Supervised Machine Learning Techniques Research Articles

Symbol detection using extreme learning machine network for MU‐SIMO systems

This paper focuses on symbol detection for a multi-user single-input multiple-output (MU-SIMO) uplink using supervised machine learning techniques. An extreme learning machine (ELM) is chosen as the equalizer at the receiver end due to its super-fast learning ability and minimum training error. To tune the parameters, a pilot based online training for an ELM network is proposed, and those parameters are further used to detect the unknown transmitted symbols. Conventionally, the channel matrix is estimated using known pilot symbols explicitly and then the transmitted symbol is recovered using a linear equalizer. Here, instead of directly estimating the channel coefficients, the symbols are detected using an ELM based direct equalizer followed by a hard decision rule. To reduce the pilot requirement of the proposed model, a method is discussed where the channel is estimated using a small number of pilot symbols and the ELM network is trained using a separate training dataset along with the channel estimate. Furthermore, the performance comparison between the proposed ELM based equalizers, linear equalizers, and nonlinear detection techniques is shown.

The Impact of Machine Learning on Total Joint Arthroplasty Patient Outcomes: A Systemic Review

BackgroundSupervised machine learning techniques have been increasingly applied to predict patient outcomes after hip and knee arthroplasty procedures. The purpose of this study was to systematically review the applications of supervised machine learning techniques to predict patient outcomes after primary total hip and knee arthroplasty. MethodsA comprehensive literature search using the electronic databases MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews was conducted in July of 2021. The inclusion criteria were studies that utilized supervised machine learning techniques to predict patient outcomes after primary total hip or knee arthroplasty. ResultsSearch criteria yielded n = 30 relevant studies. Topics of study included patient complications (n = 6), readmissions (n = 1), revision (n = 2), patient-reported outcome measures (n = 4), patient satisfaction (n = 4), inpatient status and length of stay (LOS) (n = 9), opioid usage (n = 3), and patient function (n = 1). Studies involved TKA (n = 12), THA (n = 11), or a combination (n = 7). Less than 35% of predictive outcomes had an area under the receiver operating characteristic curve (AUC) in the excellent or outstanding range. Additionally, only 9 of the studies found improvement over logistic regression, and only 9 studies were externally validated. ConclusionSupervised machine learning algorithms are powerful tools that have been increasingly applied to predict patient outcomes after total hip and knee arthroplasty. However, these algorithms should be evaluated in the context of prognostic accuracy, comparison to traditional statistical techniques for outcome prediction, and application to populations outside the training set. While machine learning algorithms have been received with considerable interest, they should be critically assessed and validated prior to clinical adoption.

Application of Supervised Machine Learning Techniques to Forecast the COVID-19 U.S. Recession and Stock Market Crash.

Machine learning (ML), a transformational technology, has been successfully applied to forecasting events down the road. This paper demonstrates that supervised ML techniques can be used in recession and stock market crash (more than 20% drawdown) forecasting. After learning from strictly past monthly data, ML algorithms detected the Covid-19 recession by December 2019, six months before the official NBER announcement. Moreover, ML algorithms foresaw the March 2020 S&P500 crash two months before it happened. The current labor market and housing are harbingers of a future U.S. recession (in 3 months). Financial factors have a bigger role to play in stock market crashes than economic factors. The labor market appears as a top-two feature in predicting both recessions and crashes. ML algorithms detect that the U.S. exited recession before December 2020, even though the official NBER announcement has not yet been made. They also do not anticipate a U.S. stock market crash before March 2021. ML methods have three times higher false discovery rates of recessions compared to crashes.

