Search

The rapid development of the Internet of Things (IoT) has brought many conveniences to our daily life. However, it has also introduced various security risks that need to be addressed. The proliferation of IoT botnets is one of these risks. Most of researchers have had some success in IoT botnet detection using artificial intelligence (AI). However, they have not considered the impact of dynamic network data streams on the models in real-world environments. Over time, existing detection models struggle to cope with evolving botnets. To address this challenge, we propose an incremental learning approach based on Gradient Boosting Decision Trees (GBDT), called GBDT-IL, for detecting botnet traffic in IoT environments. It improves the robustness of the framework by adapting to dynamic IoT data using incremental learning. Additionally, it incorporates an enhanced Fisher Score feature selection algorithm, which enables the model to achieve a high accuracy even with a smaller set of optimal features, thereby reducing the system resources required for model training. To evaluate the effectiveness of our approach, we conducted experiments on the BoT-IoT, N-BaIoT, MedBIoT, and MQTTSet datasets. We compared our method with similar feature selection algorithms and existing concept drift detection algorithms. The experimental results demonstrated that our method achieved an average accuracy of 99.81% using only 25 features, outperforming similar feature selection algorithms. Furthermore, our method achieved an average accuracy of 96.88% in the presence of different types of drifting data, which is 2.98% higher than the best available concept drift detection algorithms, while maintaining a low average false positive rate of 3.02%. [ABSTRACT FROM AUTHOR]

Using implicit responses to determine consumers' response to different stimuli is becoming a popular approach, but research is still needed to understand the outputs of the different technologies used to collect data. During the present research, electroencephalography (EEG) responses and self-reported liking and emotions were collected on different stimuli (odor, taste, flavor samples) to better understand sweetness perception. Artificial intelligence analytics were used to classify the implicit responses, identifying decision trees to discriminate the stimuli by activated sensory system (odor/taste/flavor) and by nature of the stimuli ('sweet' vs. 'non-sweet' odors; 'sweet-taste', 'sweet-flavor', and 'non-sweet flavor'; and 'sweet stimuli' vs. 'non-sweet stimuli'). Significant differences were found among self-reported-liking of the stimuli and the emotions elicited by the stimuli, but no clear relationship was identified between explicit and implicit data. The present research sums interesting data for the EEG-linked research as well as for EEG data analysis, although much is still unknown about how to properly exploit implicit measurement technologies and their data. [ABSTRACT FROM AUTHOR]

Purpose: The aim of this study was to analyze the relevance of asymmetry features between both eyes of the same patient for glaucoma screening using optical coherence tomography. Methods: Spectral-domain optical coherence tomography was used to estimate the thickness of the peripapillary retinal nerve fiber layer in both eyes of the patients in the study. These measurements were collected in a dataset from healthy and glaucoma patients. Several metrics for asymmetry in the retinal nerve fiber layer thickness between the two eyes were then proposed. These metrics were evaluated using the dataset by performing a statistical analysis to assess their significance as relevant features in the diagnosis of glaucoma. Finally, the usefulness of these asymmetry features was demonstrated by designing supervised machine learning models that can be used for the early diagnosis of glaucoma. Results: Machine learning models were designed and optimized, specifically decision trees, based on the values of proposed asymmetry metrics. The use of these models on the dataset provided good classification of the patients (accuracy 88%, sensitivity 70%, specificity 93% and precision 75%). Conclusions: The obtained machine learning models based on retinal nerve fiber layer asymmetry are simple but effective methods which offer a good trade-off in classification of patients and simplicity. The fast binary classification relies on a few asymmetry values of the retinal nerve fiber layer thickness, allowing their use in the daily clinical practice for glaucoma screening. [ABSTRACT FROM AUTHOR]

Electroencephalography (EEG) is a brain imaging technique in which electrodes are placed on the scalp. EEG signals are commonly decomposed into frequency bands called delta, theta, alpha, and beta. While these bands have been shown to be useful for characterizing various brain states, their utility as a one-size-fits-all analysis tool remains unclear. The goal of this work is to outline an objective strategy for discovering optimal EEG bands based on signal power spectra. A two-step data-driven methodology is presented for objectively determining the best EEG bands for a given dataset. First, a decision tree is used to estimate the optimal frequency band boundaries for reproducing the signal's power spectrum for a predetermined number of bands. The optimal number of bands is then determined using an Akaike Information Criterion (AIC)-inspired quality score that balances goodness-of-fit with a small band count. This data-driven approach led to better characterization of the underlying power spectrum by identifying bands that outperformed the more commonly used band boundaries by a factor of two. Additionally, key spectral components were isolated in dedicated frequency bands. The proposed method provides a fully automated and flexible approach to capturing key signal components and possibly discovering new indices of brain activity. [ABSTRACT FROM AUTHOR]

To address the limitations of LiDAR dynamic target detection methods, which often require heuristic thresholding, indirect computational assistance, supplementary sensor data, or postdetection, we propose an innovative method based on multidimensional features. Using the differences between the positions and geometric structures of point cloud clusters scanned by the same target in adjacent frame point clouds, the motion states of the point cloud clusters are comprehensively evaluated. To enable the automatic precision pairing of point cloud clusters from adjacent frames of the same target, a double registration algorithm is proposed for point cloud cluster centroids. The iterative closest point (ICP) algorithm is employed for approximate interframe pose estimation during coarse registration. The random sample consensus (RANSAC) and four-parameter transformation algorithms are employed to obtain precise interframe pose relations during fine registration. These processes standardize the coordinate systems of adjacent point clouds and facilitate the association of point cloud clusters from the same target. Based on the paired point cloud cluster, a classification feature system is used to construct the XGBoost decision tree. To enhance the XGBoost training efficiency, a Spearman's rank correlation coefficient-bidirectional search for a dimensionality reduction algorithm is proposed to expedite the optimal classification feature subset construction. After preliminary outcomes are generated by XGBoost, a double Boyer–Moore voting-sliding window algorithm is proposed to refine the final LiDAR dynamic target detection accuracy. To validate the efficacy and efficiency of our method in LiDAR dynamic target detection, an experimental platform is established. Real-world data are collected and pertinent experiments are designed. The experimental results illustrate the soundness of our method. The LiDAR dynamic target correct detection rate is 92.41%, the static target error detection rate is 1.43%, and the detection efficiency is 0.0299 s. Our method exhibits notable advantages over open-source comparative methods, achieving highly efficient and precise LiDAR dynamic target detection. [ABSTRACT FROM AUTHOR]

This paper presents a state-of-the-art estimation technique by cross-combining a number n of filters for high-precision, reliable and robust vehicle sideslip angle state estimation, over a full range of vehicle operations irrespective of the driving mission and disruptions that may occur in the system. A machine-learning algorithm based on decision trees connects several filters together to switch between them according to the driving context, ensuring the best possible state estimate for relatively small and large sideslip angle values. In conjunction with the above-mentioned aspects, a seamless transition between different vehicle models is attained by observing the key parameters characterizing the lateral motion of the vehicle. The tests conducted using a prototype vehicle on a snow-covered track confirm the effectiveness and reliability of the proposed approach. [ABSTRACT FROM AUTHOR]

