Search

Abstract: As the industrial systems are growing complex these-days and data related to the system performance are recorded/collected from various resources under various practical constraints. If the collected data are used as such in the analysis, then they have high range of uncertainties occurred in the analysis and hence performance of the system cannot be done up to desired levels. Thus the main objective of the present work is to remove the uncertainties in the data up to a desired degree of accuracy by utilizing the uncertain, vague and limited data. For analysis of this, an artificial bee colony based Lambda–Tau (ABCBLT) methodology has been used in which expression of the reliability parameters are computed by using Lambda–Tau methodology and their membership functions are formulated by solving a nonlinear optimization problem with artificial bee colony (ABC) algorithm. A time varying failure rate has been used in the analysis instead of constant failure rate. A new RAM-Index has been proposed for ranking the systems’ components based on its performance. The technique has been demonstrated through a case study of press unit of a paper industry, situated in Northern part of India, producing 200tons of paper per day. The results computed by the proposed approach are compared with the Lambda–Tau methodology and concluded that they have a reduced region of prediction in comparison of existing technique region, i.e. uncertainties involved in the analysis are reduced. Thus, it may be a more useful analysis tool to assess the current system conditions and involved uncertainties. [Copyright &y& Elsevier]

Abstract: Objectives: The goal of this study was to provide recommended steps to assess measurement comparability using a crossover study design and to demonstrate these steps using a short patient-reported outcome (PRO) instrument as an example. Methods: The example PRO instrument was administered via paper, Web, interactive voice response system, and interview; a randomized crossover design was used to gather data across the multiple administration types. Participants completed the PRO instrument, demographic and health questions, and a short preference questionnaire. Evaluation included comparisons of the item-level responses and agreement, comparison of mean scale scores, score classifications, and questions designed to collect usability and administration preference. Here the authors provide a four-step evaluation guide to evaluate measurement comparability and illustrate these steps using a case-finding tool. Results: In the example, item-level kappa statistics between the paper and the alternate versions ranged from good to excellent, intraclass correlation coefficient for mean scores were above 0.70, and the rate of disagreement ranged from 2% to 14%. In addition, although participants had an administration preference, they reported few difficulties with the versions they were assigned. Conclusions: The steps described in this article provide a guide for evaluating whether to combine scores across administration versions to simplify analyses and interpretation under a crossover design. The guide recommends the investigation of item-level responses, summary scores, and participant usability/preference when comparing versions, with each step providing unique information to support comprehensive evaluation and informed decisions regarding whether to combine data. [Copyright &y& Elsevier]

Isotope ratio mass spectrometry (IRMS) using carbon isotopes has previously been shown to be a robust and discriminating technique for the comparison of document papers. This study aims to examine the inter and intra sample variability for oxygen isotopes measured in standard 80gsm white document papers, to inform the comparison of document papers in forensic casework. 123 paper samples collected from Australia and New Zealand over a 24-month period were measured for their bulk oxygen isotopic abundance and were found to sit within a range of 15‰. A homogeneity study was undertaken which included examining the variability of samples at the sheet, ream and brand source levels. The results of this study were used to construct guidelines for sample comparison and as such, 95% confidence intervals were observed to be inappropriate for use given the high intra sample variability. Instead, a 1.4‰ discrimination range (0.7‰ either side of the measured value) was defined for use as a benchmark for discrimination when samples were measured in the same sequence. Utilising this value, 82% of the samples could be discriminated using a paired comparison, demonstrating a strong potential for use within forensic casework. [ABSTRACT FROM AUTHOR]

Background: Electronic data collection is increasingly being used for discrete choice experiments (DCEs).Objectives: To study whether paper or electronic administration results in measurement effects.Methods: Respondents were drawn from the same sample frame (an Internet panel) and completed a nearly identical DCE survey either online or on paper during the same period. A DCE on preferences for basic health insurance served as a case study. We used panel mixed logit models for the analysis.Results: In total, 898 respondents completed the survey: 533 respondents completed the survey online, whereas 365 respondents returned the paper survey. There were no significant differences with respect to sociodemographic characteristics between the respondents in both samples. The median response time was shorter for the online sample than for the paper sample, and a smaller proportion of respondents from the online sample were satisfied with the number of choice sets. Although some willingness- to-pay estimates were higher for the online sample, the elicited preferences for basic health insurance characteristics were similar between both modes of administration.Conclusions: We find no indication that online surveys yield inferior results compared with paper-based surveys, whereas the price per respondent is lower for online surveys. Researchers might want to include fewer choice sets per respondent when collecting DCE data online. Because our findings are based on a nonrandomized DCE that covers one health domain only, research in other domains is needed to support our findings. [ABSTRACT FROM AUTHOR]

Objective: This study examines the impact of data collection method on the sociodemographic and health profile of samples of people with diabetes who complete either an online or postal patient-reported outcome measure (PROM) validation survey.Methods: A longitudinal survey of people with diabetes was conducted using online and postal survey versions. The survey consisted of sociodemographic and health questions, a health and self-management PROM (Health and Self-Management in Diabetes [HASMID]), and 5-level version of EQ-5D. Dose adjustment for normal eating Online, Diabetes UK, and social media were used to recruit online survey participants. A panel of patients at a local National Health Service Trust was randomly allocated to participate in either survey version (two-thirds to postal version). Participants were asked to complete the survey again approximately 3 months later.Results: A total of 2784 participants completed the survey (1908 online, 876 postal). The samples (online versus postal) differed; the online sample was younger, with a larger proportion of women and respondents with type 1 diabetes. There were significant differences in sociodemographic characteristics by type of diabetes across data collection mode. The proportion of respondents who responded again at point 2 was higher in the postal sample (525 postal, 698 online).Conclusion: The sociodemographic and health profile of samples of people with diabetes differed depending on whether they completed the online or postal survey. Differences are likely due to different recruitment methods and differences in those choosing to respond to different survey versions. Future PROM validation surveys should select data collection methods carefully because these can affect sample characteristics and results. [ABSTRACT FROM AUTHOR]

• A novel agent marketplace is built on top of blockchain-based Smart Contracts. • The agent marketplace is integrated into an agent composition framework, to support automated agent selection and payment. The chemical industry is increasingly relying on agents for data acquisition, optimization, and simulation. In order to enable efficient management of agents, Knowledge Graphs (KG) together with agent composition frameworks are therefore applied. However, a method to assess the reliability of agents for such systems is absent. Therefore, this paper proposes a Smart Contract-based agent marketplace for composition frameworks to estimate the reliability of agents. In this agent marketplace, we improved the feedback-based reputation system by leveraging Smart Contracts to eliminate fraudulent ratings and to enable automation. The marketplace incorporates a rating-dependent payment mechanism as well, to further enhance trust. The paper also illustrates how this marketplace is integrated into the J-Park Simulator (JPS) agent composition framework for the automated agent selection and transaction. [ABSTRACT FROM AUTHOR]