Natural Language Processing Application on Commit Messages: A Case Study on HEP Software

Version Control and Source Code Management Systems, such as GitHub, contain a large amount of unstructured historical information of software projects. Recent studies have introduced Natural Language Processing (NLP) to help software engineers retrieve information from a very large collection of unstructured data. In this study, we have extended our previous study by increasing our datasets and machine learning and clustering techniques. We have followed a complex methodology made up of various steps. Starting from the raw commit messages we have employed NLP techniques to build a structured database. We have extracted their main features and used them as input of different clustering algorithms. Once each entry was labelled, we applied supervised machine learning techniques to build a prediction and classification model. We have developed a machine learning-based model to automatically classify commit messages of a software project. Our model exploits a ground-truth dataset that includes commit messages obtained from various GitHub projects belonging to the High Energy Physics context. The contribution of this paper is two-fold: it proposes a ground-truth database and it provides a machine learning prediction model that automatically identifies the more change-prone areas of code. Our model has obtained a very high average accuracy (0.9590), precision (0.9448), recall (0.9382), and F1-score (0.9360).

Hydrogeochemical characterization of the groundwater of Lahore region using supervised machine learning technique.

The cationic and anionic composition in groundwater can be better understood by identifying the type of hydrogeochemical processes influencing groundwater chemistry. This research deals with the characterization of groundwater samples by considering the likely role of hydrogeochemical processes and the factors responsible for the weathering process. The study applies statistical methods and supervised machine learning algorithm (i.e., logistic regression model) on the large data set of 1300 water samples from the Lahore district of Punjab, Pakistan. All the water samples were collected by the local authorities from a deep unconfined aquifer (> 350 ft in depth) for the years of 2005 to 2016. The characterization of groundwater quality parameters includes pH, total dissolved solids (TDS), electrical conductivity (EC), total hardness (TH), calcium (Ca2+), magnesium (Mg2+), sodium (Na+), potassium (K+), chloride (Cl-), bicarbonate (HCO3-), nitrate (NO3-), and sulfate (SO42-). The results show the sequence of the major ion in the following order: Na+ > Ca2+ > Mg+ > K+ and HCO32- > SO42- > Cl- > NO3-. The ionic ratios and Gibb's plot revealed that the prominent hydrogeochemical facies of aquifer water is Ca-HCO3, Ca-Na-HCO3, and mixed Ca-Mg-Cl type rock-weathering process, especially carbonate and silicate weathering, as significant process controlling water chemistry. The statistical evaluation of the prepared regression model determined its prediction accuracy as 92.2%, which means the model is highly efficient and satisfies the analysis. The outcomes of this study favor the utilization of such methods for other areas with large data sets.

Intelligent Real-Time Modelling of Rider Personal Attributes for Safe Last-Mile Delivery to Provide Mobility as a Service

This paper develops an intelligent real-time learning framework for the last-mile delivery of mobility as a service in city planning, based upon safe infrastructure use. Through a hybrid approach integrating statistics and supervised machine learning techniques, knowledge-driven solutions based on the specific user rather than generalized safe mobility practices are suggested. One of the most important aspects influencing transport mode and route selection, and safe infrastructure usage, i.e., the age of the user, is simulated. This is because this variable has been described in the literature as a significant variable. Nonetheless, few works deal with such modelling or the learning system. The learning system was applied in the Northumbria region of England’s northeast as a case study. It comprised four building toolkits: (a) Input toolkit, (b) Safety Predictive toolkit, (c) Variable causation toolkit, and (d) Route choice toolkit. An accurate dynamic road safety model and understanding of the critical parameters influencing bicycle rider safety is created. The developed deep learning model’s average distinguishing power to reliably predict the riskiest age group was 95%, with a standard deviation of 0.02, suggesting a good prediction accuracy across all age groups. According to the study’s findings, different infrastructural networks represent varying risks to bicycle riders of different ages. The rider’s age impacts how other road users engage with them. The regional diversity in trip intent and traffic flow conditions were significant elements influencing the safe use of infrastructure for a specific age group. The study’s findings have the potential to considerably influence infrastructure route selection, modelling, and planning. The constructed model, which integrates the rider’s fragility, sensitivity to externalities, and the varied safety impact dependent on its features, may even be used for the infrastructure still in the planning/design phase. It is envisaged that this research would aid in adopting sustainable (green) transportation options and the last-mile delivery of mobility as a service. Future work should aim to uncover the sensitivities of a rider from different countries and make a baseline comparison scenario.