This study addresses the problem of accurately predicting azimuth and elevation angles of signals impinging on an antenna array employing Machine Learning (ML). Using the information obtained at a receiving system when a transmitter's signal hits it, a Decision Tree (DT) model is trained to estimate azimuth and elevation angles simultaneously. Simulation results demonstrate the robustness of the proposed DT-based method, showcasing its ability to predict the Direction of Arrival (DOA) in diverse conditions beyond the ones present in the training dataset, i.e., the results display the model's generalization capability. Additionally, the comparative analysis reveals that DT-based DOA estimation outperforms the state-of-the-art MUltiple SIgnal Classification (MUSIC) algorithm. Our results demonstrate an average reduction of over 90% in the prediction error and 50% in the prediction time achieved by our proposal when compared to the MUSIC algorithm. These results establish DTs as competitive alternatives for DOA estimation in signal reception systems. [ABSTRACT FROM AUTHOR]

In recent years, human activity recognition (HAR) has gained significant interest from researchers in the sports and fitness industries. In this study, the authors have proposed a cascaded method including two classifying stages to classify fitness exercises, utilizing a decision tree as the first stage and a one-dimension convolutional neural network as the second stage. The data acquisition was carried out by five participants performing exercises while wearing an inertial measurement unit sensor attached to a wristband on their wrists. However, only data acquired along the z-axis of the IMU accelerator was used as input to train and test the proposed model, to simplify the model and optimize the training time while still achieving good performance. To examine the efficiency of the proposed method, the authors compared the performance of the cascaded model and the conventional 1D-CNN model. The obtained results showed an overall improvement in the accuracy of exercise classification by the proposed model, which was approximately 92%, compared to 82.4% for the 1D-CNN model. In addition, the authors suggested and evaluated two methods to optimize the clustering outcome of the first stage in the cascaded model. This research demonstrates that the proposed model, with advantages in terms of training time and computational cost, is able to classify fitness workouts with high performance. Therefore, with further development, it can be applied in various real-time HAR applications. [ABSTRACT FROM AUTHOR]

There are high concentrations of urban spaces and increasingly complex land use types. Providing an efficient and scientific identification of building types has become a major challenge in urban architectural planning. This study used an optimized gradient-boosted decision tree algorithm to enhance a decision tree model for building classification. Through supervised classification learning, machine learning training was conducted using a business-type weighted database. We innovatively established a form database to store input items. During parameter optimization, parameters such as the number of nodes, maximum depth, and learning rate were gradually adjusted based on the performance of the verification set to achieve optimal performance on the verification set under the same conditions. Simultaneously, a k-fold cross-validation method was used to avoid overfitting. The model clusters trained in the machine learning training corresponded to various city sizes. By setting the parameters to determine the size of the area of land for a target city, the corresponding classification model could be invoked. The experimental results show that this algorithm has high accuracy in building recognition. Especially in R, S, and U-class buildings, the overall accuracy rate of recognition reaches over 94%. [ABSTRACT FROM AUTHOR]

In this paper, we propose a fast labeling algorithm based on block-based concepts. Because the number of memory access points directly affects the time consumption of the labeling algorithms, the aim of the proposed algorithm is to minimize neighborhood operations. Our algorithm utilizes a block-based view and correlates a raster scan to select the necessary pixels generated by a block-based scan mask. We analyze the advantages of a sequential raster scan for the block-based scan mask, and integrate the block-connected relationships using two different procedures with binary decision trees to reduce unnecessary memory access. This greatly simplifies the pixel locations of the block-based scan mask. Furthermore, our algorithm significantly reduces the number of leaf nodes and depth levels required in the binary decision tree. We analyze the labeling performance of the proposed algorithm alongside that of other labeling algorithms using high-resolution images and foreground images. The experimental results from synthetic and real image datasets demonstrate that the proposed algorithm is faster than other methods. [ABSTRACT FROM AUTHOR]

Due to the rapid development of machine-learning technology, companies can build complex models to provide prediction or classification services for customers without resources. A large number of related solutions exist to protect the privacy of models and user data. However, these efforts require costly communication and are not resistant to quantum attacks. To solve this problem, we designed a new secure integer-comparison protocol based on fully homomorphic encryption and proposed a client-server classification protocol for decision-tree evaluation based on the secure integer-comparison protocol. Compared to existing work, our classification protocol has a relatively low communication cost and requires only one round of communication with the user to complete the classification task. Moreover, the protocol was built on a fully homomorphic-scheme-based lattice that is resistant to quantum attacks, as opposed to conventional schemes. Finally, we conducted an experimental analysis comparing our protocol with the traditional approach on three datasets. The experimental results showed that the communication cost of our scheme was 20 % of the cost of the traditional scheme. [ABSTRACT FROM AUTHOR]

A recent increase in the importance of Rare Earth Elements (REEs), proportional to advancements in modern technology, green energy, and defense, has urged researchers to look for more sophisticated and efficient exploration methods for their host rocks, such as carbonatites. Hyperspectral remote sensing has long been recognized as having great potential to identify the REEs based on their sharp and distinctive absorption features in the visible near-infrared (VNIR) and shortwave infrared (SWIR) electromagnetic spectral profiles. For instance, neodymium (Nd), one of the most abundant Light Rare Earth Elements (LREEs), has among the most distinctive absorption features of REEs in the VNIR part of the electromagnetic spectrum. Centered at ~580, ~745, ~810, and ~870 nm in the VNIR, the positions of these absorption features have been proved to be independent of the mineralogy that hosts Nd, and the features can be observed in samples as low in Nd as 1000 ppm. In this study, a neodymium index (NI) is proposed based on the 810 nm absorption feature and tested on the hyperspectral images of the Sillai Patai carbonatite samples to identify Nd pixels and to decipher the relative concentration of Nd in the samples based on the depth of the absorption feature. A preliminary spectral study of the carbonatite samples was carried out using a spectroradiometer to determine the presence of Nd in the samples. Only two of the absorption features of Nd, centered at ~745 and ~810 nm, are prominent in the Nd-rich samples. The other absorption features are either weak or suppressed by the featureless spectra of the associated minerals. Similar absorption features are found in the VNIR and SWIR images of the rock samples captured by the laboratory-based hyperspectral cameras that are processed through Minimum Noise Fraction (MNF) and Fast Fourier Transform (FFT) to filter the signal and noise from the reflectance data. An RGB false-color composite of continuum-removed VNIR reflectance bands covering wavelengths of 587.5, 747.91, and 810.25 nm efficiently displayed the spatial distribution of Nd-rich hotspots in the hyperspectral image. The depth of the 810 nm absorption feature, which corresponds to the concentration of Nd in a pixel, is comparatively greater in these zones and is quantified using the proposed NI such that the deeper the absorption feature, the higher the NI. To quantify the Nd-rich pixels in the continuum-removed VNIR images, different threshold values of NI are introduced into a decision tree classifier which generates the number of pixels in each class. The strength of the proposed NI coupled with the decision tree classifier is further supported by the accuracy assessment of the classified images generating the Kappa coefficient of 0.82. Comparing the results of the remote sensing data obtained in this study with some of the previously published studies suggests that the Sillai Patti carbonatite is rich in Nd and associated REEs, with some parts of the samples as high in Nd concentration as 1000 ppm. [ABSTRACT FROM AUTHOR]

