Search

Among patients with epilepsy, 30–40% experience recurrent seizures even after adequate antiseizure medications therapies, making them refractory. The early identification of refractory epilepsy is important to provide timely surgical treatment for these patients. In this study, we analyze interictal electroencephalography (EEG) data to predict drug refractoriness in patients with temporal lobe epilepsy (TLE) who were treated with monotherapy at the time of the first EEG acquisition. Various EEG features were extracted, including statistical measurements and interchannel coherence. Feature selection was performed to identify the optimal features, and classification was conducted using different classifiers. Functional connectivity and graph theory measurements were calculated to identify characteristics of refractory TLE. Among the 48 participants, 34 (70.8%) were responsive, while 14 (29.2%) were refractory over a mean follow-up duration of 38.5 months. Coherence feature within the gamma frequency band exhibited the most favorable performance. The light gradient boosting model, employing the mutual information filter-based feature selection method, demonstrated the highest performance (AUROC = 0.821). Compared to the responsive group, interchannel coherence displayed higher values in the refractory group. Interestingly, graph theory measurements using EEG coherence exhibited higher values in the refractory group than in the responsive group. Our study has demonstrated a promising method for the early identification of refractory TLE utilizing machine learning algorithms. [ABSTRACT FROM AUTHOR]

Malaria is a deadly disease that can take a person's life if not predicted or cured correctly. Numerous factors like temperature, humidity, precipitation, etc., impact India's increasing cases of malaria diseases. This research presents an advanced machine learning regression technique recently developed to anticipate the prevalence of malaria in Tripura using a real-world data. The proposed structure uses nine different regression methods, such as multilayer perceptron (MLP), random forest, support vector regressor, gradient boosting regressor, Bayesian ridge, kernel ridge, extreme gradient boost regressor (XGB), light gradient boosting machine (LGBM regressor and linear regression, to predict malaria using the most affecting factors of malaria diseases, namely temperature, humidity, precipitation, month, and years as input. Furthermore, to opt out the best suited technology for malaria cases prediction, the range of value method (ROVM)–multi-criteria decision methods (MCDM) technique has been applied, considering various statistical measurements as criteria. Ultimately, a comparison of various MCDM techniques reveals that MLP, XGB, and RF emerge as the top three choices. MLP regression, with root-mean-square error (RMSE) value of 0.03357, yields the lowest RMSE value, and the coefficient of determination (R2) is 0.97616, yielding the maximum among other regressions. To effectively battle the illness in Tripura, it could be useful for continuing intervention tactics by governmental, profit and nonprofit organizations. [ABSTRACT FROM AUTHOR]

In this study, a deep learning algorithm namely long short-term memory (LSTM) has been developed for forecasting daily pan evaporation at Sydney airport, Australia. The accuracy of the developed LSTM model has been compared with a commonly used machine learning model, namely multilayer perceptron neural network (MLP-NN). The evaporation rate as a single parameter was used with one time-lag based on autocorrelation function (ACF). The utilized data duration was from January 2021 to February 2022 (during Covid-19 pandemic). Different statistical measurements have been applied in order to evaluate the performance of the proposed models. The results showed that the developed LSTM model outperformed MLP-NN. The LSTM performed well with RMSE = 1.074, MAE = 0.771, R2 = 0.97, while the MLP-NN had least performance with RMSE = 2.801, MAE = 1.994, and R2 = 0.57. The developed LSTM model could be utilized in other locations for forecasting daily pan evaporation. [ABSTRACT FROM AUTHOR]

Congestive heart failure (CHF) is one of the primary sources of mortality and morbidity among the global population. Over 26 million individuals globally are affected by heart disease, and its prevalence is rising by 2% yearly. With advances in healthcare technologies, if we predict CHF in the early stages, one of the leading global mortality factors can be reduced. Therefore, the main objective of this study is to use machine learning applications to enhance the diagnosis of CHF and to reduce the cost of diagnosis by employing minimum features to forecast the possibility of a CHF occurring. We employ a deep neural network (DNN) classifier for CHF classification and compare the performance of DNN with various machine learning classifiers. In this research, we use a very challenging dataset, called the Cardiovascular Health Study (CHS) dataset, and a unique pre-processing technique by integrating C4.5 and K-nearest neighbor (KNN). While the C4.5 technique is used to find significant features and remove the outlier data from the dataset, the KNN algorithm is employed for missing data imputation. For classification, we compare six state-of-the-art machine learning (ML) algorithms (KNN, logistic regression (LR), naive Bayes (NB), random forest (RF), support vector machine (SVM), and decision tree (DT)) with DNN. To evaluate the performance, we use seven statistical measurements (i.e., accuracy, specificity, sensitivity, F1-score, precision, Matthew's correlation coefficient, and false positive rate). Overall, our results reflect our proposed integrated approach, which outperformed other machine learning algorithms in terms of CHF prediction, reducing patient expenses by reducing the number of medical tests. The proposed model obtained 97.03% F1-score, 95.30% accuracy, 96.49% sensitivity, and 97.58% precision. [ABSTRACT FROM AUTHOR]