Urban resilience and livability performance of European smart cities: A novel machine learning approach

Smart cities are centres of economic opulence and hope for standardized living. Understanding the shades of urban resilience and livability in smart city models is of paramount importance. This study presents a novel two-stage data-driven framework combining a multivariate metric-distance analysis with machine learning (ML) techniques for resilience and livability assessment of smart cities. A longitudinal dataset for 35 top-ranked European smart cities from 2015 till 2020 applied as the case study under the proposed framework. Initially, a metric distance-based weighting approach is used to weight the indicators and quantify the scores across each aspect under city resilience and urban livability. The key aspects under city resilience include social, economic, infrastructure and built environment and, institutional resilience, while under urban livability, the aspects include accessibility, community well-being, and economic vibrancy. Fuzzy c-means clustering as an unsupervised machine learning technique is used to sort smart cities based on the degree of performance. In addition, an intelligent approach is presented for the prediction of the degree of livability, resilience, and aggregate performance of smart cities based on various supervised ML techniques. Classification models such as Naïve Bayes, k-nearest neighbors (kNN), support vector machine (SVM), Classification and Regression Tree (CART) and, ensemble models including Random Forest (RF) and Gradient Boosting machine (GBM) were used. Three coefficients (accuracy, Cohen's Kappa (κ) and average area under the precision-recall curve (AUC-PR)) along with confusion matrix were used to appraise the performance of the classifier ML models. The results revealed GBM as the best classification and predictive model for the resilience, livability, and aggregate performance assessment. The study also revealed Copenhagen, Geneva, Stockholm, Munich, Helsinki, Vienna, London, Oslo, Zurich, and Amsterdam as the smart cities that co-create resilience and livability in their development model with superior performance.

Large-scale monitoring of Indo-Pacific finless porpoises (Neophocaena phocaenoides) using multiple static acoustic sensors

The distribution of Indo-Pacific finless porpoise (IPFP, Neophocaena phocaenoides) around the Hainan Island (China) is unknown, but evidences for a resident population have recently emerged. Passive acoustic data were collected from 10 locations along the south-west area of the island using static sensors (about 200 km of coastline), over a total period of 476 days from 25th February 2018 to 16th June 2019 (154,884 5-min recordings). A supervised machine learning technique was used to classify IPFP clicks and acoustic features were measured for 2,654 short-narrowband pulses detected within 47 days across the 10 sites. Detections were verified by a trained operator using spectrogram analysis and the detection range of the clicks was estimated using a Transmission Loss (TL) model. The animals were not detected at all locations and the site with the highest number of detections was characterized by low levels of ambient noise, abundance of soniferous fishes and the presence of another marine mammal species, the Indo-Pacific humpback dolphin (Sousa chinensis). The northernmost region had no detections and only two recorded files containing IPFP clicks were found at the two nearby locations (one file each). In general, the low number of detections in the study area are most likely a result of the limited detection range (about 200 m) of IPFP echolocation signals. This is a nationally protected species and the Hainan population is increasingly subject to many threats. The large-scale and long-term monitoring approach provides information regarding the spatial acoustic occurrence of this vulnerable, nearshore and small-size cetacean.

Augmented Time Regularized Generative Adversarial Network (ATR-GAN) for Data Augmentation in Online Process Anomaly Detection