Quality-relevant process monitoring provides important guarantees for the safety of industrial operation, which is based on the assumption that data collection is complete and low-order autocorrelated. However, real industrial processes always exhibit complex characteristics such as multi-rate sampling and high-order dynamic, which pose great challenges for process monitoring. To this end, a multi-rate high-order dynamic twin-latent-variable probabilistic (MHDTVP) model is presented in this paper to extract data correlations among multi-rate measurements from quality-relevant and irrelevant perspectives. Moreover, to reveal the dynamics in the multi-rate sampling process, an autoregressive twin-latent-variable structure is designed to extract both quality-relevant and quality-irrelevant high-order dynamic features. In the MHDTVP model, parameters are trained through an efficient expectation maximization (EM) iteration framework. Finally, the performance conclusions of MHDTVP are validated with the Tennessee Eastman process (TEP) and Thermal Power Plant (TPP). The experimental results demonstrate that the proposed model exhibits superior monitoring efficiency for multi-rate dynamic processes compared to similar approaches. [Display omitted] • A novel twin-latent-variable probabilistic model is proposed for the purpose of addressing quality-relevant and quality-irrelevant dynamic process monitoring. • To handle measurements with multiple sampling rates, a multi-rate strategy is designed based on the proposed model. • To reveal high-order autocorrelation, the model breaks the bottleneck of traditional first-order dynamic modeling in multi-rate processes. [ABSTRACT FROM AUTHOR]

This paper presents scientific merits and design challenges associated with a novel idea to perform rocket sounding of the Martian atmosphere. Limited data sets of the Mars atmosphere profile are available, while ambitious exploration plans for the next years and decades require achieving game-changing capabilities in terms of extensive atmospheric data acquisition. Planetary landers, orbiters or rovers have limited capabilities regarding gathering data of the atmosphere. Aircraft and rotorcraft are among other solutions for Mars exploration, but they enable conducting soundings only at relatively low altitudes. It is therefore of uttermost importance to ensure the capability for further data acquisition supporting safe and sustainable exploration of Mars. This paper proposes the concept of use of Mars-dedicated sounding rockets. Unlike other concepts, considering rocket propulsion on vehicles capable of doing tasks including i.e. descent and landing, bringing samples of Martian soil to orbit and serving as a long-range Mars surface transportation system, the proposed concept is focused only on conducting ballistic flights with the main purpose of atmospheric research, aiding exploration and potential future activities. Rocket design drivers are discussed and compared with the state-of-the-art approach regarding Earth atmospheric sounding. Means for propelling suborbital flights on Mars are discussed. Optimization of vehicles for a reference mission is described and effects of sizing rockets for Martian application instead of Terrestrial utilization are given. Finally, recommendations for future dedicated Martian sounding rocket developments and a technical scenarios outlook are provided. • Mars atmospheric research history and its current exploration-related objectives. • Growing need of Martian atmospheric research using in-situ measurements. • Rocket payload mass fractions and propulsion system trade offs. • Performance of dedicated sounding vehicles on Mars and on Earth. • Mars mission concept utilizing dozens of sounding rockets within a single lander. [ABSTRACT FROM AUTHOR]

EU-funded project FRANCIS focuses on involving citizens in the development of so-called Frugal Innovations, i.e. simple, affordable and sustainable solutions. It does so by organising open-innovations challenges that are managed through an IT-platform and supported by various recruitment and ideation formats. Usually, Citizen Science only involves citizens for minor tasks such as data collection. FRANCIS jump ahead in science is that the project invites them to propose and work on new solutions that have a real chance to enter the market. Before starting the challenges, the team of FRANCIS already made some assumptions about how to encourage and support participants in the best possible way based on existing publications and first findings from primary research. Now that the first challenge is running, the team had the opportunity to check whether they have proven to be true so far. This paper presents the interim lessons learned and details the way forward [ABSTRACT FROM AUTHOR]

Collecting and structuring data on which to perform specific data analyses is a long and complicated process. It requires substantial advanced planning of what data and what data characteristics will be needed for the study. It is important to specify in what format they should be collected. On the other side, there are huge amounts of data stored in a variety of applications. This data is clean and consistent. In many cases, it can be used for new analyses, different from the original purposes for which the information was collected. This is particularly true for data in healthcare. Medical information accumulated in multiple systems can be used for further research and analysis. This paper presents a methodology, appropriate data models, and relevant software decisions that support the reuse and explorations of medical data for specific treatment assessments based on data routinely collected from clinical systems. Some initial results are presented and discussed. [ABSTRACT FROM AUTHOR]

Automatic detection and prediction of falls specifically in elderly population who live alone can prevent untoward incidents by taking action at appropriate time. Validating these automated systems requires a labeled dataset. The available datasets are mostly based on inertial features. A health-related concern may also be a reason behind a fall. Considering physiological features along with inertial features can increase the accuracy of the system. In this paper, a labeled multiclass dataset is presented based on physiological and Inertial features of the body (PIF v2). Two trials are performed by each participant. The participants are young as well as elderly performing activities and falls in real-time using the prototype containing various sensors designed for collecting inertial as well as physiological features. Two trials are performed by each participant. 27 subjects (15 males and 12 females) between the age 10 years to 75 years participated in data collection. A detailed comparison between PIF v1 where the data were collected individually using each sensor and PIF v2 is given along with statistical analysis of PIF v2. The statistical analysis of various features with a t-test shows that the different activities and falls can be differentiated using inertial as well as physiological features. [ABSTRACT FROM AUTHOR]

IoT devices which include wireless sensors, software, actuators, and computer devices operated through the Internet, enable the transfer of data among objects or people automatically without human intervention. Since these devices are resource constraint embedded devices, security policies are not implemented adequately upon these devices. The connectivity with the Internet, diversity of hardware, varied operating platforms, and surge in attack surface increases the target space for malicious cyber actors. The threat probability increases substantially since the attacker takes advantage of less secure, vulnerable devices to perform the massive-scale attack on the critical infrastructure. It has been observed that the majority of embedded IoT devices operate upon Linux-flavoured operating environments. This paper reviews the Linux-based IoT malware analysis techniques and tools employed for malware detection, analysis, and classification. Various threat data collection methods have been discussed at length and a thorough study of tools and techniques used in static and dynamic analysis of the Linux malware has been provided. A review of the machine learning methods developed using discrete features to classify the malicious program is one of the essential components of this paper. The paper concluded with a discussion on various open issues and challenges that need to be addressed by the research community at large. [ABSTRACT FROM AUTHOR]