Key leaf functional traits, such as chlorophyll and carotenoids content (Cab and Cxc), equivalent water thickness (EWT), and leaf mass per area (LMA), are essential to the characterization and monitoring of ecosystem function. Spectroscopy provides access to these four leaf traits by relying on their specific spectral absorptions over the 0.4–2.5 µm domain. In this study, we compare the performance of three categories of estimation methods to retrieve these four leaf traits from laboratory directional-hemispherical leaf reflectance and transmittance measurements: statistical, physical, and hybrid methods. To this aim, a dataset pooling samples from 114 deciduous and evergreen oak trees was collected on four sites in California (woodland savannas and mixed forests) over three seasons (spring, summer and fall) and was used to assess the performance of each method. Physical and hybrid methods were based on the PROSPECT leaf radiative transfer model. Physical methods included inversion of PROSPECT from iterative algorithms and look-up table (LUT)-based inversion. For LUT-based methods, two distance functions and two sampling schemes were tested. For statistical and hybrid methods, four distinct machine learning regression algorithms were compared: ridge, partial least squares regression (PLSR), Gaussian process regression (GPR), and random forest regression (RFR). In addition, we evaluated the transferability of statistical methods using an independent dataset (ANGERS Leaf optical properties database) to train the regression algorithms. Thus, a total of 17 estimations were compared. Firstly, we studied the PROSPECT leaf structural parameter N retrieved by iterative inversions and its distribution over our oak-specific dataset. N showed a more pronounced seasonal dependency for the deciduous species than for the evergreen species. For the four traits, the statistical methods trained on our dataset outperformed the PROSPECT-based methods. More particularly, statistical methods using GPR yielded the most accurate estimates (RMSE = 5.0 µg·cm−2; 1.3 µg·cm−2; 0.0009 cm; and 0.0009 g·cm−2 for Cab, Cxc, EWT, and LMA, respectively). Among the PROSPECT-based methods, the iterative inversion of this model led to the most accurate results for Cab, Cxc, and EWT (RMSE = 7.8 µg·cm−2; 2.0 µg·cm−2; and 0.0035 cm, respectively), while for LMA, a hybrid method with RFR (RMSE = 0.0030 g·cm−2) was the most accurate. These results showed that estimation accuracy is independent of the season. Considering the transferability of statistical methods, for the four leaf traits, estimation performance was inferior for estimators built on the ANGERS database compared to estimators built exclusively on our dataset. However, for EWT and LMA, we demonstrated that these types of statistical methods lead to better estimation accuracy than PROSPECT-based methods (RMSE = 0.0016 cm and 0.0013 g·cm−2 respectively). Finally, our results showed that more differences were observed between plant functional types than between species or seasons. [ABSTRACT FROM AUTHOR]

In various engineering applications, remote sensing images such as digital elevation models (DEMs) and orthomosaics provide a convenient means of generating 3D representations of physical assets, enabling the discovery of new insights and analyses. However, the presence of noise and artefacts, particularly unwanted natural features, poses significant challenges, and their removal requires the application of filtering techniques prior to conducting analysis. Unmanned aerial vehicle-based photogrammetry is used at Melbourne Water's Western Treatment Plant as a cost-effective and efficient method of inspecting the floating covers on the anaerobic lagoons. The focus of interest is the elevation profile of the floating covers for these sewage-processing lagoons and its implications for sub-surface scum accumulation, which can compromise the structural integrity of the engineered assets. However, unwanted artefacts due to trapped rainwater, debris, dirt, and other irrelevant structures can significantly distort the elevation profile. In this study, a machine learning algorithm is utilised to group distinct features on the floating cover based on an image segmentation process. An unsupervised k-means clustering algorithm is employed, which operates on a stacked 4D array composed of the elevation of the DEM and the RGB channels of the associated orthomosaic. In the cluster validation process, seven cluster groups were considered optimal based on the Calinski–Harabasz criterion. Furthermore, by utilising the k-means method as a filtering technique, three clusters contain features related to the elevations associated with the floating cover membrane, collectively representing 84% of the asset, with each cluster contributing at least 19% of the asset. The artefact groups constitute less than 6% of the asset and exhibit significantly different features, colour characteristics, and statistical measurements from those of the membrane groups. The study found notable improvements using the k-means filtering method, including a 59.4% average reduction in outliers and a 36.3% decrease in standard deviation compared to raw data. Additionally, employing the proposed method in the scum hardness analysis improved correlation strength by 13.1%, removing approximately 16% of the artefacts in total assets, in contrast to a 3.6% improvement with the median filtering method. This improved imaging will lead to significant benefits when integrating imagery into deep learning models for structural health monitoring and asset performance. [ABSTRACT FROM AUTHOR]

