrial process is connected to a network called the industrial Internet of things (IIOT). With IIOT manufacturers being capable of tracking every device, it has become easier to prevent or quickly solve failures. Specifically, the large amount of available data has allowed the use of artificial intelligence (AI) algorithms to improve industrial applications in many ways (e.g., failure detection, process optimization, and abnormality detection). Although data are abundant, their access has raised problems due to privacy concerns of manufacturers. Censoring sensitive information is not a desired approach because it negatively impacts the AI performance. To increase trust, there is also the need to understand how AI algorithms make choices, i.e., to no longer regard them as black boxes. This paper focuses on recent advancements related to the challenges mentioned above, discusses the industrial impact of proposed solutions, and identifies challenges for future research. It also presents examples related to privacy-preserving and explainable AI solutions, and comments on the interaction between the identified challenges in the conclusions.

Machine Learning for Healthcare Wearable Devices: The Big Picture

Using artificial intelligence and machine learning techniques in healthcare applications has been actively researched over the last few years. It holds promising opportunities as it is used to track human activities and vital signs using wearable devices and assist in diseases' diagnosis, and it can play a great role in elderly care and patient's health monitoring and diagnostics. With the great technological advances in medical sensors and miniaturization of electronic chips in the recent five years, more applications are being researched and developed for wearable devices. Despite the remarkable growth of using smart watches and other wearable devices, a few of these massive research efforts for machine learning applications have found their way to market. In this study, a review of the different areas of the recent machine learning research for healthcare wearable devices is presented. Different challenges facing machine learning applications on wearable devices are discussed. Potential solutions from the literature are presented, and areas open for improvement and further research are highlighted.

Ciphertext Database Audit Technology Under Searchable Encryption Algorithm and Blockchain Technology

The study aims to solve the problems in auditing ciphertext data, improve audit efficiency, and increase the security of audit data in the audit server. First, the existing encryption algorithms are analyzed. Second, the searchable encryption algorithm is proposed to audit the ciphertext data, and an audit server scheme is made based on blockchain technology (BT). Finally, the two schemes are compared with the traditional audit technology. The results show that the server's inspection efficiency of the searchable encryption algorithm is higher.

Differentially private data aggregating with relative error constraint

Privacy preserving methods supporting for data aggregating have attracted the attention of researchers in multidisciplinary fields. Among the advanced methods, differential privacy (DP) has become an influential privacy mechanism owing to its rigorous privacy guarantee and high data utility. But DP has no limitation on the bound of noise, leading to a low-level utility. Recently, researchers investigate how to preserving rigorous privacy guarantee while limiting the relative error to a fixed bound. However, these schemes destroy the statistical properties, including the mean, variance and MSE, which are the foundational elements for data aggregating and analyzing. In this paper, we explore the optimal privacy preserving solution, including novel definitions and implementing mechanisms, to maintain the statistical properties while satisfying DP with a fixed relative error bound. Experimental evaluation demonstrates that our mechanism outperforms current schemes in terms of security and utility for large quantities of queries.

Detecting Cybersecurity Attacks in Internet of Things Using Artificial Intelligence Methods: A Systematic Literature Review

In recent years, technology has advanced to the fourth industrial revolution (Industry 4.0), where the Internet of things (IoTs), fog computing, computer security, and cyberattacks have evolved exponentially on a large scale. The rapid development of IoT devices and networks in various forms generate enormous amounts of data which in turn demand careful authentication and security. Artificial intelligence (AI) is considered one of the most promising methods for addressing cybersecurity threats and providing security. In this study, we present a systematic literature review (SLR) that categorize, map and survey the existing literature on AI methods used to detect cybersecurity attacks in the IoT environment. The scope of this SLR includes an in-depth investigation on most AI trending techniques in cybersecurity and state-of-art solutions. A systematic search was performed on various electronic databases (SCOPUS, Science Direct, IEEE Xplore, Web of Science, ACM, and MDPI). Out of the identified records, 80 studies published between 2016 and 2021 were selected, surveyed and carefully assessed. This review has explored deep learning (DL) and machine learning (ML) techniques used in IoT security, and their effectiveness in detecting attacks. However, several studies have proposed smart intrusion detection systems (IDS) with intelligent architectural frameworks using AI to overcome the existing security and privacy challenges. It is found that support vector machines (SVM) and random forest (RF) are among the most used methods, due to high accuracy detection another reason may be efficient memory. In addition, other methods also provide better performance such as extreme gradient boosting (XGBoost), neural networks (NN) and recurrent neural networks (RNN). This analysis also provides an insight into the AI roadmap to detect threats based on attack categories. Finally, we present recommendations for potential future investigations.

Attribute-Based Adaptive Homomorphic Encryption for Big Data Security

There is a drastic increase in Internet usage across the globe, thanks to mobile phone penetration. This extreme Internet usage generates huge volumes of data, in other terms, big data. Security and privacy are the main issues to be considered in big data management. Hence, in this article, Attribute-based Adaptive Homomorphic Encryption (AAHE) is developed to enhance the security of big data. In the proposed methodology, Oppositional Based Black Widow Optimization (OBWO) is introduced to select the optimal key parameters by following the AAHE method. By considering oppositional function, Black Widow Optimization (BWO) convergence analysis was enhanced. The proposed methodology has different processes, namely, process setup, encryption, and decryption processes. The researcher evaluated the proposed methodology with non-abelian rings and the homomorphism process in ciphertext format. Further, it is also utilized in improving one-way security related to the conjugacy examination issue. Afterward, homomorphic encryption is developed to secure the big data. The study considered two types of big data such as adult datasets and anonymous Microsoft web datasets to validate the proposed methodology. With the help of performance metrics such as encryption time, decryption time, key size, processing time, downloading, and uploading time, the proposed method was evaluated and compared against conventional cryptography techniques such as Rivest-Shamir-Adleman (RSA) and Elliptic Curve Cryptography (ECC). Further, the key generation process was also compared against conventional methods such as BWO, Particle Swarm Optimization (PSO), and Firefly Algorithm (FA). The results established that the proposed method is supreme than the compared methods and can be applied in real time in near future.