Supervised machine learning techniques, such as classification models, have been widely applied to online process anomaly detection in advanced manufacturing. However, since abnormal process states rarely occur in regular manufacturing settings, the data collected for model training may be highly imbalanced, which may result in significant training bias for supervised learning and, thus, further deteriorate the anomaly detection accuracy. To reduce the training bias, a natural idea is to incorporate data augmentation techniques to generate effective artificial sample data for the abnormal process states. However, most of the existing data augmentation methods do not effectively consider the temporal orders of the sensor signals, and they also usually require large amounts of actual samples to ensure satisfactory augmentation performance. To address these limitations, this article developed a novel data-driven methodology termed augmented time regularized generative adversarial network (ATR-GAN). By incorporating a proposed augmented generator, ATR-GAN is capable of generating more effective artificial samples for training supervised learning models. The novelty of this augmented generator in the proposed methodology can be summarized into three aspects: 1) an augmented filter layer is introduced in the augmented generator to identify the high-quality artificial samples; 2) in the augmented filter layer, a new distance metric termed time-regularized Hausdorff (TRH) distance is developed to accurately measure the similarity between actual samples and the generated artificial samples; and 3) batching techniques are also employed in the proposed augmented generator to further increase the diversity of the artificial data and fully utilize the relatively limited training data. In addition, the effectiveness of the proposed ATR-GAN is also validated by both numerical simulation and a real-world case study in additive manufacturing. Note to Practitioners—Online process anomaly detection currently plays a significant role in advanced manufacturing since unexpected anomalies may damage product quality and even result in catastrophic loss. In practice, processes are mostly under normal conditions, and anomalies rarely occur. Therefore, the data collected under abnormal conditions are very limited compared to normal conditions, which causes the data imbalanced issue, leading to deterioration in detection accuracy. Many existing data augmentation methods, such as generative adversarial network (GAN), cannot synthesize diversified high-quality artificial samples only using relatively limited actual samples. There is an urgent need in developing an effective data augmentation methodology to address the data imbalanced issue in process anomaly detection. This article developed a novel approach called augmented time regularized GAN (ATR-GAN) for online sensor data augmentation. With this new approach, the applications in additive manufacturing demonstrate that the performance of data augmentation can be improved effectively, and thereafter, the anomaly detection accuracy is also increased significantly. Moreover, the developed methodology is inherently integrated into a generic framework. Thus, it can be further transformed for applications in many other areas that need data augmentation.

Machine Learning Techniques for Device-Free Localization Using Low-Resolution Thermopiles

Indoor device-free localization (DFL) has many uses, including aged care, location-based services, ambient-assisted living, and fire safety management. In recent publications, thermopile sensors (very low-resolution infrared cameras) have been shown as being able to localize individuals while preserving their privacy. This article reports the performance evaluation of a large number of supervised machine learning techniques for the localization of a target using a ceiling-mounted thermopile. The algorithms were trained and validated using a large data set constructed from an individual walking arbitrary paths with the accurate ground truth provided by a virtual reality system. For robust performance evaluation, the algorithms were tested with data sets collected on a different day with several other subjects. A 2-D convolutional neural network exploiting spatial correlation and several recurrent neural network structures exploiting temporal correlation among the captured data provided the most accurate localization performance. Several data sets, constructed from the thermopile’s readings for four individual targets, were made available online for other researchers to use.

DIABETES PREDICTION USING MECHINE LEARNING

Diabetes, sometimes known as diabetes mellitus or just diabetes, is a condition brought on by elevated blood glucose levels. The diagnosis of diabetes can be made using a variety of conventional techniques based on physical and chemical examinations. However, it can be difficult for medical professionals to predict diabetes early.due to the complicated interconnectedness of numerous factors, including how diabetes affects people,kidney, eye, heart, nerves, foot, and so forth. Methods of data science have the potential toaid other scientific disciplines by providing fresh insights into established issues. Among them is toassist in forecasting using medical data. Datascience has an emerging topic called machine learning that studies how machines learn from experience. The purpose of this research is to create a system that can detect diabetes in a patient with a higher risk.by merging the outcomes of many machine learning methods. Goals of this projectusing three different supervised machine learning techniques, such as SVM, ANN and logistic regression (Artificial neural network). This also seeks to suggest a successful a method for diabetic illness early diagnosis.

Exploring Socioeconomic Status as a Global Determinant of COVID-19 Prevalence, Using Exploratory Data Analytic and Supervised Machine Learning Techniques: Algorithm Development and Validation Study.