COVID-19 occurs due to infection through respiratory droplets containing the SARS-CoV-2 virus, which are released when someone sneezes, coughs, or talks. The gold-standard exam to detect the virus is Real-Time Polymerase Chain Reaction (RT-PCR); however, this is an expensive test and may require up to 3 days after infection for a reliable result, and if there is high demand, the labs could be overwhelmed, which can cause significant delays in providing results. Biomedical data (oxygen saturation level—SpO2, body temperature, heart rate, and cough) are acquired from individuals and are used to help infer infection by COVID-19, using machine learning algorithms. The goal of this study is to introduce the Integrated Portable Medical Assistant (IPMA), which is a multimodal piece of equipment that can collect biomedical data, such as oxygen saturation level, body temperature, heart rate, and cough sound, and helps infer the diagnosis of COVID-19 through machine learning algorithms. The IPMA has the capacity to store the biomedical data for continuous studies and can be used to infer other respiratory diseases. Quadratic kernel-free non-linear Support Vector Machine (QSVM) and Decision Tree (DT) were applied on three datasets with data of cough, speech, body temperature, heart rate, and SpO2, obtaining an Accuracy rate (ACC) and Area Under the Curve (AUC) of approximately up to 88.0% and 0.85, respectively, as well as an ACC up to 99% and AUC = 0.94, respectively, for COVID-19 infection inference. When applied to the data acquired with the IMPA, these algorithms achieved 100% accuracy. Regarding the easiness of using the equipment, 36 volunteers reported that the IPMA has a high usability, according to results from two metrics used for evaluation: System Usability Scale (SUS) and Post Study System Usability Questionnaire (PSSUQ), with scores of 85.5 and 1.41, respectively. In light of the worldwide needs for smart equipment to help fight the COVID-19 pandemic, this new equipment may help with the screening of COVID-19 through data collected from biomedical signals and cough sounds, as well as the use of machine learning algorithms. [ABSTRACT FROM AUTHOR]

One of the essential requirements of injection molding is to ensure the stable quality of the parts produced. However, numerous processing conditions, which are often interrelated in quite a complex way, make this challenging. Machine learning (ML) algorithms can be the solution, as they work in multidimensional spaces by learning the structure of datasets. In this study, we used four ML algorithms (kNN, naïve Bayes, linear discriminant analysis, and decision tree) and compared their effectiveness in predicting the quality of multi-cavity injection molding. We used pressure-based quality indexes (features) as inputs for the classification algorithms. We proved that all the examined ML algorithms adequately predict quality in injection molding even with very little training data. We found that the decision tree algorithm was the most accurate one, with a computational time of only 8–10 s. The average performance of the decision tree algorithm exceeded 90%, even for very little training data. We also demonstrated that feature selection does not significantly affect the accuracy of the decision tree algorithm. [ABSTRACT FROM AUTHOR]

This paper presents different prediction methods based on decision tree and ensemble learning to suggest possible next assembly steps. The predictor is designed to be a component of a sensor-based assembly assistance system whose goal is to provide support via adaptive instructions, considering the assembly progress and, in the future, the estimation of user emotions during training. The assembly assistance station supports inexperienced manufacturing workers, but it can be useful in assisting experienced workers, too. The proposed predictors are evaluated on the data collected in experiments involving both trainees and manufacturing workers, as well as on a mixed dataset, and are compared with other existing predictors. The novelty of the paper is the decision tree-based prediction of the assembly states, in contrast with the previous algorithms which are stochastic-based or neural. The results show that ensemble learning with decision tree components is best suited for adaptive assembly support systems. [ABSTRACT FROM AUTHOR]

There is an evident increase in the importance that remote sensing sensors play in the monitoring and evaluation of natural hazards susceptibility and risk. The present study aims to assess the flash-flood potential values, in a small catchment from Romania, using information provided remote sensing sensors and Geographic Informational Systems (GIS) databases which were involved as input data into a number of four ensemble models. In a first phase, with the help of high-resolution satellite images from the Google Earth application, 481 points affected by torrential processes were acquired, another 481 points being randomly positioned in areas without torrential processes. Seventy percent of the dataset was kept as training data, while the other 30% was assigned to validating sample. Further, in order to train the machine learning models, information regarding the 10 flash-flood predictors was extracted in the training sample locations. Finally, the following four ensembles were used to calculate the Flash-Flood Potential Index across the Bâsca Chiojdului river basin: Deep Learning Neural Network–Frequency Ratio (DLNN-FR), Deep Learning Neural Network–Weights of Evidence (DLNN-WOE), Alternating Decision Trees–Frequency Ratio (ADT-FR) and Alternating Decision Trees–Weights of Evidence (ADT-WOE). The model's performances were assessed using several statistical metrics. Thus, in terms of Sensitivity, the highest value of 0.985 was achieved by the DLNN-FR model, meanwhile the lowest one (0.866) was assigned to ADT-FR ensemble. Moreover, the specificity analysis shows that the highest value (0.991) was attributed to DLNN-WOE algorithm, while the lowest value (0.892) was achieved by ADT-FR. During the training procedure, the models achieved overall accuracies between 0.878 (ADT-FR) and 0.985 (DLNN-WOE). K-index shows again that the most performant model was DLNN-WOE (0.97). The Flash-Flood Potential Index (FFPI) values revealed that the surfaces with high and very high flash-flood susceptibility cover between 46.57% (DLNN-FR) and 59.38% (ADT-FR) of the study zone. The use of the Receiver Operating Characteristic (ROC) curve for results validation highlights the fact that FFPIDLNN-WOE is characterized by the most precise results with an Area Under Curve of 0.96. [ABSTRACT FROM AUTHOR]

Smart production ecosystems are a valuable target for attackers. In particular, due to the high level of connectivity introduced by Industry 4.0, attackers can potentially attack individual components of production systems from the outside. One approach to strengthening the security of industrial control systems is to perform black box security tests such as network fuzzing. These are applicable, even if no information on the internals of the control system is available. However, most security testing strategies assume a gray box setting, in which some information on the internals are available. We propose a new approach to bridge the gap between these gray box strategies and the real-world black box setting in the domain of industrial control systems. This approach involves training an adaptive machine learning model that approximates the information that is missing in a black box setting. We propose three different approaches for the model, combine them with an evolutionary testing approach, and perform an evaluation using a System under Test with known vulnerabilities. Our evaluation shows that the model is indeed able to learn valuable information about a previously unknown system, and that more vulnerabilities can be uncovered with our approach. The model-based approach using a Decision Tree was able to find a significantly higher number of vulnerabilities than the two baseline fuzzers. [ABSTRACT FROM AUTHOR]

With the advancement of engineering techniques, underground shield tunneling projects have also started incorporating emerging technologies to monitor the forces and displacements during the construction and operation phases of shield tunnels. Monitoring devices installed on the tunnel segment components generate a large amount of data. However, due to various factors, data may be missing. Hence, the completion of the incomplete data is imperative to ensure the utmost safety of the engineering project. In this research, a missing data imputation technique utilizing Random Forest (RF) is introduced. The optimal combination of the number of decision trees, maximum depth, and number of features in the RF is determined by minimizing the Mean Squared Error (MSE). Subsequently, complete soil pressure data are artificially manipulated to create incomplete datasets with missing rates of 20%, 40%, and 60%. A comparative analysis of the imputation results using three methods—median, mean, and RF—reveals that this proposed method has the smallest imputation error. As the missing rate increases, the mean squared error of the Random Forest method and the other two methods also increases, with a maximum difference of about 70%. This indicates that the random forest method is suitable for imputing monitoring data. [ABSTRACT FROM AUTHOR]

