A blockchain-based audit approach for encrypted data in federated learning

The development of data-driven artificial intelligence technology has given birth to a variety of big data applications. Data has become an essential factor to improve these applications. Federated learning, a privacy-preserving machine learning method, is proposed to leverage data from different data owners. It is typically used in conjunction with cryptographic methods, in which data owners train the global model by sharing encrypted model updates. However, data encryption makes it difficult to identify the quality of these model updates. Malicious data owners may launch attacks such as data poisoning and free-riding. To defend against such attacks, it is necessary to find an approach to audit encrypted model updates. In this paper, we propose a blockchain-based audit approach for encrypted gradients. It uses a behavior chain to record the encrypted gradients from data owners, and an audit chain to evaluate the gradients’ quality. Specifically, we propose a privacy-preserving homomorphic noise mechanism in which the noise of each gradient sums to zero after aggregation, ensuring the availability of aggregated gradient. In addition, we design a joint audit algorithm that can locate malicious data owners without decrypting individual gradients. Through security analysis and experimental evaluation, we demonstrate that our approach can defend against malicious gradient attacks in federated learning.     

Prediction of fish quality level with machine learning

In this study, sea bream, sea bass, anchovy and trout were captured and recorded using a digital camera during refrigerated storage for 7 days. In addition, their total viable counts (TVC) were determined on a daily basis. Based on the TVC, each fish was classified as ‘fresh’ when it was <5 log cfu per g, and as ‘not fresh’ when it was >7 log cfu per g. They were uploaded on a web-based machine learning software called Teachable Machine (TM), which was trained about the pupils and heads of the fish. In addition, images of each species from different angles were uploaded to the software in order to ensure the recognition of fish species by TM. The data of the study indicated that the TM was able to distinguish fish species with high accuracy rates and achieved over 86% success in estimating the freshness of the fish species tested.     

Memory-based dynamic event-triggered control for networked interval type-2 fuzzy systems subject to DoS attacks

In this article, the memory-based dynamic event-triggered controller design issue is investigated for networked interval type-2 (IT2) fuzzy systems under non-periodic denial-of-service (DoS) attacks. For saving limited network bandwidth, a novel memory-based dynamic event-triggered mechanism (DETM) is proposed to schedule data communication. Unlike existing event-triggered generators, the developed memory-based DETM can utilize a series of newly released signals and further save network resources by introducing interval dynamic variables. Moreover, to improve design flexibility, an IT2 fuzzy controller with freely selectable fuzzy rule number and premise membership functions (MFs) is synthesized. Then, a new switched time-delay system with imperfectly matched MFs is established under the consideration of memory-based DETM and DoS attacks simultaneously. Besides, based on the property of MFs, the boundary information of membership grades and slack matrices are introduced in the stability analysis. Furthermore, by using a piecewise Lyapunov–Krasovskii method, membership-functions-dependent criteria are deduced to ensure the asymptotic stability of built fuzzy switched systems. Finally, the effectiveness of proposed control strategies is demonstrated by simulation examples.     

Cheese ripening in nonconventional conditions: A multiparameter study applied to Protected Geographical Indication Canestrato di Moliterno cheese

A multiparameter study was performed to evaluate the effect of fondaco, a traditional ripening cellar without any artificial temperature and relative humidity control, on the chemical, microbiological, and sensory characteristics of Protected Geographical Indication Canestrato di Moliterno cheese. Ripening in such a nonconventional environment was associated with lower counts of lactococci, lactobacilli, and total viable bacteria, and higher presence of enterococci, in comparison with ripening in a controlled maturation room. Moreover, fondaco cheese underwent accelerated maturation, as demonstrated by faster casein degradation, greater accumulation of free AA, and higher formation of volatile organic compounds. Secondary proteolysis, as assessed by liquid chromatography-mass spectrometry of free AA and low molecular weight peptides, did not show any qualitative difference among cheeses, but fondaco samples evidenced an advanced level of peptidolysis. On the other hand, significant qualitative differences were observed in the free fatty acid profiles and in the sensory characteristics. Principal component analysis showed a clear separation of the fondaco and control cheeses, indicating that ripening in the natural room conferred unique sensory features to the product.     