BackgroundThe COVID-19 pandemic represents the most unprecedented global challenge in recent times. As the global community attempts to manage the pandemic in the long term, it is pivotal to understand what factors drive prevalence rates and to predict the future trajectory of the virus.ObjectiveThis study had 2 objectives. First, it tested the statistical relationship between socioeconomic status and COVID-19 prevalence. Second, it used machine learning techniques to predict cumulative COVID-19 cases in a multicountry sample of 182 countries. Taken together, these objectives will shed light on socioeconomic status as a global risk factor of the COVID-19 pandemic.MethodsThis research used exploratory data analysis and supervised machine learning methods. Exploratory analysis included variable distribution, variable correlations, and outlier detection. Following this, the following 3 supervised regression techniques were applied: linear regression, random forest, and adaptive boosting (AdaBoost). Results were evaluated using k-fold cross-validation and subsequently compared to analyze algorithmic suitability. The analysis involved 2 models. First, the algorithms were trained to predict 2021 COVID-19 prevalence using only 2020 reported case data. Following this, socioeconomic indicators were added as features and the algorithms were trained again. The Human Development Index (HDI) metrics of life expectancy, mean years of schooling, expected years of schooling, and gross national income were used to approximate socioeconomic status.ResultsAll variables correlated positively with the 2021 COVID-19 prevalence, with R2 values ranging from 0.55 to 0.85. Using socioeconomic indicators, COVID-19 prevalence was predicted with a reasonable degree of accuracy. Using 2020 reported case rates as a lone predictor to predict 2021 prevalence rates, the average predictive accuracy of the algorithms was low (R2=0.543). When socioeconomic indicators were added alongside 2020 prevalence rates as features, the average predictive performance improved considerably (R2=0.721) and all error statistics decreased. Thus, adding socioeconomic indicators alongside 2020 reported case data optimized the prediction of COVID-19 prevalence to a considerable degree. Linear regression was the strongest learner with R2=0.693 on the first model and R2=0.763 on the second model, followed by random forest (0.481 and 0.722) and AdaBoost (0.454 and 0.679). Following this, the second model was retrained using a selection of additional COVID-19 risk factors (population density, median age, and vaccination uptake) instead of the HDI metrics. However, average accuracy dropped to 0.649, which highlights the value of socioeconomic status as a predictor of COVID-19 cases in the chosen sample.ConclusionsThe results show that socioeconomic status is an important variable to consider in future epidemiological modeling, and highlights the reality of the COVID-19 pandemic as a social phenomenon and a health care phenomenon. This paper also puts forward new considerations about the application of statistical and machine learning techniques to understand and combat the COVID-19 pandemic.

Machine Learning-Based Prediction of Specific Energy Consumption for Cut-Off Grinding.

Cut-off operation is widely used in the manufacturing industry and is highly energy-intensive. Prediction of specific energy consumption (SEC) using data-driven models is a promising means to understand, analyze and reduce energy consumption for cut-off grinding. The present article aims to put forth a novel methodology to predict and validate the specific energy consumption for cut-off grinding of oxygen-free copper (OFC–C10100) using supervised machine learning techniques. State-of-the-art experimental setup was designed to perform the abrasive cutting of the material at various cutting conditions. First, energy consumption values were predicted on the bases of input process parameters of feed rate, cutting thickness, and cutting tool type using the three supervised learning techniques of Gaussian process regression, regression trees, and artificial neural network (ANN). Among the three algorithms, Gaussian process regression performance was found to be superior, with minimum errors during validation and testing. The predicted values of energy consumption were then exploited to evaluate the specific energy consumption (SEC), which turned out to be highly accurate, with a correlation coefficient of 0.98. The relationship of the predicted specific energy consumption (SEC) with material removal rate agrees well with the relationship depicted in physical models, which further validates the accuracy of the prediction models.