With the massive, worldwide, smart metering roll-out, both energy suppliers and users are starting to tap into the potential of higher resolution energy readings for accurate billing, improved demand response, improved tariffs better tuned to users and the grid, and empowering end-users to know how much their individual appliances contribute to their electricity bills via nonintrusive load monitoring (NILM). A number of NILM approaches, based on machine learning (ML), have been proposed over the years, focusing on improving the NILM model performance. However, the trustworthiness of the NILM model itself has hardly been addressed. It is important to explain the underlying model and its reasoning to understand why the model underperforms in order to satisfy user curiosity and to enable model improvement. This can be done by leveraging naturally interpretable or explainable models as well as explainability tools. This paper adopts a naturally interpretable decision tree (DT)-based approach for a NILM multiclass classifier. Furthermore, this paper leverages explainability tools to determine local and global feature importance, and design a methodology that informs feature selection for each appliance class, which can determine how well a trained model will predict an appliance on any unseen test data, minimising testing time on target datasets. We explain how one or more appliances can negatively impact classification of other appliances and predict appliance and model performance of the REFIT-data trained models on unseen data of the same house and on unseen houses on the UK-DALE dataset. Experimental results confirm that models trained with the explainability-informed local feature importance can improve toaster classification performance from 65% to 80%. Additionally, instead of one five-classifier approach incorporating all five appliances, a three-classifier approach comprising a kettle, microwave, and dishwasher and a two-classifier comprising a toaster and washing machine improves classification performance for the dishwasher from 72% to 94% and the washing machine from 56% to 80%. [ABSTRACT FROM AUTHOR]

This study aims to analyze the asymmetry between both eyes of the same patient for the early diagnosis of glaucoma. Two imaging modalities, retinal fundus images and optical coherence tomographies (OCTs), have been considered in order to compare their different capabilities for glaucoma detection. From retinal fundus images, the difference between cup/disc ratio and the width of the optic rim has been extracted. Analogously, the thickness of the retinal nerve fiber layer has been measured in spectral-domain optical coherence tomographies. These measurements have been considered as asymmetry characteristics between eyes in the modeling of decision trees and support vector machines for the classification of healthy and glaucoma patients. The main contribution of this work is indeed the use of different classification models with both imaging modalities to jointly exploit the strengths of each of these modalities for the same diagnostic purpose based on the asymmetry characteristics between the eyes of the patient. The results show that the optimized classification models provide better performance with OCT asymmetry features between both eyes (sensitivity 80.9%, specificity 88.2%, precision 66.7%, accuracy 86.5%) than with those extracted from retinographies, although a linear relationship has been found between certain asymmetry features extracted from both imaging modalities. Therefore, the resulting performance of the models based on asymmetry features proves their ability to differentiate healthy from glaucoma patients using those metrics. Models trained from fundus characteristics are a useful option as a glaucoma screening method in the healthy population, although with lower performance than those trained from the thickness of the peripapillary retinal nerve fiber layer. In both imaging modalities, the asymmetry of morphological characteristics can be used as a glaucoma indicator, as detailed in this work. [ABSTRACT FROM AUTHOR]

In this paper, an approach that can fast classify the data from the electronic nose is presented. In this approach the gradient tree boosting algorithm is used to classify the gas data and the experiment results show that the proposed gradient tree boosting algorithm achieved high performance on this classification problem, outperforming other algorithms as comparison. In addition, electronic nose we used only requires a few seconds of data after the gas reaction begins. Therefore, the proposed approach can realize a fast recognition of gas, as it does not need to wait for the gas reaction to reach steady state. [ABSTRACT FROM AUTHOR]

The increasing geriatric population across the world has necessitated the early detection of frailty through the analysis of daily-life behavioral patterns. This paper presents a system for ambient, automatic, and the continuous measurement and analysis of ascent and descent motions and long-term handrail-use behaviors of participants in their homes using an RGB-D camera. The system automatically stores information regarding the environment and three-dimensional skeletal coordinates of the participant only when they appear within the camera's angle of view. Daily stair ascent and descent motions were measured in two houses: one house with two participants in their 20s and two in their 50s, and another with two participants in their 70s. The recorded behaviors were analyzed in terms of the stair ascent/descent speed, handrail grasping points, and frequency determined using the decision tree algorithm. The participants in their 70s exhibited a decreased stair ascent/descent speed compared to other participants; those in their 50s and 70s exhibited increased handrail usage area and frequency. The outcomes of the study indicate the system's ability to accurately detect a decline in physical function through the continuous measurement of daily stair ascent and descent motions. [ABSTRACT FROM AUTHOR]

The article presents the implementation of artificial intelligence algorithms for the problem of discretization in Electrical Impedance Tomography (EIT) adapted for urinary tract monitoring. The primary objective of discretization is to create a finite element mesh (FEM) classifier that will separate the inclusion elements from the background. In general, the classifier is designed to detect the area of elements belonging to an inclusion revealing the shape of that object. We show the adaptation of supervised learning methods such as logistic regression, decision trees, linear and quadratic discriminant analysis to the problem of tracking the urinary bladder using EIT. Our study focuses on developing and comparing various algorithms for discretization, which perfectly supplement methods for an inverse problem. The innovation of the presented solutions lies in the originally adapted algorithms for EIT allowing for the tracking of the bladder. We claim that a robust measurement solution with sensors and statistical methods can track the placement and shape change of the bladder, leading to effective information about the studied object. This article also shows the developed device, its functions and working principle. The development of such a device and accompanying information technology came about in response to particularly strong market demand for modern technical solutions for urinary tract rehabilitation. [ABSTRACT FROM AUTHOR]

In wells that operate by electrical submersible pump (ESP), the use of automation tools becomes essential in the interpretation of data. However, the fact that the wells work with automated systems does not guarantee the early diagnosis of operating conditions. The analysis of amperimetric charts is one of the ways to identify fail conditions. Generally, the analysis of these histographics is performed by operators who are often overloaded, generating a decrease in the efficiency of observing the well operating conditions. Currently, technologies based on machine learning (ML) algorithms create solutions to early diagnose abnormalities in the well's operation. Thus, this work aims to provide a proposal for detecting the operating conditions of the ESP pump from electrical current data from 24 wells in the city of Mossoró, Rio Grande do Norte state, Brazil. The algorithms used were Decision Tree, Support Vector Machine, K-Nearest Neighbor and Neural Network. The algorithms were tested without and with hyperparameter tuning based on a training dataset. The results confirm that the application of the ML algorithm is feasible for classifying the operating conditions of the ESP pump, as all had an accuracy greater than 87%, with the best result being the application of the SVM model, which reached an accuracy of 93%. [ABSTRACT FROM AUTHOR]