• system identification method based on hysteresis loop bias relay feedback without using excitation is designed • hysteresis loop bias relay instead of a controller is proposed • a parameter scheme for the hysteresis loop bias relay to guide the selection of parameters is proposed This paper proposes a closed-loop system identification method based on hysteresis loop bias relay feedback without using excitation. The method utilizes hysteresis loop bias relay instead of a controller to ensure the informativity of data without external excitation and with only noisy feedback. Additionally, this paper proposes a parameter scheme for the hysteresis loop bias relay to guide the selection of parameters more rationally, thus improving the accuracy of the identification model. Compared to recently proposed switching controller methods, the advantage of this method lies in the fact that it requires fewer samples and shortens data collection time. The method's effectiveness is verified through simulation comparisons with typical methods. [ABSTRACT FROM AUTHOR]

• In this paper, a videogrammetric method using four high-speed cameras is proposed to measure the full-field 3D dynamic deformation for evaluating the impact response characteristics of the test article in the full-scale crash test. • A global calibration method based on the high-precision total station is proposed to unify the data of multi-camera videogrammetric system. • The implementation of the proposed method is illustrated through a 6.0 m/s full-scale crash test using a typical civil aircraft approximately 24 m in length. • The results show that the test data are complete and reliable. After the vertical crash at 5.71 m/s, the lower structure of the cabin floor is seriously deformed, and the upper structure of the fuselage in the central wing area is obviously deformed due to the inertia effect of the wing. The stiffness difference of the different fuselage segments results in significant differences in dynamic response. Crashworthiness has been and will continue to be the main concern in aircraft design. Crash tests are essential approaches to investigate the crashworthiness of civil aircraft. Videogrammetric methods have been employed in crash tests to overcome the disadvantages of laborious, time-consuming measurement data collection using contact sensors. However, previous methods using a single pair of cameras still face challenges in large-scale applications due to the inherent contradiction between measurement range and accuracy. In this paper, a videogrammetric method using four high-speed cameras is proposed to measure the full-field three-dimensional (3D) dynamic deformation for evaluating the impact response characteristics of the full-scale crash test aircraft. Increasing the number of high-speed cameras not only can enhance the resolution and then measurement accuracy of object, but also expands the measurement range significantly. To ensure the performance of digital image correlation (DIC) used for large-scale measurement range, a speckle pattern optimization design and fabrication method is presented to generate the random speckle with uniform spraying size and density. A global calibration method based on the high-precision total station is proposed to unify the data of different stereovision subsystems. The implementation of the proposed method is illustrated through a 6.0 m/s full-scale crash test using a civil aircraft approximately 24 m in length. The impact velocity, full-field displacement and strain of the fuselage structure are measured. The results show that the test data are complete and reliable. After the vertical crash at 5.71 m/s, the lower structure of the cabin floor is seriously deformed, and the upper structure of the fuselage in the central wing area is obviously deformed due to the inertia effect of the wing. The stiffness difference of the different fuselage segments results in significant differences in dynamic response. The higher the stiffness is, the smaller the deformation. This work may provide some valuable technical insights that could support the crashworthy design, verification, and certification of civil aircraft. [ABSTRACT FROM AUTHOR]

In today's dynamic business environment, achieving and sustaining operational efficiency is paramount for organizations. To remain competitive and meet evolving customer demands, companies strive to optimize their operations and enhance efficiency. Over time, various indicators have been developed to assess efficiency across multiple operational dimensions, aligning with technological advancements and industrial practices. One such indicator is Overall Equipment Effectiveness (OEE), which evaluates equipment efficiency based on availability, performance, and quality. However, with the advent of the digital industry, the management of data for efficiency calculations has become increasingly complex. In this context, this paper proposes a digitized, real-time OEE calculation system that leverages advanced technologies to continuously monitor OEE, enable real-time decision-making, and drive performance improvements. Harnessing the potential of digitalization, the system facilitates seamless data collection, analysis, and reporting while automating manual processes. By focusing on OEE as a comprehensive efficiency indicator, the proposed system aims to enhance organizational performance and establish a foundation for calculating other efficiency measures. [ABSTRACT FROM AUTHOR]

The implementation of intelligent techniques produces good results in automating fault finding and predicting future outcomes. These approaches have been on the increase in the past years, especially so to detect faults within Compressed Air Systems (CASs). With the use of intelligent techniques, one could minimise the manual and time-consuming aspect of CAS maintenance, improve the environmental impact of the system, while minimising downtime. This paper proposes a general framework for the implementation of intelligent analysis techniques within a real-world system. Such an approach has been implemented on the demand-side of a CAS. In literature, no open datasets are available for use by artificial intelligence models. Hence, as part of this research, a fault generating and monitoring system has been connected to an existing production machine in a manufacturing site to collect the required data. Two classification machine learning methods were implemented and compared across a number of performance metrics. Both the general framework, and its implementation, provide a stepping stone in integrating smart systems with real-time intelligent data analytics for the demand-side of a CAS. These systems would provide a sustainable CAS operation through the effective detection of anomalies and their timely repair. [ABSTRACT FROM AUTHOR]

This paper deals with the state fault detection scheme for distribution flow networks subject to continuously varying conditions at boundaries. A robust PDE detection observer for transport flow systems is designed. Directly built on the nonlinear hyperbolic systems of balance laws model with anti-collocated setup, the PDE observer based on backstepping theory provide the on-line estimation of signals that are not measured. The stability of the error equation is proved. The estimation and the observability time are used for fault detection; an adaptive threshold is defined for the purpose. The performances of the observer and the fault detection method are validated on actual flow data collected from a real water distribution system (WDS) for leakage detection. The leak detection time corresponds to the first alarm activation, confirms the effectiveness of proposed approach. [Display omitted] • Robust persistent state fault detection PDE-based observer for transport flow systems is built. • The nonlinearity derives from a class of hyperbolic systems of balance laws is kept. • Backstepping boundary observer with anti-collocated setup is designed. • The online estimation of unmeasured signals and the observability time are used for fault detection. • An adaptive threshold based on disturbances bound, performance index and input signal is set. • Data are collected from a realwater distribution network. • The performances of the observer and the fault detection scheme are validated on actual flow data. [ABSTRACT FROM AUTHOR]

Emotion recognition based on electroencephalography (EEG) signal features is now one of the booming big data research areas. As the number of commercial EEG devices in the current market increases, there is a need to understand current trends and provide researchers and young practitioners with insights into future investigations of emotion recognition systems. This paper aims to evaluate popular consumer-grade EEG devices' status and review relevant studies that examined the reliability of these low-cost devices for emotion recognition over the last five years. Additionally, a comparison with research-grade devices is conducted. This paper also highlights EEG-based emotion recognition research's key areas, including different feature extraction capabilities, characteristics, and machine learning algorithms. Finally, the main challenges for building an EEG-based emotion recognition system, focusing on the data collection process with commercial EEG devices and machine learning algorithms' performance, are presented. [ABSTRACT FROM AUTHOR]