Currently, a substantial portion of images snapped at exhibitions and galleries on social media demonstrates that aesthetic experience is not restricted to the confines of cultural institutions. The primary objective of this paper is to examine whether the content or aspect of an artwork influences the aesthetic experience of the viewer and to measure the artwork's social media popularity. To compare controversial works of art with those whose design, qualities, or intended message are non-controversial, we first sought out controversial works. A variety of artworks were revealed on Instagram; thus, the objective was to identify a non-controversial artwork published in the same year as each controversial artwork. We adhered to the complete procedure for cleansing, standardizing, and transforming the data to ensure comparability. Popularity was measured using a ranking algorithm and quantitative approaches for the recognition and statistical measurement of emotions. In addition, the exhaustive literature survey on models of aesthetic experience revealed no link between the experience of art and its social media popularity. Considering this, we have proposed, among other things, a new framework for interacting with art that integrates these parameters. According to the findings, controversial artworks elicited stronger emotions than non-controversial artworks. Furthermore, investigations have determined the three most popular works of art in each category. Under the scrutiny of social media, these results may inspire future research on the popularity of museum artworks and the design of aesthetic experiences. [ABSTRACT FROM AUTHOR]

As a result of global climate change, sustainable water supply management is becoming increasingly difficult. Dams and reservoirs are key tools for controlling and managing water resources; they have benefited human cultures in a variety of ways, including enhanced human health, increased food production, water supply for domestic and industrial use, economic growth, irrigation, hydro-power generation, and flood control. This study aims to compare the application of deep learning and conventional machine learning algorithms for predicting daily reservoir inflow. Long short-term memory (LSTM) has been applied as a deep learning algorithm and boosted regression tree (BRT) has been implemented as a machine learning algorithm. Five statistical indices have been selected to evaluate the performance of the proposed models. The selected statistical measurements are mean absolute error (MAE), root mean square error (RMSE), correlation coefficient (R), coefficient of determination (R2), mean square error (MSE), Nash Sutcliffe Model Efficiency Coefficient (NSE), and the RMSE-observations standard deviation ratio (RSR). The findings showed that LSTM outperformed BRT with a significant difference in terms of accuracy. [ABSTRACT FROM AUTHOR]

Yarn quality prediction is vital in cotton spinning mills for stable production, quality assurance, and cost control. This paper built a machine-learning-based yarn quality prediction method, which contains two core parts: a novel data characterization measurement and a one-dimensional convolutional neural network (YQP-1D-CNN). Firstly, apply a statistical measurement to extract the variable-size feature indicators from the spinning technical parameters, including raw cotton blending scheme, yarn structure parameters, and production process parameters. Then a 1D-CNN prediction model is established based on the prior structural relationships between the feature indicators. Compared with the traditional ANNs, YQP-1D-CNN possesses fewer parameters, requires fewer training samples, and represents better generalization performance. In the experiment, 200 samples are collected from a spinning mill. The prediction performance of the proposed method is compared with six models in existing research. The YQP-1D-CNN shows the highest prediction R2 and the lowest MSE on the overall performance. The better performance of YQP-1D-CNN on poor sample quality is verified better as well. The ablation experiment proves the advantages of the data characterization measurement and the 1D-CNN structure on yarn quality prediction. Finally, the YQP-1D-CNN has been tested in the spinning mill and met practical production requirements. [ABSTRACT FROM AUTHOR]

Topic modeling is a machine learning algorithm based on statistics that follows unsupervised machine learning techniques for mapping a high-dimensional corpus to a low-dimensional topical subspace, but it could be better. A topic model's topic is expected to be interpretable as a concept, i.e., correspond to human understanding of a topic occurring in texts. While discovering corpus themes, inference constantly uses vocabulary that impacts topic quality due to its size. Inflectional forms are in the corpus. Since words frequently appear in the same sentence and are likely to have a latent topic, practically all topic models rely on co-occurrence signals between various terms in the corpus. The topics get weaker because of the abundance of distinct tokens in languages with extensive inflectional morphology. Lemmatization is often used to preempt this problem. Gujarati is one of the morphologically rich languages, as a word may have several inflectional forms. This paper proposes a deterministic finite automaton (DFA) based lemmatization technique for the Gujarati language to transform lemmas into their root words. The set of topics is then inferred from this lemmatized corpus of Gujarati text. We employ statistical divergence measurements to identify semantically less coherent (overly general) topics. The result shows that the lemmatized Gujarati corpus learns more interpretable and meaningful subjects than unlemmatized text. Finally, results show that lemmatization curtails the size of vocabulary decreases by 16% and the semantic coherence for all three measurements—Log Conditional Probability, Pointwise Mutual Information, and Normalized Pointwise Mutual Information—from −9.39 to −7.49, −6.79 to −5.18, and −0.23 to −0.17, respectively. [ABSTRACT FROM AUTHOR]