The article presents an attempt to identify an appropriate regression model for the estimation of cutting tool lifespan in the milling process based on the analysis of the R2 parameters of these models. The work is based on our own experiments and the accumulated database (which we make available for further use). The study uses a Haas VF-1 milling machine equipped with vibration sensors and based on a Beckhoff PLC data collector. As the acquired sensor data are continuous, and in order to account for dependencies between them, regression models were used. Support Vector Regression (SVR), decision trees and neural networks were tested during the work. The results obtained show that the best prediction results with the lowest error values were obtained for two-dimensional neural networks using the LBFGS solver (93.9%). Very similar results were also obtained for SVR (93.4%). The research carried out is related to the realisation of intelligent manufacturing dedicated to Industry 4.0 in the field of monitoring production processes, planning service downtime and reducing the level of losses resulting from damage to materials, semi-finished products and tools. [ABSTRACT FROM AUTHOR]

The event sensor provides high temporal resolution and generates large amounts of raw event data. Efficient low-complexity coding solutions are required for integration into low-power event-processing chips with limited memory. In this paper, a novel lossless compression method is proposed for encoding the event data represented as asynchronous event sequences. The proposed method employs only low-complexity coding techniques so that it is suitable for hardware implementation into low-power event-processing chips. A first, novel, contribution consists of a low-complexity coding scheme which uses a decision tree to reduce the representation range of the residual error. The decision tree is formed by using a triplet threshold parameter which divides the input data range into several coding ranges arranged at concentric distances from an initial prediction, so that the residual error of the true value information is represented by using a reduced number of bits. Another novel contribution consists of an improved representation, which divides the input sequence into same-timestamp subsequences, wherein each subsequence collects the same timestamp events in ascending order of the largest dimension of the event spatial information. The proposed same-timestamp representation replaces the event timestamp information with the same-timestamp subsequence length and encodes it together with the event spatial and polarity information into a different bitstream. Another novel contribution is the random access to any time window by using additional header information. The experimental evaluation on a highly variable event density dataset demonstrates that the proposed low-complexity lossless coding method provides an average improvement of 5.49 % , 11.45 % , and 35.57 % compared with the state-of-the-art performance-oriented lossless data compression codecs Bzip2, LZMA, and ZLIB, respectively. To our knowledge, the paper proposes the first low-complexity lossless compression method for encoding asynchronous event sequences that are suitable for hardware implementation into low-power chips. [ABSTRACT FROM AUTHOR]

Despite the great technological advances in ADAS, autonomous driving still faces many challenges. Among them is improving decision-making algorithms so that vehicles can make the right decision inspired by human driving. Not only must these decisions ensure the safety of the car occupants and the other road users, but they have to be understandable by them. This article focuses on decision-making algorithms for autonomous vehicles, specifically for lane changing on highways and sub-urban roads. The challenge to overcome is to develop a decision-making algorithm that combines fidelity to human behavior and that is based on machine learning, with a global structure that allows understanding the behavior of the algorithm and that is not opaque such as black box algorithms. To this end, a three-step decision-making method was developed: trajectory prediction of the surrounding vehicles, risk and gain computation associated with the maneuver and based on the predicted trajectories, and finally decision making. For the decision making, three algorithms: decision tree, random forest, and artificial neural network are proposed and compared based on a naturalistic driving database and a driving simulator. [ABSTRACT FROM AUTHOR]

Physical exercise has become an essential tool for treating various non-communicable diseases (also known as chronic diseases). Due to this, physical exercise allows to counter different symptoms and reduce some risk of death factors without medication. A solution to support people in doing exercises is to use artificial systems that monitor their exercise progress. While one crucial aspect is to monitor the correct physical motions for rehabilitative exercise, another essential element is to give encouraging feedback during workouts. A coaching system can track a user's exhaustion and give motivating feedback accordingly to boost exercise adherence. For this purpose, this research investigates whether it is possible to predict the subjective exhaustion level based on non-invasive and non-wearable technology. A novel data set was recorded with the facial record as the primary predictor and individual exhaustion levels as the predicted variable. 60 participants (30 male, 30 female) took part in the data recording. 17 facial action units (AU) were extracted as predictor variables for the perceived subjective exhaustion measured using the BORG scale. Using the predictor and the target variables, several regression and classification methods were evaluated aiming to predict exhaustion. The results showed that the decision tree and support vector methods provide reasonable prediction results. The limitation of the results, depending on participants being in the training data set and subjective variables (e.g., participants smiling during the exercises) were further discussed. [ABSTRACT FROM AUTHOR]

In this paper, we propose an object-cooperated decision method for efficient ternary tree (TT) partitioning that reduces the encoding complexity of versatile video coding (VVC). In most previous studies, the VVC complexity was reduced using decision schemes based on the encoding context, which do not apply object detecion models. We assume that high-level objects are important for deciding whether complex TT partitioning is required because they can provide hints on the characteristics of a video. Herein, we apply an object detection model that discovers and extracts the high-level object features—the number and ratio of objects from frames in a video sequence. Using the extracted features, we propose machine learning (ML)-based classifiers for each TT-split direction to efficiently reduce the encoding complexity of VVC and decide whether the TT-split process can be skipped in the vertical or horizontal direction. The TT-split decision of classifiers is formulated as a binary classification problem. Experimental results show that the proposed method more effectively decreases the encoding complexity of VVC than a state-of-the-art model based on ML. [ABSTRACT FROM AUTHOR]

Due to the recent advance in the industrial Internet of Things (IoT) in manufacturing, the vast amount of data from sensors has triggered the need for leveraging such big data for fault detection. In particular, interpretable machine learning techniques, such as tree-based algorithms, have drawn attention to the need to implement reliable manufacturing systems, and identify the root causes of faults. However, despite the high interpretability of decision trees, tree-based models make a trade-off between accuracy and interpretability. In order to improve the tree's performance while maintaining its interpretability, an evolutionary algorithm for discretization of multiple attributes, called Decision tree Improved by Multiple sPLits with Evolutionary algorithm for Discretization (DIMPLED), is proposed. The experimental results with two real-world datasets from sensors showed that the decision tree improved by DIMPLED outperformed the performances of single-decision-tree models (C4.5 and CART) that are widely used in practice, and it proved competitive compared to the ensemble methods, which have multiple decision trees. Even though the ensemble methods could produce slightly better performances, the proposed DIMPLED has a more interpretable structure, while maintaining an appropriate performance level. [ABSTRACT FROM AUTHOR]

According to the World Health Organisation, falling is a major health problem with potentially fatal implications. Each year, thousands of people die as a result of falls, with seniors making up 80% of these fatalities. The automatic detection of falls may reduce the severity of the consequences. Our study focuses on developing a vision-based fall detection system. Our work proposes a new feature descriptor that results in a new fall detection framework. The body geometry of the subject is analyzed and patterns that help to distinguish falls from non-fall activities are identified in our proposed method. An AlphaPose network is employed to identify 17 keypoints on the human skeleton. Thirteen keypoints are used in our study, and we compute two additional keypoints. These 15 keypoints are divided into five segments, each of which consists of a group of three non-collinear points. These five segments represent the left hand, right hand, left leg, right leg and craniocaudal section. A novel feature descriptor is generated by extracting the distances from the segmented parts, angles within the segmented parts and the angle of inclination for every segmented part. As a result, we may extract three features from each segment, giving us 15 features per frame that preserve spatial information. To capture temporal dynamics, the extracted spatial features are arranged in the temporal sequence. As a result, the feature descriptor in the proposed approach preserves the spatio-temporal dynamics. Thus, a feature descriptor of size [ m × 15 ] is formed where m is the number of frames. To recognize fall patterns, machine learning approaches such as decision trees, random forests, and gradient boost are applied to the feature descriptor. Our system was evaluated on the UPfall dataset, which is a benchmark dataset. It has shown very good performance compared to the state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