Technology has become ubiquitous, and humans are increasingly dependent on it. Concepts like Smart and Cyber–Physical Systems (CPS), and the Internet of Things (IoT), are frequently used to describe applications and systems that surround us. These concepts commonly encompass applications that are able to sense their environment, collect and process data, make inferences, and communicate with other applications. However, most of the systems that currently exist do not account for human actions and intents, treating humans as mere sources of data. Furthermore, even the few systems that consider the human factor only do so to a very limited extent. On the premise that technology is made by people for people, new human-centric paradigms are needed, in which emotions, human states, and actions can be conveyed into the system. In this paper, we delve into the concept of Human-in-the-Loop Cyber–Physical Systems (HiTLCPS) and the challenges it presents. Building upon this foundation, we propose a novel approach that integrates humans into all phases of the cyber–physical loop. This comprehensive integration entails influencing human emotions and states, and incorporating human actions into the system architecture and functionalities. Finally, to demonstrate the feasibility and effectiveness of our proposed model, we provide concrete implementation examples in this paper. These examples, along with associated case studies, offer insights into how our human-centric paradigm enhances system performance, user experience, and overall effectiveness. [Display omitted] • Smart and Cyber–Physical Systems (CPS), and IoT are prevalent, but often overlook human actions and intents. • Need for human-centric paradigms in technology design. • Introduction of Human-in-the-Loop Cyber–Physical System (HiTLCPS) concept. • Proposal for a comprehensive approach integrating humans in all phases of the cyber–physical loop. • Concrete implementation examples, research work, and case studies with results. [ABSTRACT FROM AUTHOR]

Strip-shaped Wireless Sensor Networks (WSNs) with branches are commonly used in various long and narrow applications, such as mines, factories, subways, and pipelines, and they face serious energy hole problems caused by multi-hop communication. The Mobile Data Collector (MDC) can alleviate the energy hole problem. Current solutions have two limitations: one is the balance between energy consumption and delay, and the other is the overly ideal network model, e.g., the square region or circular area. This paper focuses on strip-shaped networks and proposes a novel mobile data collection method to find a trade-off between energy preservation and data delivery delay. Firstly, the MDC path is planned by solving the diameter of the tree in the network, resulting in reduced delay. Secondly, the network energy consumption is further reduced by clustering and optimal transmission distance adjustment. Then, a network lifetime balancing mechanism is designed to balance network energy between backbone and branches. Finally, the performance of the algorithm proposed in this paper has been studied in four types of strip-shaped WSNs and compared with four existing MDC methods with evaluation metrics of maximum node energy consumption, network delay and weighted sum of both. The simulation results demonstrate that the proposed algorithm is applicable to different types of strip-shaped WSNs with branches and achieves excellent network performance, which can effectively balance network energy consumption and data collection delay. [ABSTRACT FROM AUTHOR]

This paper provides an overview of crowdsensing and some of its applications. Crowdsensing is a part of the collecting data situations also; it's built on a data system on multiple customer interactions. Moreover, writing the general information of the smart cities can be used to boost to received number frequency to send messages. This work mentioned the Crowdsensing layers that describe Mobile crowdsensing. The article focuses on crowdsensing layers, developed an application in Coalition Game using crowdsensing in terms of GPS. In addition, this paper discussed the Mobile crowdsensing system and how important the cloud is in serving the wireless network, the Internet of things (IoT), and data collection. Furthermore, this research also has developed a smart crowdsensing parking system that will help by reducing the time-wasting users. [ABSTRACT FROM AUTHOR]

In this paper, we developed a novel data collection and analysis approach that uses manual and automated processing of large datasets and combines them to offset the limitations of past research and practice when it comes to identifying actors in complex environmental governance systems. Our empirical context is wildfire governance in Colorado. Wildfire governance represents a pressing challenge for interested actors in the western United States. We characterize Colorado's wildfire governance system by using big data triangulation techniques to identify key actors and their topical domains to reveal meaningful subgroups. Our results highlight the importance of understanding the pitfalls of each source of data and the potential benefits of combining them to gain a more complete understanding of the system. The true potential of multi-source analysis becomes apparent when one pays special attention to the degree of overlap that exists between sources. Each source reveals a unique makeup of actors (both in terms of individuals and organizational types) and different forms of engagement in the system. In addition to providing different sets of actors, we found evidence of different emergent topics in each data source. • We develop a novel method for processing large quantities of publicly available data for actor identification. • Multi-source data triangulation provides specific insights on the structure of the governance system. • Each source reveals a unique makeup of actors, their roles and topical domains in the system. • Minimal overlap between sources supports the need to critically consider biases and pitfalls associated to data sources. [ABSTRACT FROM AUTHOR]

This paper describes the concept of a novel research planned to be carried out in Polish accounting firms providing accounting services to micro and small enterprises. The accounting firms consist of around 36,000 active entities providing services to as many as about 2 million entrepreneurs. The principal objective of this research is to determine the extent of robotic process automation in Polish accounting firms, as well as to identify the benefits of and obstacles to its implementation. The research is to be structured into 6 tasks, while the methods to be applied include a literature review, interviews with the owners and accountants working in accounting firms, survey methods/pen-and-paper personal interview and computer-assisted web interview, as well as raw data collection and statistical analysis. The results of a preliminary pilot study in two accounting firms are also presented in the paper. [ABSTRACT FROM AUTHOR]

One of the most significant sources of information from customers is customer complaints. Successful and effective complaint management can end complaint crises and ensure client loyalty, which is a sign of great service performance. In this paper, we proposed a novel customer temperament-centered and e-CCH system-based data collection and data mining method titled "3D" model for customer complaint data analysis. Three phases are (1) Development and launch of e-Customer Complaint Handling system, (2) Data collection and transfer of learning by e-Customer Complaint Handling system, and (3) Data mining by e-Customer Complaint Handling system. An advanced e lectronic C ustomer C omplaint H andling System called the e-CCH system was then developed and launched. This system adapts the seasonal associations model based on Hippocrates's customer temperament theory to the whole stages of customer complaint reporting and handling. With this system, we conducted a dataset collection work from restaurant chains of two brands over four years. As a result, we collect thousands of real-world temperament-centred customer complaint cases by four years to form the one-of-a-kind CCH dataset. This one-of-a-kind CCH dataset was open-sourced with detailed customer complaint attributes and heuristic decision-making for valuable industrial handling manner. After further analysis of this dataset, we found that customers with different temperament types tend to have different types of complaints. In addition, adapting the temperament theory to the e-CCH system can classify customer types better and provide personalized solutions. To our best knowledge, this rich and the one-of-a-kind CCH dataset reported in this paper is the first comprehensive study of customer complaint handling in an industrial service management context. Meanwhile, data mining with cross analysis and correspondence analysis and an ChatGPT experiment for transfer of learning based on this yearly and one-of-a-kind industrial customer complaint dataset was analyzed and discussed. In addition, how this dataset may contribute to more realistic complaint-handling theoretic studies for better service failure recovery and interactive marketing is discussed in-depth. [ABSTRACT FROM AUTHOR]