One of the core elements of Machine Learning (ML) is statistics and its embedded foundational rules and without its appropriate integration, ML as we know would not exist. Various aspects of ML platforms are based on statistical rules and most notably the end results of the ML model performance cannot be objectively assessed without appropriate statistical measurements. The scope of statistics within the ML realm is rather broad and cannot be adequately covered in a single review article. Therefore, here we will mainly focus on the common statistical concepts that pertain to supervised ML (i.e. classification and regression) along with their interdependencies and certain limitations. [ABSTRACT FROM AUTHOR]

Predicting the tire-road contact triaxial stress is significant in assessing performance of vehicle and road surface material. However, both the Finite Element Method and the direct measurement method can only obtain contact stresses under several specific conditions, which is difficult to generalize. In this paper, a chain regression model is proposed, which hybridizes εSVR and ANN to improve forecasting accuracy in all directions by predicting uniaxial and triaxial stresses step by step. Meanwhile, the specific type of tire (185/65R15) under different conditions are simulated by using the 3 D finite element method for the dataset, and the factors affecting triaxial stress are analyzed by correlation analysis. Numerical examples from the above dataset reveal that the proposed εSVR-ANN chain model outperforms other multi-output regression models in all five directions in terms of forecasting accuracy. In addition, statistical tests verify the efficacy of the proposed method. This study provides a reference for the design of tire-road contact stress measurement and statistical scheme. [ABSTRACT FROM AUTHOR]

Manufacturability analysis is a critical step before manufacturing to reduce costs and risks. It is used widely in conventional manufacturing (CM) processes. However, to the best of our knowledge, there is no natural method to evaluate the manufacturability of additive manufacturing (AM) processes that have more uncertainty-derived risks and costs than CM processes. A clear definition of the manufacturability of AM processes has not been established, and there is no standard to check whether a component is manufactured successfully by an AM process, particularly for porous complex components. This study introduces the development of a new machine learning-based method to solve the problem mentioned above. It is based on the statistical measurement of experimental samples. The proposed method can be used to perform the manufacturability analysis for periodic cellular structures printed by a selective laser melting (SLM) process. A novel definition of the manufacturability of the SLM-ed periodic cellular structure was proposed. Experimental results indicate that the developed learning model (ANN model) can achieve up to 94% classification accuracy and 96% prediction accuracy, which satisfies the application requirements of the AM industry. Moreover, the developed model can be adapted for the manufacturability analysis of different AM processes. [ABSTRACT FROM AUTHOR]

• Machine learning-based prediction model for body fat percentage (BFP). • Artificial intelligence-based BFP prediction model for men and women. • BFP prediction model with photoplethysmography signal. • Low-cost BFP prediction model. • High accuracy BFP prediction model. Calculation of body fat percentage (BFP) is a frequently encountered problem in the literature. BFP is one of the most significant parameters which should be processed in body weight control programs. Anthropometric measurements and statistical methods are being used generally in the literature for BFP estimation. Artificial intelligence and gender-based models with a photoplethysmography signal (PPG) were proposed for BFP estimation in this study. In the study, the PPG signal is divided into lower frequency bands, and 25 features are taken out from each frequency band. Artificial intelligence algorithms were created by reducing the extracted features with the help of a feature selection algorithm. According to the results obtained, models with performance values of R M S E = 0.35 , R = 1 for men, R M S E = 0.87 , R = 1 for women were created. In the best performing models, the PPG signal's high-frequency components are used for men, whereas the low-frequency band of the PPG signal is used for women. As a result, the proposed model in this study is considered to be used for BFP measurement. [ABSTRACT FROM AUTHOR]

Assembly of biomolecules at solid-water interfaces requires molecules to traverse complex orientation-dependent energy landscapes through processes that are poorly understood, largely due to the dearth of in situ single-molecule measurements and statistical analyses of the rotational dynamics that define directional selection. Emerging capabilities in high-speed atomic force microscopy and machine learning have allowed us to directly determine the orientational energy landscape and observe and quantify the rotational dynamics for protein nanorods on the surface of muscovite mica under a variety of conditions. Comparisons with kinetic Monte Carlo simulations show that the transition rates between adjacent orientation-specific energetic minima can largely be understood through traditional models of in-plane Brownian rotation across a biased energy landscape, with resulting transition rates that are exponential in the energy barriers between states. However, transitions between more distant angular states are decoupled from barrier height, with jump-size distributions showing a power law decay that is characteristic of a nonclassical Levy-flight random walk, indicating that large jumps are enabled by alternative modes of motion via activated states. The findings provide insights into the dynamics of biomolecules at solid-liquid interfaces that lead to selfassembly, epitaxial matching, and other orientationally anisotropic outcomes and define a general procedure for exploring such dynamics with implications for hybrid biomolecular-inorganic materials design. [ABSTRACT FROM AUTHOR]