This paper presents a method for the multi-criteria classification of data in terms of identifying pneumatic wheel imbalance on the basis of vehicle body vibrations in normal operation conditions. The paper uses an expert system based on search graphs that apply source features of objects and distances from points in the space of classified objects (the metric used). Rules generated for data obtained from tests performed under stationary and road conditions using a chassis dynamometer were used to develop the expert system. The recorded linear acceleration signals of the vehicle body were analyzed in the frequency domain for which the power spectral density was determined. The power field values for selected harmonics of the spectrum consistent with the angular velocity of the wheel were adopted for further analysis. In the developed expert system, the Kamada–Kawai model was used to arrange the nodes of the decision tree graph. Based on the developed database containing learning and testing data for each vehicle speed and wheel balance condition, the probability of the wheel imbalance condition was determined. As a result of the analysis, it was determined that the highest probability of identifying wheel imbalance equal to almost 100% was obtained in the vehicle speed range of 50 km/h to 70 km/h. This is known as the pre-resonance range in relation to the eigenfrequency of the wheel vibrations. As the vehicle speed increases, the accuracy of the data classification for identifying wheel imbalance in relation to the learning data decreases to 50% for the speed of 90 km/h. [ABSTRACT FROM AUTHOR]

Intrusion detection system (IDS) design for mobile adhoc networks (MANET) is a crucial component for maintaining the integrity of the network. The need for rapid deployment of IDS capability with minimal data availability for training and testing is an important requirement of such systems, especially for MANETs deployed in highly dynamic scenarios, such as battlefields. This work proposes a two-level detection scheme for detecting malicious nodes in MANETs. The first level deploys dedicated sniffers working in promiscuous mode. Each sniffer utilizes a decision-tree-based classifier that generates quantities which we refer to as correctly classified instances (CCIs) every reporting time. In the second level, the CCIs are sent to an algorithmically run supernode that calculates quantities, which we refer to as the accumulated measure of fluctuation (AMoF) of the received CCIs for each node under test (NUT). A key concept that is used in this work is that the variability of the smaller size population which represents the number of malicious nodes in the network is greater than the variance of the larger size population which represents the number of normal nodes in the network. A linear regression process is then performed in parallel with the calculation of the AMoF for fitting purposes and to set a proper threshold based on the slope of the fitted lines. As a result, the malicious nodes are efficiently and effectively separated from the normal nodes. The proposed scheme is tested for various node velocities and power levels and shows promising detection performance even at low-power levels. The results presented also apply to wireless sensor networks (WSN) and represent a novel IDS scheme for such networks. [ABSTRACT FROM AUTHOR]

Application of deep learning (DL) to the field of healthcare is aiding clinicians to make an accurate diagnosis. DL provides reliable results for image processing and sensor interpretation problems most of the time. However, model uncertainty should also be thoroughly quantified. This paper therefore addresses the employment of Monte Carlo dropout within the DL structure to automatically discriminate presymptomatic signs of spinocerebellar ataxia type 2 in saccadic samples obtained from electrooculograms. The current work goes beyond the common incorporation of this special type of dropout into deep neural networks and uses the uncertainty derived from the validation samples to construct a decision tree at the register level of the patients. The decision tree built from the uncertainty estimates obtained a classification accuracy of 81.18% in automatically discriminating control, presymptomatic and sick classes. This paper proposes a novel method to address both uncertainty quantification and explainability to develop reliable healthcare support systems. [ABSTRACT FROM AUTHOR]

The Brillouin Optical Time-Domain Analyzer assisted by the AdaBoost Algorithm for Brillouin frequency shift (BFS) extraction is proposed and experimentally demonstrated. The Brillouin gain spectrum classification under different BFS is realized by iteratively updating the weak classifier in the form of a decision tree, forming several base classifiers and combining them into a strong classifier. Based on the pseudo-Voigt curve training set with noise, the performance of the AdaBoost Algorithm is studied, and the influence of different signal-to-noise ratio (SNR), frequency range, and frequency step is also studied. Results show that the performance of BFS extraction decreases with the decrease in SNR, the reduction in frequency range, and the increase in frequency step. [ABSTRACT FROM AUTHOR]

In recent years, many methods for intrusion detection systems (IDS) have been designed and developed in the research community, which have achieved a perfect detection rate using IDS datasets. Deep neural networks (DNNs) are representative examples applied widely in IDS. However, DNN models are becoming increasingly complex in model architectures with high resource computing in hardware requirements. In addition, it is difficult for humans to obtain explanations behind the decisions made by these DNN models using large IoT-based IDS datasets. Many proposed IDS methods have not been applied in practical deployments, because of the lack of explanation given to cybersecurity experts, to support them in terms of optimizing their decisions according to the judgments of the IDS models. This paper aims to enhance the attack detection performance of IDS with big IoT-based IDS datasets as well as provide explanations of machine learning (ML) model predictions. The proposed ML-based IDS method is based on the ensemble trees approach, including decision tree (DT) and random forest (RF) classifiers which do not require high computing resources for training models. In addition, two big datasets are used for the experimental evaluation of the proposed method, NF-BoT-IoT-v2, and NF-ToN-IoT-v2 (new versions of the original BoT-IoT and ToN-IoT datasets), through the feature set of the net flow meter. In addition, the IoTDS20 dataset is used for experiments. Furthermore, the SHapley additive exPlanations (SHAP) is applied to the eXplainable AI (XAI) methodology to explain and interpret the classification decisions of DT and RF models; this is not only effective in interpreting the final decision of the ensemble tree approach but also supports cybersecurity experts in quickly optimizing and evaluating the correctness of their judgments based on the explanations of the results. [ABSTRACT FROM AUTHOR]

The article presents an extensive analysis of the literature related to the diagnosis of the extrusion process and proposes a new, unique method. This method is based on the observation of the punch displacement signal in relation to the die, and then approximation of this signal using a polynomial. It is difficult to find in the literature even an attempt to solve the problem of diagnosing the extrusion process by means of a simple distance measurement. The dominant feature is the use of strain gauges, force sensors or even accelerometers. However, the authors managed to use the displacement signal, and it was considered a key element of the method presented in the article. The aim of the authors was to propose an effective method, simple to implement and not requiring high computing power, with the possibility of acting and making decisions in real time. At the input of the classifier, authors provided the determined polynomial coefficients and the SSE (Sum of Squared Errors) value. Based on the SSE values only, the decision tree algorithm performed anomaly detection with an accuracy of 98.36%. With regard to the duration of the experiment (single extrusion process), the decision was made after 0.44 s, which is on average 26.7% of the extrusion experiment duration. The article describes in detail the method and the results achieved. [ABSTRACT FROM AUTHOR]