The EU has set itself demanding objectives to achieve complete decarbonisation by 2050. The building sector plays an important role for this purpose, hence monitoring its progress towards decarbonisation is essential to identify measures that work and to readjust policies when necessary. However, collecting information to monitor the building sector is a complex task due to the scarcity and low quality of available data. The European digital building logbook (DBL) has great potential in this regard because it allows the collection of data at the building scale, which could be combined at a national level. Thus, this paper proposes a new functionality for the DBL -as an enabler for the collection of data for the assessment of the progress of the main EU policy objectives related to the building sector decarbonisation-, and it evaluates the potential of the DBL for it. To do so, the main policy objectives were associated to a set of progress indicators –mandatory and optional- and matched to DBL indicators. Mandatory indicators are for all Member States and they have been included in the 2023 amended version of the proposal for the Energy Performance of Buildings Directive recast. Optional indicators are considered country-specific in this paper and are sourced from various literature references. Here we have studied the optional progress indicators for the case of Spain. Results show that the DBL has great potential to contribute to the collection of more than half of the mandatory indicators and a very high share of the optional ones for Spain. • Monitoring the decarbonisation progress of national building sectors is crucial. • Mandatory and optional indicators are needed to monitor buildings decarbonisation. • The DBL has great potential for monitoring national building stocks decarbonisation. • The DBL will provide the majority of the decarbonisation progress indicators. [ABSTRACT FROM AUTHOR]

Production planning poses major challenges due to high product diversity, high fluctuations in demand and strong dynamic fluctuations in delivery quantities and times. This makes automation of production planning increasingly necessary to enable optimized sequence planning based on current production data. Production control systems (PRS) offer the possibility of automatic corrective actions in the event of deviations from the specified target dates. Production, PRS and data acquisition are considered as a single control loop and are the subject of various development and research topics. However, automatic interventions in planning and control can lead to serious disruptions if there are errors in the software of the PRS. Therefore, testing and validation of PRS is a crucial factor in development, but is hardly possible in real production environments due to the availability of physical manufacturing facilities and the associated risks. Therefore, in the scope of the paper, a concept for a software framework is proposed to provide a test environment for a PRS. A virtual production environment, which can be combined with the PRS and parameterized in form of a discrete-event simulation, offers the possibility to test the PRS. It is also possible to couple the simulation with a reference calculation to show the behavior of the production environment. The entire framework is designed to enable the identification of critical machines and times as well as, in further steps, the detection of software errors in the PRS. [ABSTRACT FROM AUTHOR]

Current studies on compliment responses are largely confined to the dyadic model of communication, without taking into consideration the influence of third parties present. The most commonly used methods of data collection in compliment response research are discourse completion tasks (DCT), role play, etc., the objectivity and accuracy of which remain unconfirmed. With the method of natural conversation recording and targeted elicitation, this article collects 233 cases of female college students' compliment responses in both dyadic and triadic contexts in Beijing and analyzes the data in the model of "acceptance-mitigation-evasion-rejection". The results show that the frequency of accepting compliments among Chinese college students has increased greatly compared to the findings of Chen (1993), indicating noticeable Western influence on Chinese compliment response strategies. However, there is a significant difference between the acceptance ratio in dyadic and triadic contexts. In dyadic communication, the compliment receivers generally prefer the acceptance strategy, while in triadic contexts, they choose evasion and rejection. This finding indicates that the westernization of compliment response strategies in Chinese culture is a gradual and heterogeneous process and the Self-denigration Maxim still plays an important role in the culture at a deeper level. This study also indicates that the use of DCT tends to raise the ratio of acceptance strategy in general in Chinese compliment response research. • First, the result of this paper shows that in triadic contexts Chinese college students accept much less compliments than they do in dyadic contexts, which challenges a dominating view that under the influence of westernization, Chinese have mostly shifted from rejection to acceptance in responding to compliments. • Second, this paper takes into account the presence of a third party, which calls for and inspires reconsideration of other speech acts such as apology, invitation and so on. • Third, the findings of our study suggest that DCT as a method of data collection systematically increases the percentage of Chinese interactants' acceptance of compliments in naturally-occurring conversations. [ABSTRACT FROM AUTHOR]

In the era of Big Data, the Internet has become one of the main data sources: Data can be collected for relatively low costs and can be used for a wide range of purposes. To be able to timely support solid decisions in any field, it is essential to increase data production efficiency, data accuracy, and reliability. In this framework, our paper aims at identifying an optimized and flexible method to collect and, at the same time, geolocate social media information over a whole country. In particular, the target of this paper is to compare three alternative methods to collect data from the social media Twitter. This is achieved considering four main comparison criteria: Collection time, dataset size, pre-processing phase load, and geographic distribution. Our findings regarding Great Britain identify one of these methods as the best option, since it is able to collect both the highest number of tweets per hour and the highest percentage of unique tweets per hour. Furthermore, this method reduces the computational effort needed to pre-process the collected tweets (e.g., showing the lowest collection times and the lowest number of duplicates within the geographical areas) and enhances the territorial coverage (if compared to the population distribution). At the same time, the effort required to set up this method is feasible and less prone to the arbitrary decisions of the researcher. [ABSTRACT FROM AUTHOR]

Escalation in the real world applications using Internet of Things (IoT) now demands low cost and minimum power consuming devices having sensing, communicating, and computing capabilities. Low-rate Wireless Personal Area Network (LR-WPAN) is an emerging solution which can provide these capabilities with low data rate and less complexity. The longevity and connectivity of the network is a critical issue during the data collection in LR-WPAN. An Energy aware Load Balanced Clustering (ELBC) approach is proposed in this paper that enhances the network life time by minimizing the energy consumption. Clustering is done using the Golden Eagle Optimization (GEO) and by taking connectivity, Average energy of cluster, balancing factor and cohesion as parameters for optimal cluster head selection. Mobile Sink based data collection strategy using Gravitational Search Algorithm (GSA) is used to collect data from clusters. The simulation is done in network simulator-3 and the result shows that, the performance of ELBC-GEO is better in terms of network lifetime, average energy of nodes, and data collection delay. [ABSTRACT FROM AUTHOR]