This paper presents a simple, fast and efficacious system to promote the human action classification outcome using the depth action sequences. Firstly, the motion history images (MHIs) and static history images (SHIs) are created from the front (XOY), side (YOZ) and top (XOZ) projected scenes of each depth sequence in a 3D Euclidean space through engaging the 3D Motion Trail Model (3DMTM). Then, the Local Binary Patterns (LBPs) algorithm is operated on the MHIs and SHIs to learn motion and static hierarchical features to represent the action sequence. The motion and static hierarchical feature vectors are then fed into a classifier ensemble to classify action classes, where the ensemble comprises of two classifiers. Thus, each ensemble includes a pair of Kernel-based Extreme Learning Machine (KELM) or |${\mathrm{l}}_{\mathrm{2}}$| -regularized Collaborative Representation Classifier (⁠|${\mathrm{l}}_{\mathrm{2}}$| -CRC) or Multi-class Support Vector Machine. To extensively assess the framework, we perform experiments on a couple of standard available datasets such as MSR-Action3D , UTD-MHAD and DHA. Experimental consequences demonstrate that the proposed approach gains a state-of-the-art recognition performance in comparison with other available approaches. Several statistical measurements on recognition results also indicate that the method achieves superiority when the hierarchical features are adopted with the KELM ensemble. In addition, to ensure real-time processing capability of the algorithm, the running time of major components is investigated. Based on machine dependency of the running time, the computational complexity of the system is also shown and compared with other methods. Experimental results and evaluation of the computational time and complexity reflect real-time compatibility and feasibility of the proposed system. [ABSTRACT FROM AUTHOR]

During the phase of periodic asphalt pavement survey, patched and unpatched potholes need to be accurately detected. This study proposes and verifies a computer vision-based approach for automatically distinguishing patched and unpatched potholes. Using two-dimensional images, patched and unpatched potholes may have similar shapes. Therefore, this study relies on image texture descriptors to delineate these two objects of interest. The texture descriptors of statistical measurement of color channels, the gray-level cooccurrence matrix, and the local ternary pattern are used to extract texture information from image samples of asphalt pavement roads. To construct a classification model based on the extracted texture-based dataset, this study proposes and validates an integration of the Support Vector Machine Classification (SVC) and the Forensic-Based Investigation (FBI) metaheuristic. The SVC is used to generalize a classification boundary that separates the input data into two class labels of patched and unpatched potholes. To optimize the SVC performance, the FBI algorithm is utilized to fine-tune the SVC hyperparameters. To establish the hybrid FBI-SVC framework, an image dataset consisting of 600 samples has been collected. The experiment supported by the Wilcoxon signed-rank test demonstrates that the proposed computer vision is highly suitable for the task of interest with a classification accuracy rate = 94.833%. [ABSTRACT FROM AUTHOR]

This study assessed the performance of modeling approaches to estimate personal exposure in Kenyan homes where cooking fuel combustion contributes substantially to household air pollution (HAP). We measured emissions (PM2.5, black carbon, CO); household air pollution (PM2.5, CO); personal exposure (PM2.5, CO); stove use; and behavioral, socioeconomic, and household environmental characteristics (eg, ventilation and kitchen volume). We then applied various modeling approaches: a single‐zone model; indirect exposure models, which combine person‐location and area‐level measurements; and predictive statistical models, including standard linear regression and ensemble machine learning approaches based on a set of predictors such as fuel type, room volume, and others. The single‐zone model was reasonably well‐correlated with measured kitchen concentrations of PM2.5 (R2 = 0.45) and CO (R2 = 0.45), but lacked precision. The best performing regression model used a combination of survey‐based data and physical measurements (R2 = 0.76) and a root mean‐squared error of 85 µg/m3, and the survey‐only‐based regression model was able to predict PM2.5 exposures with an R2 of 0.51. Of the machine learning algorithms evaluated, extreme gradient boosting performed best, with an R2 of 0.57 and RMSE of 98 µg/m3. [ABSTRACT FROM AUTHOR]