Cultural Differences in an Artificial Representation of the Human Emotional Brain System: A Deep Learning Study

Cultural differences between countries are customarily calculated by comparing mean scores on multiple dimensions, which ignores heterogeneity within societies. This article proposes a fundamentally new way of measuring cultural differences by evaluating the functioning of the emotional brain—that is, by comparing how people think, rather than what they think. Deep learning as a supervised machine learning technique is used to mathematically represent the emotional brain of the Volksgeist (“folk spirit”) with a complex system of nonlinear combinations of value priorities, opinions, and other factors with subjective feelings of well-being. Cross-validating how well one Volksgeist's artificial brain fits another country provides a novel measure for emotional cultural differences, which also considers heterogeneity within a society. This measure is conceptually and statistically unrelated to the Kogut–Singh index calculated from mean scores on cultural dimensions. However, it shows good face validity and supports international marketing and business theory as a negative and highly significant predictor for bilateral trade flows within Europe, using a classic gravity model.

Employment of instrumented vehicles to identify real-time snowy weather conditions on freeways using supervised machine learning techniques – A naturalistic driving study

This study introduces the idea of using vehicles as weather sensors to identify real-time weather on freeways in the same context as Road Weather Information System (RWIS) but in a continuous, trajectory-level, and for road segments allocated in the vehicles paths. The study developed a novel approach to detect snowy and clear weather conditions by utilizing real-time data collected from vehicles' external sensors and CANbus. The proposed approach used time series datasets from the SHRP2 Naturalistic Driving Study (NDS), collected during normal driving conditions on freeways. Trips occurring in snowy weather alongside matched trips in clear weather were segmented into time- and distance-based segments such as a one-minute, one-mile, and half a mile. Three assemblies of the input data are considered in the modeling step: data collected from external sensors, CANbus data, and these two data combined. Data analysis was implemented using the Deep Learning Artificial Neural Network, Decision Tree, Random Forest, and Gradient Boosted Trees models. The results indicate that using different segmentation levels provides decent results in detecting snowy weather. The accuracy in estimating the real-time snowy weather was in ranges of 80% to 85%, 71% to 79%, and 73% to 83% for the one-minute, one-mile, and half mile segmentation types, respectively. The GBT model performed the best among all models based on the area under the Receiver Operating Characteristics (ROC) curve, the highest cumulative percentage in estimating the snowy weather using the lower portion of the population, and the highest overall accuracy. Results indicated that an accuracy of 83% in estimating snowy weather conditions could be accomplished using the data collected from external sensors only without accessing CANbus data.

Predicting and optimizing the heat transfer performance of radiator with functionalized graphene nanofluids

AbstractAccurately predicting the heat transfer characteristics of coolants used in thermal management of energy systems like heat exchangers, power electronics, and heating, ventilation, and air conditioning is indispensable in maintaining its operating conditions within safety limits. Apart from safety, factors such as power consumption and operating cost are the most important constraints to be considered in designing an energy‐efficient and cost‐effective cooling solution. In this study, the experimental data available from previous research on the use of functionalized graphene‐based nanofluids in compact heat exchangers such as the automotive radiator is used to optimize the heat transfer performance parameters like Nusselt number of the nanofluid, the friction factor, and effectiveness of the heat exchanger. A supervised machine learning technique like the artificial neural network is used to obtain the objective functions of the response variables in terms of input features such as Reynolds number, Prandtl number, the volume concentration of nanoparticles in the base fluid, number of transfer units, heat capacity, the density of nanofluid, pressure drop and velocity. On the current dataset, it is found that by using the Bayesian regularization training algorithm and tangent sigmoidal activation function in the neural network, the best accuracies in the prediction can be achieved. Well‐known nature‐inspired optimization algorithms like genetic algorithms and simulated annealing are used in optimizing the above‐mentioned response variables. Both algorithms converged to the same values of the objective functions. The optimum values of Nusselt number, effectiveness, and friction factor are 105.65, 0.506, and 0.0038, respectively, for the given composition of the nanofluid and radiator configuration.