Energy minimization is one of the significant tasks in the WSN which can be achieved through clustering process. Clustering in the WSN is an efficient method to increase the lifespan of the network. In the proposed work the main concentration is given to the network lifetime which is very necessary for the less node failures and high packet deliveries. In this paper the hierarchical structuring is done using clustering approach based on efficient data aggregation process. The aggregation process is defined into two phases. The first phase defines the arrangement of the nodes in the cluster using uniform distribution and second phase defines the selection of the cluster head and aggregation of the data using linear programming. The main objective is to minimize the energy consumption and increase of the robustness of the sensor network. Eventually the performance of the network will be evaluated in terms of low energy consumption, high packet deliveries and low congestions in the WSN. [ABSTRACT FROM AUTHOR]

• Koopman operator theory is used to establish an explicit control-oriented driver-vehicle dynamic model. • An online hybrid delay-embedded extended dynamic mode decomposition algorithm is proposed. • Time-delay characteristic of driver-vehicle system dynamics is considered. • Multiple local Koopman driver models are constructed to improve driver modelling accuracy. • The Koopman driver model is updated online to capture the driver's adaptive behaviour. Driver-automation shared control has recently proven to be a promising technical scheme to enhance the safety of vehicles, but desire to improve the user acceptance of the shared control systems requires a high-fidelity driver-vehicle dynamic model for characterizing the closed-loop driver-vehicle interaction. In this paper, an emergent pure data-driven modelling approach, i.e., the Koopman operator, is adopted to capture the intrinsic characteristics of the driver-vehicle system dynamics and yield an explicit control-oriented driver-vehicle model in an infinite-dimensional space. Moreover, an online hybrid delay-embedded extended dynamic mode decomposition algorithm is proposed to improve the finite-dimensional approximation accuracy of the Koopman operator. The resulting driver-vehicle model has a simple linear structure which facilitates further shared controller design. Driver-in-the-loop experiments on a driving simulator are conducted to collect driving data for the identification and verification of the proposed modelling approach, and the effectiveness and superiority are demonstrated by experimental results. In particular, it is found that an online updating mechanism for the Koopman model is of great importance to capture the driver's adaptive behaviour in the course of driver-automation interaction. [ABSTRACT FROM AUTHOR]

The blockchain-based access control mechanism (BACM) is gradually becoming an essential paradigm for solving dynamic and trusted access control problems in the open network environment. However, since the current open network environment has such features as dynamic variability and the uncertainty of user identity, most of the existing BACM cannot solve the access control problems in the current open network environment in a dynamic, flexible, proactive, efficient, and fine-grained approach. In this paper, we propose a novel BACM scheme to address such problems. Specifically, we first design a new, proactive, and fine-grained access control model, by utilizing the dynamicity and fine-grain of the attribute-based access control model, flexibility shown by the trust evaluation mechanism in evaluating the trust level of users, and proactivity shown by the game evaluation mechanism in curbing malicious users who suddenly launch malicious access requests. Second, based on the above access control model, we propose a dynamic, flexible, and proactive BACM for the current open network environment, exploiting the trustworthiness and transparency that the smart contract and the transaction mechanism in blockchain technology show during program execution. Further, a double sliding storage window is built, guaranteeing accurate data acquisition by BACM while efficiently allowing it to acquire time-sensitive data during the permission management process. Meanwhile, a pre-authorization concept is introduced to improve the efficiency and flexibility of BACM in processing access control problems. Security analysis demonstrates that our proposed BACM scheme satisfies the simple security issue and the simple availability issue. Experiments on a real user trust record dataset demonstrate the high effectiveness of the proposed BACM scheme in evaluating and deciding on access requests and the superiorities over most existing schemes in dynamicity, fine granularity, flexibility, and proactivity. [ABSTRACT FROM AUTHOR]

• This paper analyzes data collection and representation in CPSS and proposes to use tensors to represent CPSS data. • This paper reviews some representative data fusion methods related to CPSS. • This paper proposes a series of tensor based data fusion methods for CPSS data. • This paper reviews the design of data fusion frameworks and proposes a comprehensive data fusion framework for CPSS. Cyber-Physical-Social systems (CPSSs) are the extension of Cyber-Physical systems (CPS), which seamlessly integrate cyber space, physical space and social space. CPSSs promote the information resource from single space to tri-space, so as to lead a revolution in data science (DS). This paper aims to provide a comprehensive review of data fusion in CPSSs for readers. We firstly analyze data collection and representation in CPSS and propose to use tensors to represent CPSS data, then a general definition of CPSS data fusion is proposed to clarify the concept of information fusion in CPSS. After that, some representative data fusion methods related to CPSS are reviewed. Furthermore, we propose a series of tensor based data fusion methods for CPSS data. Also, we review the design of data fusion frameworks and propose a comprehensive data fusion framework for CPSS. Some challenges and future works are discussed as well. [ABSTRACT FROM AUTHOR]

This article has been retracted: please see Elsevier Policy on Article Withdrawal (https://www.elsevier.com/about/policies/article-withdrawal). This article has been retracted at the request of the Authors. The paper included data that was used without obtaining consent from a team member who actively contributed to the data collection process and therefore this paper is being retracted due to the use of data without permission. Furthermore, the manuscript was submitted to the Journal of Membrane Science without adhering to the journal's authorship policies as outlined in the JMS Submission Guidelines. [ABSTRACT FROM AUTHOR]

The efficient processing of mass stream data has attracted wide attention in the database field. The skyline query on the sensor data stream can monitor multiple targets in real time, to avoid abnormal events such as fire and explosion, which is very useful in the practical application of sensor data monitoring. However, real-world stream data may often contain incomplete data attributes due to faulty sensing devices or imperfect data collection techniques. Skyline queries over incomplete data streams may lead to a lack of transitivity and loop domination issues. To solve the problem of the skyline query over incomplete data streams, firstly, this paper uses differential dependency rule (DD) to fill the missing attribute values of data in the incomplete data stream. Then, the hierarchical grid index (HGrid) is introduced into the field of skyline query to improve pruning efficiency. In the process of skyline calculation, this paper only keeps as few calculation results as possible for the data that may affect the result to avoid a large number of repeated calculations. Thus, S_IDS (Skyline query algorithm over Incomplete Data Stream) is proposed to query skyline results with high confidence from the incomplete data stream. Finally, by comparing with the most advanced skyline query algorithms over incomplete data streams, the correctness and efficiency of the proposed S_IDS algorithm are proved. [ABSTRACT FROM AUTHOR]