Background: Sowing time is commonly used as the temporal reference for Arabidopsis thaliana (Arabidopsis) experiments in high throughput plant phenotyping (HTPP) systems. This relies on the assumption that germination and seedling establishment are uniform across the population. However, individual seeds have different development trajectories even under uniform environmental conditions. This leads to increased variance in quantitative phenotyping approaches. We developed the Digital Adjustment of Plant Development (DAPD) normalization method. It normalizes time-series HTPP measurements by reference to an early developmental stage and in an automated manner. The timeline of each measurement series is shifted to a reference time. The normalization is determined by cross-correlation at multiple time points of the time-series measurements, which may include rosette area, leaf size, and number. Results: The DAPD method improved the accuracy of phenotyping measurements by decreasing the statistical dispersion of quantitative traits across a time-series. We applied DAPD to evaluate the relative growth rate in Arabidopsis plants and demonstrated that it improves uniformity in measurements, permitting a more informative comparison between individuals. Application of DAPD decreased variance of phenotyping measurements by up to 2.5 times compared to sowing-time normalization. The DAPD method also identified more outliers than any other central tendency technique applied to the non-normalized dataset. Conclusions: DAPD is an effective method to control for temporal differences in development within plant phenotyping datasets. In principle, it can be applied to HTPP data from any species/trait combination for which a relevant developmental scale can be defined. [ABSTRACT FROM AUTHOR]

Knowledge of subject-specific muscle volumes may be used as surrogates for evaluating muscle strength and power generated by 'fat-free' muscle mass. This study presents population-based statistical learning approaches for predicting 'fat-free' muscle volume from known anthropometric measurements. Using computed tomography (CT) imaging data to obtain lower-limb muscle volumes from 50 men and women, this study evaluated six statistical learning methods for predicting muscle volumes from anthropometric measurements: (i) stepwise regression, (ii) linear support vector machine (SVM), (iii) 2nd-order polynomial SVM, (iv) linear partial least squares regression (PLSR), (v) quadratic PLSR, and (vi) 3rd-order spline fit PLSR. These techniques have successfully been demonstrated in bioengineering applications and freely available in open-source toolkits. Analysis revealed that separating a general population into sexes and/or cohorts based on adipose level may improve prediction accuracies. The most important measures that statistically influence muscle volume predictions were shank girth, followed by sex and finally leg length, as identified using stepwise regression. SVM learning predicted muscle volume with an accuracy of 85 ± 4% when using linear interpolation, but performed poorly with an accuracy of 59 ± 6% using polynomial interpolation. The simpler linear PLSR exhibited muscle volume prediction accuracy of 87 ± 2%, while quadratic PLSR was slightly reduced at 82 ± 3%. For the spline fit PLSR, high accuracy was observed on the training data set (~ 99%) but over-fitting (a drawback of high-interpolation methods) resulted in erroneous predictions on testing data, and hence, the model was deemed unsuitable. In conclusion, use of linear PLSR models with variables of sex, leg length, and shank girth is a useful tool for predicting 'fat-free' muscle volume. [ABSTRACT FROM AUTHOR]

In recent years, deep neural networks have been utilized in a wide variety of applications including image generation. In particular, generative adversarial networks (GANs) are able to produce highly realistic pictures as part of tasks such as image compression. As with standard compression, it is desirable to be able to automatically assess the perceptual quality of generative images to monitor and control the encode process. However, existing image quality algorithms are ineffective on GAN generated content, especially on textured regions and at high compressions. Here we propose a new “naturalness”-based image quality predictor for generative images. Our new GAN picture quality predictor is built using a multi-stage parallel boosting system based on structural similarity features and measurements of statistical similarity. To enable model development and testing, we also constructed a subjective GAN image quality database containing (distorted) GAN images and collected human opinions of them. Our experimental results indicate that our proposed GAN IQA model delivers superior quality predictions on the generative image datasets, as well as on traditional image quality datasets. [ABSTRACT FROM AUTHOR]

We propose to perform wearable sensors-based human physical activity recognition. This is further extended to an Internet-of-Things (IoT) platform which is based on a web-based application that integrates wearable sensors, smartphones, and activity recognition. To this end, a smartphone collects the data from wearable sensors and sends it to the server for processing and recognition of the physical activity. We collect a novel data set of 13 physical activities performed both indoor and outdoor. The participants are from both the genders where their number per activity varies. During these activities, the wearable sensors measure various body parameters via accelerometers, gyroscope, magnetometers, pressure, and temperature. These measurements and their statistical are then represented in features vectors that used to train and test supervised machine learning algorithms (classifiers) for activity recognition. On the given data set, we evaluate a number of widely known classifiers such random forests, support vector machine, and many others using the WEKA machine learning suite. Using the default settings of these classifiers in WEKA, we attain the highest overall classification accuracy of 90%. Consequently, such a recognition rate is encouraging, reliable, and effective to be used in the proposed platform. [ABSTRACT FROM AUTHOR]

To maintain the serviceability of buildings, the owners need to be informed about the current condition of the water supply and waste disposal systems. Therefore, timely and accurate detection of corrosion on pipe surface is a crucial task. The conventional manual surveying process performed by human inspectors is notoriously time consuming and labor intensive. Hence, this study proposes an image processing-based method for automating the task of pipe corrosion detection. Image texture including statistical measurement of image colors, gray-level co-occurrence matrix, and gray-level run length is employed to extract features of pipe surface. Support vector machine optimized by differential flower pollination is then used to construct a decision boundary that can recognize corroded and intact pipe surfaces. A dataset consisting of 2000 image samples has been collected and utilized to train and test the proposed hybrid model. Experimental results supported by the Wilcoxon signed-rank test confirm that the proposed method is highly suitable for the task of interest with an accuracy rate of 92.81%. Thus, the model proposed in this study can be a promising tool to assist building maintenance agents during the phase of pipe system survey. [ABSTRACT FROM AUTHOR]