A Tradeoff Between Accuracy and Speed for K-Means Seed Determination

With a sharp increase in the information volume, analyzing and retrieving this vast data volume is much more essential than ever. One of the main techniques that would be beneficial in this regard is called the Clustering method. Clustering aims to classify objects so that all objects within a cluster have similar features while other objects in different clusters are as distinct as possible. One of the most widely used clustering algorithms with the well and approved performance in different applications is the k-means algorithm. The main problem of the k-means algorithm is its performance which can be directly affected by the selection in the primary clusters. Lack of attention to this crucial issue has consequences such as creating empty clusters and decreasing the convergence time. Besides, the selection of appropriate initial seeds can reduce the cluster’s inconsistency. In this paper, we present a new method to determine the initial seeds of the k-mean algorithm to improve the accuracy and decrease the number of iterations of the algorithm. For this purpose, a new method is proposed considering the average distance between objects to determine the initial seeds. Our method attempts to provide a proper tradeoff between the accuracy and speed of the clustering algorithm. The experimental results showed that our proposed approach outperforms the Chithra with 1.7% and 2.1% in terms of clustering accuracy for Wine and Abalone detection data, respectively. Furthermore, achieved results indicate that comparing with the Reverse Nearest Neighbor (RNN) search approach, the proposed method has a higher convergence speed.     

Development of a stakeholder identification and analysis method for human factors integration in work system design interventions – Change Agent Infrastructure

In any work system design intervention—for example, a physical workplace re-design, a work process change, or an equipment upgrade—it is often emphasized how important it is to involve stakeholders in the process of analysis and design, to gain their perspectives as input to the development, and ensure their future acceptance of the solution. While the users of an artifact or workplace are most often regarded as being the most important stakeholders in a design intervention, in a work-system context there may be additional influential stakeholders who influence and negotiate the design intervention's outcomes, resource allocation, requirements, and implementation. Literature shows that it is uncommon for empirical ergonomics and human factors (EHF) research to apply and report the use of any structured stakeholder identification method at all, leading to ad-hoc selections of whom to consider important. Conversely, other research fields offer a plethora of stakeholder identification and analysis methods, few of which seem to have been adopted in the EHF context. This article presents the development of a structured method for identification, classification, and qualitative analysis of stakeholders in EHF-related work system design intervention. It describes the method's EHF-related theoretical underpinnings, lessons learned from four use cases, and the incremental development of the method that has resulted in the current method procedure and visualization aids. The method, called Change Agent Infrastructure (abbreviated CHAI), has a mainly macroergonomic purpose, set on increasing the understanding of sociotechnical interactions that create the conditions for work system design intervention, and facilitating participative efforts.     

Microscopic Methods for Identification of Sulfate-Reducing Bacteria from Various Habitats

This paper is devoted to microscopic methods for the identification of sulfate-reducing bacteria (SRB). In this context, it describes various habitats, morphology and techniques used for the detection and identification of this very heterogeneous group of anaerobic microorganisms. SRB are present in almost every habitat on Earth, including freshwater and marine water, soils, sediments or animals. In the oil, water and gas industries, they can cause considerable economic losses due to their hydrogen sulfide production; in periodontal lesions and the colon of humans, they can cause health complications. Although the role of these bacteria in inflammatory bowel diseases is not entirely known yet, their presence is increased in patients and produced hydrogen sulfide has a cytotoxic effect. For these reasons, methods for the detection of these microorganisms were described. Apart from selected molecular techniques, including metagenomics, fluorescence microscopy was one of the applied methods. Especially fluorescence in situ hybridization (FISH) in various modifications was described. This method enables visual identification of SRB, determining their abundance and spatial distribution in environmental biofilms and gut samples.     

安卓平台上的医用图像信息采集系统的开发设计

为了探讨在安卓平台上构建医用图像采集系统的开发个案,分析通过以智能手机、平板电脑为核心安卓设备通过拍照获得化验单数据后进行文本识别并提交智慧医疗系统的解决方案。本文首先通过二值化算法形成低阈值图像数据,使用卷积神经元网络算法对文本进行逐一识别,使用K-means算法对识别后的单字文本进行字段记录值的整合并形成元数据库服务于其他智慧医疗系统模块。在使用9000组数据对神经元网络进行前期训练的前提下,该系统的识别准确率达到了99.5%以上。本系统具有一定的可行性,对未来智慧医疗的系统开发有实践意义。     

人工智能赋能在线实验教学行为分析

教学行为分析作是教学质量分析的重要组成部分，也是教学引导与反馈机制的重要依据。文章阐述了人工智能（AI）的含义及发展历程，重点分析总结了教学行为分析方法及AI在教学行为分析上的应用。文章以东南大学电工电子在线实验为研究平台，探索分析了AI技术在实验教学行为分析上的可行性，梳理了基于专家系统的在线实验分析系统的设计思路，充分探讨了“智能”教育在实验教学中的深刻内涵。