In real world, decision-makers' regret psychology often affects decision outcomes due to uncertain risks. Moreover, decision information may be missing in the process of data acquisitions or data storages. Three-way decision has been widely explored in the risky decision-making area by providing effective strategies to divide objects into three mutually disjoint regions. Existing three-way decision methods in fuzzy incomplete information systems rarely consider the influence of decision-makers' psychological states on decision outcomes. In the current paper, we primarily study a new decision-making method that combines regret theory with three-way decision in fuzzy incomplete information systems. First, a prior probability tolerance dominance relation in a fuzzy incomplete information system is defined to handle a binary relation among evaluation values, and a method to calculate objective weights is designed as well. When an incomplete information system does not contain a fuzzy decision attribute value, we put forward a new method to calculate the decision attribute value of each object in the incomplete information system. Then, integrated utility perception values are obtained by combining with regret theory. Further, a regret theory-based three-way decision method with a priori probability tolerance dominance relation is proposed for fuzzy incomplete information systems. At last, the stability and validity of the presented method are verified via corresponding experimental and comparative analysis of realistic cases. • A new algorithm in a fusion decision system is designed. • A prior probability tolerance dominance relation in an FIIS is defined. • A RT-3WD-PPTDR method is established. • RT in behavior psychology is considered. • The feasibility, stability, and superiority of the constructed method are verified. [ABSTRACT FROM AUTHOR]

This paper proposes an incipient chatter detection method to meet high dynamic applications' time and reliability constraints, such as high-speed milling involving heavy noise. The herein introduced method relies on a multiple sampling per revolution (MSPR) technique, coupled with two data preprocessing techniques, a modified adaptive cumulative chatter indicator, and a two-risk levels-based threshold. The MSPR technique enables collecting information-rich enough data to characterize the chatter dynamics thanks to a significant amount of data collected in each revolution. Therefore, the MSPR technique allows for acquiring the data using a short-time window, thus reducing the detection delay. Two data preprocessing techniques, i.e., Z-score normalization and mean-centered, are implemented for data integration and chatter information consolidation. The modified adaptive cumulative chatter indicator has three advantages: (a) it accumulates the information on the chatter feature and highlights the appearance of an incipient chatter; (b) it adapts to the variation of the environmental disturbance noises, resulting in enhanced detection reliability; (c) it is faster than the adaptive cumulative log-likelihood ratio (ACLLR) for decision-making statistically. The two-risk levels-based threshold overcomes the limitations of a unique threshold, and allows simultaneously assessing the two risk levels, thus improving detection reliability. We successfully applied the proposed method to detect incipient chatter in a digital high-speed milling process and assessed its effectiveness by comparing it with several existing chatter detection methods. • A method for incipient chatter detection in high-speed milling process based on the cumulative strategy. • Multiple sampling per revolution technique that consolidates incipient chatter information, reduce the detection delay. • Modified adaptive cumulative chatter indicator enables improving the detection reliability. • Threshold based on two-risk levels to enhancing the detection reliability. [ABSTRACT FROM AUTHOR]

The Mini-EUSO telescope was launched for the International Space Station on August 22 nd , 2019 to observe from the ISS orbit (∼ 400 km altitude) various phenomena occurring in the Earth's atmosphere through a UV-transparent window located in the Russian Zvezda Module. Mini-EUSO is based on a set of two Fresnel lenses of 25 cm diameter each and a focal plane of 48 × 48 pixels, for a total field of view of 44 °. Until July 2021, Mini-EUSO performed a total of 41 data acquisition sessions, obtaining UV images of the Earth in the 290 nm – 430 nm band with temporal and spatial resolution on ground of 2.5 μs and 6.3 × 6.3 km2, respectively. The data acquisition was performed with a 2.5 μs sampling rate, using a dedicated trigger looking for signals with a typical duration of tens of μs. In the present paper the analysis of the performance of the 2.5 μs trigger logic is presented, with a focus on the method used for the analysis and the categories of triggered events. The expected functioning of the trigger logic has been confirmed, with the trigger rate on spurious events that remains within the requirements in nominal background conditions. The trigger logic detected several different phenomena, including lightning strikes, elves, ground-based flashers and events with EAS-like characteristics. [ABSTRACT FROM AUTHOR]

Ultrasound-guided biopsy needle identification is a crucial step in clinical treatment planning, but remains challenging due to the difficulty in data acquisition that includes the ultrasound speckle interference pattern and the presence of strong linear anatomical structure. This paper introduces a real-time needle tracking method for visualizing 2D needle shapes and trajectory during interventions. Based on observations of the needle dynamics within a small fraction of dynamic ultrasound images, the current needle placement was estimated: (1) A subspace based background suppression technique was used to identify points representing possible needle locations using the consecutive dynamic frames in a sliding-window fashion. and (2) a Hough transform was then used to filter out false positives and fit the remaining points on the Hough space. Evaluation on datasets from 16 subjects demonstrated that the proposed method produced high-quality needle-only images, significantly reducing the mean trajectory error to 1.89°and the tip position error to 5.1 mm, outperforming temporal subtraction (2.65°and 14.3 mm) and Gabor filtering (2.94°and 13.8 mm). The attention-based U-Net achieved a comparable mean trajectory angle error of 1.82°but yielded a higher mean tip position error of 8.2 mm. Qualitative and quantitative analyses consistently indicated that the proposed method offers enhanced accuracy and robustness across subjects compared to competing methods. [ABSTRACT FROM AUTHOR]

Intelligent Internet of Things (IoT) provides services for machine learning applications by collecting real-time data, where data correctness crucially impacts the training accuracy of models. Data owners host incremental real-time data in the cloud and use blockchain as a public platform for data query, usage, and certification. Malicious attackers may manipulate data through methods like data poisoning, compromising model parameters. Blockchain-based data integrity verification (auditing) schemes can effectively detect and monitor such activities. However, managing the storage demands for massive on-chain data certificates poses significant challenges. Existing compressed certificate storage strategies, such as the Merkle-Hash tree (MHT), incur significant on-chain storage costs because the auxiliary paths and the entire tree structure must be uploaded during verification, resulting in an O(n) storage overhead. Polynomial commitments offer a promising alternative by reducing storage costs to hundreds of bytes. However, current polynomial commitment methods lack effective support for incremental data updates, which are essential for real-time applications. In this paper, we propose a novel polynomial commitment-based dynamic integrity verification scheme that implements cumulative commitment updates for the first time. Our scheme includes a new data block insertion and deletion strategy that maintains the original order of data blocks within polynomial commitments. Theoretical security analysis confirms the robustness of the proposed scheme, while simulation results demonstrate that it meets the efficiency requirements for blockchain-based distributed data integrity verification mechanisms. [ABSTRACT FROM AUTHOR]