The current state of the art approach in the simulation of particle-laden flow in turbomachinery is to handle particle–wall interactions via rebound and erosion models. Rebound models often require a priori parameter tuning to match experimental measurements. Moreover, the actual stochastic nature of the rebound is neglected, and the particle is assumed not to fracture upon impact. However, this affects the resulting particle trajectories and is particularly critical at high (normal) impact velocities, where particles in typical aero engine flow exhibit a high probability of fracture, as illustrated in our previous work. In this work, we propose a method to develop a generalized rebound model which is parameter-free for the user and considers the stochasticity of the rebound. To this end, state of the art methods from function approximation, more precisely, deep dense neural networks are employed. The networks are trained through a supervised learning approach, where the neural network maps the impacting particles' characteristics to its new particle trajectory after rebound. For this, we present an efficient method to predict probability distributions in a supervised learning context without a priori parameter tuning of known PDFs. In a second step, we extend the network to account for particle fracture, where the particle breakage is based on a fracture probability distribution to determine whether a particle breaks. The performance of the proposed framework is illustrated by the use of experimental measurements of the statistical rebound of sand particles in an erosion test rig specifically designed to match flow and impact conditions (including particle fracture) as well as particle sizes in aero engine compressors. • We present a methodology to develop rebound models which consider stochastic rebounds. • Dense neural networks to enable an extension of the model to new data and features. • The framework can include statistical and individual measurement or simulation data. • Particle fracture and physical plausibility conditions are considered. • The model can be directly incorporated into CFD. [ABSTRACT FROM AUTHOR]

Bio-composites have become the prime material selection for green concrete because of the increasing awareness of environmental issues. Due to their highly heterogenous nature, most of the existing studies have focused on the experimental investigation of thermal conductivity (TC) and compressive strength (CS) of hemp-based bio-composites (HBC). However, the tests to estimate the TC and CS may take long time and be costly. In this study, three different machine learning (ML) techniques known as artificial neural network (ANN), adaptive neuro-fuzzy interface system (ANFIS), and multi-gene expression programming (MEP) are used to predict and formulate a mathematical expression for determining the CS and TC of HBC. A total of 159 and 86 experimental records for CS and TC, respectively, were achieved from the past published literature and the ten most influential input variables were considered. The performance of the proposed models was tested using root squared error (RSE), mean absolute error (MAE), Nash-Sutcliffe efficiency (NSE), root mean square logarithmic error (RMSLE), root mean square error (RMSE), correlation coefficient (R), coefficient of determination (R2), performance index (PI), the objective function (OF), and other statistical measures recommended in the literature. Sensitivity and parametric studies were also conducted to assess the coherence of the developed MEP equations with the actual physical phenomenon. Based on R, R2 and NSE, the increasing order is ANFIS > MEP > ANN for both CS and TC. The overall R-value for CS models are 0.9929 (ANFIS), 0.9848 (MEP) and 0.9691 (ANN) while 0.9997 (ANFIS) 0.9759 (MEP) and 0.9601 (ANN) for TC. The integrated measurements of statistical performance (PI and OF) of all models approach 0, stating the outburst performance and generalization capability of the developed models. MEP also provides a simple empirical mathematical equation for the prediction of CS and TC. The increasing trend of the importance of inputs variables for CS and TC was calculated. The parametric trend of each input for CS and TC is in strong agreement and consistent with the experimental results in the database. Thus, the projected equations for estimating the CS and TC of HBC are accurate and feasible and can be used by the designer and practitioners to save the total time required of hectic laboratory tests. However, it is suggested to validate the results of this research with the latest experimental data and other ML algorithms must be studied. • Prediction of compressive strength and dry-thermal conductivity of bio-composites using three soft computing techniques (ANN, ANFIS, MEP). • The comparative performance of the developed models was assessed via nine different statistical indicators like RSE, MAE, NSE, RMSLE, RMSE, R, R2, PI, and OF. • Sensitivity and parametric studies were conducted to assess the coherence of the developed MEP equations with the actual physical phenomenon. [ABSTRACT FROM AUTHOR]