In bearing diagnostics using a data-driven modeling approach, a concern is the need for data from all possible scenarios to build a practical model for all operating conditions. This paper is a study on bearing diagnostics with the concurrent occurrence of multiple defect types. The authors are not aware of any work in the literature that studies this practical problem. A strategy based on one-versus-all (OVA) class binarization is proposed to improve fault diagnostics accuracy while reducing the number of scenarios for data collection, by predicting concurrent defects from training data of normal and single defects. The proposed OVA diagnostic approach is evaluated with empirical analysis using support vector machine (SVM) and C4.5 decision tree, two popular classification algorithms frequently applied to system health diagnostics and prognostics. Statistical features are extracted from the time domain and the frequency domain. Prediction performance of the proposed strategy is compared with that of a simple multi-class classification, as well as that of random guess and worst-case classification. We have verified the potential of the proposed OVA diagnostic strategy in performance improvements for single-defect diagnosis and predictions of BPFO plus BPFI concurrent defects using two laboratory-collected vibration data sets. [ABSTRACT FROM AUTHOR]

Foot strike detection is important when evaluating a person's gait characteristics. Accelerometer and gyroscope signals from smartphones have been used to train artificial intelligence (AI) models for automated foot strike detection in able-bodied and elderly populations. However, there is limited research on foot strike detection in lower limb amputees, who have a more variable and asymmetric gait. A novel method for automated foot strike detection in lower limb amputees was developed using raw accelerometer and gyroscope signals collected from a smartphone positioned at the posterior pelvis. Raw signals were used to train a decision tree model and long short-term memory (LSTM) model for automated foot strike detection. These models were developed using retrospective data (n = 72) collected with the TOHRC Walk Test app during a 6-min walk test (6MWT). An Android smartphone was placed on a posterior belt for each participant during the 6MWT to collect accelerometer and gyroscope signals at 50 Hz. The best model for foot strike identification was the LSTM with 100 hidden nodes in the LSTM layer, 50 hidden nodes in the dense layer, and a batch size of 64 (99.0% accuracy, 86.4% sensitivity, 99.4% specificity, and 83.7% precision). This research created a novel method for automated foot strike identification in lower extremity amputee populations that is equivalent to manual labelling and accessible for clinical use. Automated foot strike detection is required for stride analysis and to enable other AI applications, such as fall detection. [ABSTRACT FROM AUTHOR]

R peak detection is crucial in electrocardiogram (ECG) signal analysis to detect and diagnose cardiovascular diseases (CVDs). Herein, the dynamic mode selected energy (DMSE) and adaptive window sizing (AWS) algorithm are proposed for detecting R peaks with better efficiency. The DMSE algorithm adaptively separates the QRS components and all non-objective components from the ECG signal. Based on local peaks in QRS components, the AWS algorithm adaptively determines the Region of Interest (ROI). The Feature Extraction process computes the statistical properties of energy, frequency, and noise from each ROI. The Sequential Forward Selection (SFS) procedure is used to find the best subsets of features. Based on these characteristics, an ensemble of decision tree algorithms detects the R peaks. Finally, the R peak position on the initial ECG signal is adjusted using the R location correction (RLC) algorithm. The proposed method has an experimental accuracy of 99.94%, a sensitivity of 99.98%, positive predictability of 99.96%, and a detection error rate of 0.06%. Given the high efficiency in detection and fast processing speed, the proposed approach is ideal for intelligent medical and wearable devices in the diagnosis of CVDs. [ABSTRACT FROM AUTHOR]

Oscillating Water Column (OWC) systems harness wave energy using a partially submerged chamber with an underwater opening. The Savonius turbine, a vertical-axis wind turbine, is well-suited for this purpose due to its efficiency at low speeds and self-starting capability, making it an ideal power take-off (PTO) mechanism in OWC systems. This study tested an OWC device with a Savonius turbine in an air duct to evaluate its performance under varying flow directions and loads. An innovative aspect was assessing the influence of power augmenters (PAs) positioned upstream and downstream of the turbine. The experimental setup included load cells, Pitot tubes, differential pressure sensors and rotational speed sensors. Data obtained were used to calculate pressure differentials across the turbine and torque. The primary goal of using PA is to increase the CP–λ curve area without modifying the turbine geometry, potentially enabling interventions on existing turbines without rotor dismantling. Additionally, another novelty is the implementation of a regression Machine-Learning algorithm based on decision trees to analyze the influence of various features on predicting pressure differences, thereby broadening the scope for further testing beyond physical experimentation. [ABSTRACT FROM AUTHOR]

The sensors used in the Internet of Medical Things (IoMT) network run on batteries and need to be replaced, replenished or should use energy harvesting for continuous power needs. Additionally, there are mechanisms for better utilization of battery power for network longevity. IoMT networks pose a unique challenge with respect to sensor power replenishment as the sensors could be embedded inside the subject. A possible solution could be to reduce the amount of sensor data transmission and recreate the signal at the receiving end. This article builds upon previous physiological monitoring studies by applying new decision tree-based regression models to calculate the accuracy of reproducing data from two sets of physiological signals transmitted over cellular networks. These regression analyses are then executed over three different iteration varieties to assess the effect that the number of decision trees has on the efficiency of the regression model in question. The results indicate much lower errors as compared to other approaches indicating significant saving on the battery power and improvement in network longevity. [ABSTRACT FROM AUTHOR]

The growing popularity of social media has engendered the social problem of spam proliferation through this medium. New spam types that evade existing spam detection systems are being developed continually, necessitating corresponding countermeasures. This study proposes an anomaly detection-based framework to detect new Twitter spam, which works by modeling the characteristics of non-spam tweets and using anomaly detection to classify tweets deviating from this model as anomalies. However, because modeling varied non-spam tweets is challenging, the technique's spam detection and false positive (FP) rates are low and high, respectively. To overcome this shortcoming, anomaly detection is performed on known spam tweets pre-detected using a trained decision tree while modeling normal tweets. A one-class support vector machine and an autoencoder with high detection rates are used for anomaly detection. The proposed framework exhibits superior detection rates for unknown spam compared to conventional techniques, while maintaining equivalent or improved detection and FP rates for known spam. Furthermore, the framework can be adapted to changes in spam conditions by adjusting the costs of detection errors. [ABSTRACT FROM AUTHOR]

Fall-related information can help clinical professionals make diagnoses and plan fall prevention strategies. The information includes various characteristics of different fall phases, such as falling time and landing responses. To provide the information of different phases, this pilot study proposes an automatic multiphase identification algorithm for phase-aware fall recording systems. Seven young adults are recruited to perform the fall experiment. One inertial sensor is worn on the waist to collect the data of body movement, and a total of 525 trials are collected. The proposed multiphase identification algorithm combines machine learning techniques and fragment modification algorithm to identify pre-fall, free-fall, impact, resting and recovery phases in a fall process. Five machine learning techniques, including support vector machine, k-nearest neighbor (kNN), naïve Bayesian, decision tree and adaptive boosting, are applied to identify five phases. Fragment modification algorithm uses the rules to detect the fragment whose results are different from the neighbors. The proposed multiphase identification algorithm using the kNN technique achieves the best performance in 82.17% sensitivity, 85.74% precision, 73.51% Jaccard coefficient, and 90.28% accuracy. The results show that the proposed algorithm has the potential to provide automatic fine-grained fall information for clinical measurement and assessment. [ABSTRACT FROM AUTHOR]