Despite advances in utilizing physiological sensors and machine learning (ML) algorithms, accurately and consistently monitoring heat strain levels of field workers on job sites remains a challenge not fully addressed by previous research. Existing frameworks often fail to adapt to the diverse physiology of workers and their varied working conditions, leading to concerns about stability, reliability, and accuracy. To address these limitations, a worker-centered heat strain monitoring framework was introduced, leveraging physiological data from wearable biosensors for a more accurate and consistent estimation of heat strain risks. A high-fidelity virtual reality (VR) environment was developed to simulate heat-vulnerable occupations for quality data collection. Building on this foundation, EnsmTrBoost, a physiological sensing framework integrating ensemble learning and domain adaptation, was developed. This framework exhibited a remarkable prediction accuracy exceeding 93%. This paper advances heat strain monitoring and supports the development of early warning systems for heat-related fatalities at work sites. • A worker-centered heat stress monitoring framework is proposed. • Developed EnsmTrBoost algorithm predicts heat stress with 93.2% accuracy. • Framework adapts to individual variability, ensuring reliable strain assessment. • This approach can mitigate heat-related health risks among field workers. [ABSTRACT FROM AUTHOR]

Current concrete spalling detection and measurement methods are sparse; despite recent research and commercial offerings using laser scanners, manual measurement is still the industry standard. This paper presents a spalling 3D reconstruction and measurement method. The method uses images illuminated with angled and directional lighting and three neural networks for photometric stereo 3D mesh generation and spalling volume measurement. The proposed method was benchmarked on a laboratory dataset of spalled concrete slabs against a high-resolution laser scanner, yielding an average height error of 0.0 mm and a standard deviation of 1.3 mm. Volume comparisons showed that with manual input, the method achieved a mean absolute percentage error of 22%. Finally, the proposed technique was compared to manual measurements and benchmarked on a spalled concrete structure against a Trimble X12 laser scanner. This research can provide inspectors with increased data interpretability and reduced imaging time for concrete defect mapping. • We use photometric stereo to produce 3D meshes of concrete surfaces. • Data acquisition and processing is fast and human-interpretable. • The 3D meshes are automatically analysed to locate spalling and measure volume. • Our method outperformed the results of research alternatives. • We compared our method to manual and commercial alternatives used in industry. [ABSTRACT FROM AUTHOR]

Odor sensory evaluation is broadly applied in food, clothing, cosmetics, and other fields. Traditional artificial sensory evaluation has poor repeatability, and the machine olfaction represented by the electronic nose (E-nose) is difficult to reflect human feelings. Olfactory electroencephalogram (EEG) contains features associated with human olfactory preference, which has unique advantages in odor sensory evaluation. However, the difficulty of cross-subject olfactory EEG recognition dramatically limits its application. In this paper, a human–machine collaborative multimodal learning method is proposed for cross-subject olfactory preference recognition. Firstly, the olfactory EEG and E-nose multimodal data acquisition and preprocessing paradigms are established. Secondly, a complementary multimodal data mining strategy is proposed to effectively mine the individual and common features representing subjects' olfactory preferences from multimodal data. Finally, the cross-subject olfactory preference recognition is achieved in 24 subjects by fusing the extracted common and individual features, and the recognition effect is superior to the state-of-the-art recognition methods. Furthermore, the advantages of the proposed method in cross-subject olfactory preference recognition indicate its potential for practical odor evaluation applications. [ABSTRACT FROM AUTHOR]

The complexity of high-dimensional and noisy process signals reduces the effectiveness of conventional fault detection methods in industrial processes. Based on the hypothesis that data collected from normal and faulty processes has different characteristics, unsupervised deep neural networks, e.g., autoencoders, have been widely applied in process fault detection and achieved good performance. Many variants have been proposed to improve feature learning by combining different network structures. In this paper, a new transformer model, residual autoencoder-based transformer, is proposed for process fault detection. Firstly, autoencoder and transformer are integrated for better unsupervised feature learning of process signals. Secondly, linear embedding and attention mechanisms with bias are proposed to generate effective features from process signals. Finally, residual connections are constructed between the encoder and decoder of RATransformer to address overfitting in training. Four industrial cases are used to test the performance of RATransformer for process fault detection. The results show that the fault detection rate of RATransformer is at least 1 % higher than other comparison methods. Moreover, the testing results show that the model structure improves the fault detection performance of RATransformer. The complex models like RATransformer can be used in the industrial process when sufficient normal process data is available. An end-to-end training method can be further developed to improve the applicability of RATransformer in process fault detection in the future. • A new transformer model, residual autoencoder-based transformer (RATransformer) is proposed for process fault detection. • The feature learning improves fault detection performance in industrial process. • Residual autoencoder-based transformer provides an effective way for process control. • The effectiveness of RATransformer for fault detection is verified in industrial processes. [ABSTRACT FROM AUTHOR]

This paper introduces an efficient methodology for conducting rat anesthesia experiments, aimed at enhancing the quality of raw brain signals obtained. The proposed approach enables the acquisition of animal brain signals during experiments without the confounding influence of muscle noise. Initially, the use of alpha-chloralose (a-c) in conjunction with Isoflurane is introduced to induce anesthesia in rats. Subsequently, Dexdomitor is administered to prevent muscular movements during the collection of brain signals, further refining the signal quality. Experimental outcomes conclusively demonstrate that our anesthesia method produces cleaner raw signals and exhibits improved robustness during data acquisition, outperforming existing methods that rely solely on Isoflurane or the Ketamine-Xylazine combination. Notably, this improved performance is achieved with minimal alterations to vital physiological parameters, including body temperature, respiration, and heart rates. Moreover, the efficacy of a-c in maintaining anesthesia for up to 7 h stands in contrast to the shorter durations achievable with continuous Isoflurane administration or the 30-min window offered by Ketamine-Xylazine, highlighting the practical advantages of our proposed method. Finally, post-experiment observations confirmed that the animals gradually returned to normal behavior without any signs of distress or adverse effects, indicating that our method was both effective and safe. [ABSTRACT FROM AUTHOR]

In the field of machine learning, the issue of class imbalance is a common problem. It refers to an imbalance in the quantity of data collected, where one class has a significantly larger number of data compared to another class, which can negatively affect the classification efficiency of algorithms. Under-sampling methods address class imbalance by reducing the quantity of data in the majority class, thereby achieving a balanced dataset and mitigating the class imbalance problem. Traditional under-sampling methods based on k-means clustering either set the unified value of k (number of clusters) or determine it directly based on the quantity of data in the minority or majority class. This paper proposes an adaptive k-means clustering under-sampling algorithm that calculates an appropriate k for each dataset. After clustering the majority class dataset into k clusters, our algorithm calculates the distances between the data within each cluster and the cluster centroids from two perspectives and selects data based on these distances. Subsequently, the subset of the majority class dataset are combined with the minority class dataset to generate a new balanced dataset, which is then used for classification algorithms. The performance of our algorithm is evaluated on 45 datasets. Experimental results demonstrate that our algorithm can dynamically determine appropriate k for different datasets and output a balanced dataset, thus enhancing the classification efficiency of machine learning algorithms. This work can provide new algorithmic ensemble strategies for addressing class imbalance problem. [ABSTRACT FROM AUTHOR]