Wavelet transformation and vertical stacking based image classification applying machine learning

The paper aims to classify the people who are not so famous and have a controversial background such as the politicians, the criminals/most wanted people all around India, the social reformers who are bringing the changes necessary for the nation and the sports athletes who are making the nation proud around the globe. This is done to spread awareness among the inhabitants of the country as they will come to know about these unknown people and what they did in the past, and can also be used for investigation to keep a record of the past criminals. Image classification falls under supervised machine learning technique and uses various classification algorithms to identify a particular image label. The aim of the paper is to identify the person and provide information that they did before or achieved something to be so famous to be included in this paper. This is achieved by implementing three major Machine Learning algorithms namely: Logistic Regression, Random Forest and SVM (Support Vector Machine), among which SVM classifier along with wavelet transformation, and vertical stacking of cropped raw pixel image and the cropped wavelet transformed image to achieve maximum precision and accuracy of 94.61%.

Detecting Anomalies in Financial Data Using Machine Learning Algorithms

Bookkeeping data free of fraud and errors are a cornerstone of legitimate business operations. The highly complex and laborious work of financial auditors calls for finding new solutions and algorithms to ensure the correctness of financial statements. Both supervised and unsupervised machine learning (ML) techniques nowadays are being successfully applied to detect fraud and anomalies in data. In accounting, it is a long-established problem to detect financial misstatements deemed anomalous in general ledger (GL) data. Currently, widely used techniques such as random sampling and manual assessment of bookkeeping rules become challenging and unreliable due to increasing data volumes and unknown fraudulent patterns. To address the sampling risk and financial audit inefficiency, we applied seven supervised ML techniques inclusive of deep learning and two unsupervised ML techniques such as isolation forest and autoencoders. We trained and evaluated our models on a real-life GL dataset and used data vectorization to resolve journal entry size variability. The evaluation results showed that the best trained supervised and unsupervised models have high potential in detecting predefined anomaly types as well as in efficiently sampling data to discern higher-risk journal entries. Based on our findings, we discussed possible practical implications of the resulting solutions in the accounting and auditing contexts.

Privacy leakage of LoRaWAN smart parking occupancy sensors

Development of smart cities is enabled by its core concepts of smart and sustainable mobility, where Low Power Wide Area Network (LPWAN) such as Long Range Wide Area Network (LoRaWAN) became one of the most important Internet of Things (IoT) technologies. Due to its low power consumption, simple setup, and large communication range, LoRaWAN smart parking devices are already employed to reduce congestion and improve quality of life. This paper studies privacy leakage of LoRaWAN smart parking communication devices. Namely, when a vehicle as a metallic obstacle obscures the LoRaWAN smart parking device, the signal strength will be significantly reduced on the receiver side. Consequently, the variation in the signal strength of LoRaWAN parking systems transmits information about parking space occupancy, allowing the implementation of a passive side-channel attack at large distances. Using supervised machine learning techniques based on Neural Network, the attacker can estimate parking lot occupancy with accuracy up to 97%, while Random Forrest approach reaches the accuracy over 98%.

Machine-Learning-Based Reconstruction of Turbulent Vortices From Sparse Pressure Sensors in a Pump Sump

Abstract Getting access to the state of turbulent flow from limited sensor measurements in engineering systems is a major challenge. Development of technologies to accurately estimate the state of the flow is now possible with the use of machine learning. We present a supervised machine learning technique to reconstruct turbulent vortical structures in a pump sump from sparse surface pressure measurements. For the current flow reconstruction technique, a combination of multilayer perceptron and three-dimensional convolutional neural network is utilized. This technique provides accurate flow estimation from only a few sensor measurements, identifying the presence of adverse vortices. The dependence of the model performance on the amount of training data, the number of input sensors, and the noise levels are investigated. The present machine learning-based flow estimator supports safe operations of pumps and can be extended to a broad range of applications for industrial fluid-based systems.