Changes of directions and cutting maneuvers, including 180-degree turns, are common locomotor actions in team sports, implying high mechanical load. While the mechanics and neurophysiology of turns have been extensively studied in laboratory conditions, modern inertial measurement units allow us to monitor athletes directly on the field. In this study, we applied four supervised machine learning techniques (linear regression, support vector regression/machine, boosted decision trees and artificial neural networks) to predict turn direction, speed (before/after turn) and the related positive/negative mechanical work. Reference values were computed using an optical motion capture system. We collected data from 13 elite female soccer players performing a shuttle run test, wearing a six-axes inertial sensor at the pelvis level. A set of 18 features (predictors) were obtained from accelerometers, gyroscopes and barometer readings. Turn direction classification returned good results (accuracy > 98.4%) with all methods. Support vector regression and neural networks obtained the best performance in the estimation of positive/negative mechanical work (coefficient of determination R2 = 0.42–0.43, mean absolute error = 1.14–1.41 J) and running speed before/after the turns (R2 = 0.66–0.69, mean absolute error = 0.15–018 m/s). Although models can be extended to different angles, we showed that meaningful information on turn kinematics and energetics can be obtained from inertial units with a data-driven approach. [ABSTRACT FROM AUTHOR]

In this paper, a novel approach is presented to the measurement of marine icing phenomena under the presence of a two-phase condition. We have developed a sensor consisting of an electrostatic array and a signal processing based on a decision tree method. A three-element electrostatic array is employed to derive signals having linearly decoupled characteristics from which two key parameters, ice and water accretion layer dimension, can be determined for the purpose of environmental monitoring. The quantified characteristics revealed a correlation with the ice layer thickness in spite of the strong influence from the top water phase layer. The decision tree model established a relationship between the signal characteristics and the two accretion thickness parameters of water and ice layer. Through experimental verification, it has been observed that our sensor array in combination with the decision tree model based signal processing provides a simple practical solution to the challenging field of a two phase composition measurement such as in the marine icing considered in this study. [ABSTRACT FROM AUTHOR]

Smart wearable robotic system, such as exoskeleton assist device and powered lower limb prostheses can rapidly and accurately realize man–machine interaction through locomotion mode recognition system. However, previous locomotion mode recognition studies usually adopted more sensors for higher accuracy and effective intelligent algorithms to recognize multiple locomotion modes simultaneously. To reduce the burden of sensors on users and recognize more locomotion modes, we design a novel decision tree structure (DTS) based on using an improved backpropagation neural network (IBPNN) as judgment nodes named IBPNN-DTS, after analyzing the experimental locomotion mode data using the original values with a 200-ms time window for a single inertial measurement unit to hierarchically identify nine common locomotion modes (level walking at three kinds of speeds, ramp ascent/descent, stair ascent/descent, Sit, and Stand). In addition, we reduce the number of parameters in the IBPNN for structure optimization and adopted the artificial bee colony (ABC) algorithm to perform global search for initial weight and threshold value to eliminate system uncertainty because randomly generated initial values tend to result in a failure to converge or falling into local optima. Experimental results demonstrate that recognition accuracy of the IBPNN-DTS with ABC optimization (ABC-IBPNN-DTS) was up to 96.71% (97.29% for the IBPNN-DTS). Compared to IBPNN-DTS without optimization, the number of parameters in ABC-IBPNN-DTS shrank by 66% with only a 0.58% reduction in accuracy while the classification model kept high robustness. [ABSTRACT FROM AUTHOR]

Falls are seriously threatening the health of elderly. In order to reduce the potential danger caused by falls, this paper proposes a two-stage fall recognition algorithm based on human posture features. For preprocessing, we construct the new key features: deflection angles and spine ratio to describe the changes of human posture based on the human skeleton extracted by OpenPose. In the first stage, based on the variables: tendency symbol and steady symbol integrated by the scattered key features, we divide the human body state into three states: stable state, fluctuating state, and disordered state. By analyzing whether the body is in a stable state, the ADL (activities of daily living) actions with high stability can be preliminarily excluded. In the second stage: to further identify the confusing ADL actions and the fall actions, we innovatively design a time-continuous recognition algorithm. When human body is constantly in an unstable state, the human posture features: compare value γ , energy value ε , state score τ are proposed to form a feature vector, and support vector machine (SVM), K nearest neighbors (KNN), decision tree (DT), random forest (RF) are utilized for classification. Experiment results demonstrate that SVM with linear kernel function can distinguish falling actions best and our approach achieved a detection accuracy of 97.34%, precision of 98.50%, and the recall, F1 score are 97.33%, 97.91% respectively. Compared with previous state-of-art algorithms, our algorithm can achieve the highest recognition accuracy. It proves that our fall detection method is effective. [ABSTRACT FROM AUTHOR]

Monitoring the condition of rotating machinery, especially planetary gearboxes, is a challenging problem. In most of the available approaches, diagnostic procedures are related to advanced signal pre-processing/feature extraction methods or advanced data (features) analysis by using artificial intelligence. In this paper, the second approach is explored, so an application of decision trees for the classification of spectral-based 15D vectors of diagnostic data is proposed. The novelty of this paper is that by a combination of spectral analysis and the application of decision trees to a set of spectral features, we are able to take advantage of the multidimensionality of diagnostic data and classify/recognize the gearbox condition almost faultlessly even in non-stationary operating conditions. The diagnostics of time-varying systems are a complicated issue due to time-varying probability densities estimated for features. Using multidimensional data instead of an aggregated 1D feature, it is possible to improve the efficiency of diagnostics. It can be underlined that in comparison to previous work related to the same data, where the aggregated 1D variable was used, the efficiency of the proposed approach is around 99% (ca. 19% better). We tested several algorithms: classification and regression trees with the Gini index and entropy, as well as the random tree. We compare the obtained results with the K-nearest neighbors classification algorithm and meta-classifiers, namely: random forest and AdaBoost. As a result, we created the decision tree model with 99.74% classification accuracy on the test dataset. [ABSTRACT FROM AUTHOR]

School bullying is a serious problem among teenagers. School violence is one type of school bullying and considered to be the most harmful. As AI (Artificial Intelligence) techniques develop, there are now new methods to detect school violence. This paper proposes a video-based school violence detecting algorithm. This algorithm first detects foreground moving targets via the KNN (K-Nearest Neighbor) method and then preprocesses the detected targets via morphological processing methods. Then, this paper proposes a circumscribed rectangular frame integrating method to optimize the circumscribed rectangular frame of moving targets. Rectangular frame features and optical-flow features were extracted to describe the differences between school violence and daily-life activities. We used the Relief-F and Wrapper algorithms to reduce the feature dimension. SVM (Support Vector Machine) was applied as the classifier, and 5-fold cross validation was performed. The accuracy was 89.6%, and the precision was 94.4%. To further improve the recognition performance, we developed a DT–SVM (Decision Tree–SVM) two-layer classifier. We used boxplots to determine some features of the DT layer that are able to distinguish between typical physical violence and daily-life activities and between typical daily-life activities and physical violence. For the remainder of activities, the SVM layer performed a classification. For this DT–SVM classifier, the accuracy reached 97.6%, and the precision reached 97.2%, thus showing a significant improvement. [ABSTRACT FROM AUTHOR]