Chlorination has made water supply safe from bacteria and viruses, and has almost completely eliminated risks of waterborne diseases such as typhoid fever, cholera and dysentery. However, the health benefit of chlorination has introduced some possible risks from the by-products of the disinfection process. In this work, we propose the use of a learning algorithm 'support vector machine (SVM)' for the prediction of the water pollutant total exposure index (TEI) and cumulative risk (CR). The SVM is an artificial intelligent approach that could capture the input/output mapping from the given data. Support vector machines were developed based on the structural risk minimization principle from statistical learning theory in which the empirical risk and the machine complexity are optimized. The proposed algorithm was validated by using statistical measurements on experimental data different from those used for the training (out-of sample testing) to evaluate the algorithm's 'generalizability'. The experimental results demonstrate a good correlation between the SVM predicted values and those conventionally estimated using probabilistic (Monte Carlo) simulations. In both cases, the TEI and CR values were estimated using a set of data extracted from the national database IRIS. The SVM, once trained with representative data, could be used easily and effectively in comparison with other conventional estimation methods of TEI and CR. [ABSTRACT FROM AUTHOR]

A drawback of traditional data-mining methods is that they do not leverage prior knowledge of users. In prior work, we proposed a method that could discover unexpected patterns in data by using domain knowledge in a systematic manner. In this paper, we present new methods for discovering a minimal set of unexpected patterns by combining the two independent concepts of minimality and unexpectedness, both of which have been well-studied in the KDD literature. We demonstrate the strengths of this approach experimentally using a case study in a marketing domain. [ABSTRACT FROM AUTHOR]

Data mining uncovers hidden, previously unknown, and potentially useful information from large amounts of data. Compared to the traditional statistical and machine learning data analysis techniques, data mining emphasizes providing a convenient and complete environment for the data analysis. In this paper, we propose an integrated framework for visualized, exploratory data clustering, and pattern extraction from mixed data. We further discuss its implementation techniques: a generalized self-organizing map (GSOM) and an extended attribute-oriented induction (EAOI), which not only overcome the drawbacks of their original algorithms, but also provide additional analysis capabilities. Specifically, the GSOM facilitates the direct handling of mixed data, including categorical and numeric values. The EAOI enables exploration for major values hidden in the data and, in addition, offers an alternative for processing numeric attributes, instead of generalizing them. A prototype was developed for experiments with synthetic and real data sets, and comparison with those of the traditional approaches. The results confirmed the feasibility of the framework and the superiority of the extended techniques. [ABSTRACT FROM AUTHOR]

• In this study, a new data augmantetion model was proposed. • The presented model was tested by statistical methods. • This model can be used in any work where data augmantetion will be done. Cancer disease is among the leading causes of death in the world today. It has been scientifically proven that if the disease is detected in the first stages, the success rate in treatment can be close to 100%. Accordingly, it can be said that the problems caused by breast cancer can be solved to a great extent thanks to early diagnosis. Data sets that can be processed in early diagnosis are required. Researchers use artificial intelligence techniques to develop systems to assist specialist doctors. It is of great importance to have a data set on which researchers can work. The more parameters of these data sets and the number of these parameters, the more artificial learning process takes place. In this study, both a new data augmentation method for two different classes in the database of breast cancer was presented and breast cancer was diagnosed with this method. First, the database was analyzed statistically with the proposed data augmentation method. Then, the new values obtained from the data augmentation process were added to the database with a 5-times and 10-times augmentation. Experimental studies were conducted between raw database and databases with augmentatied data. The proposed model with 5 different Machine Learning Algorithms (MLA) (k Nearest Neighbors (k-NN), Decision Tree (DT), Naive Bayes (NB), Random Forest (RF)) and Support Vector Machine (SVM)) were tested. The accuracy rates in train and test operations were increased in the following order; for k-NN, same − 15%, for DT, 0.1–13.01%, for RF, same − 14.03%, for NB, 14.7%−19.7%, for SVM, 15.9%−14.06%. The rates obtained from the experimental results proved to be very promising. The proposed method can be used as an alternative data augmentation method for researchers to get more accurate results in their studies. [ABSTRACT FROM AUTHOR]

Accurate measurements of statistical properties, such as the star formation rate and the lifetime of young stellar objects (YSOs) in different stages, are essential for constraining star formation theories. However, it is a difficult task to separate galaxies and YSOs based on spectral energy distributions (SEDs) alone, because they contain both thermal emission from stars and dust around them and no reliable theories can be applied to distinguish them. Here we compare different machine learning algorithms and develop the Spectrum Classifier of Astronomical Objects (SCAO), based on Fully Connected Neural Network (FCN), to classify regular stars, galaxies, and YSOs. Superior to previous classifiers, SCAO is solely trained by high quality data labeled in Molecular Cores to Planet-forming Disks (c2d) catalog without a priori theoretical knowledge, and provides excellent results with high precision (> 96%) and recall (> 98%) for YSOs when only eight bands are included. We systematically investigate the effects of observation errors and distance effects, and show that high accuracy performance is still maintained even when using fluxes of only three bands (IRAC 3, a=IRAC 4, and MIPS 1) in the long wavelengths regime, because the silicate absorption feature is automatically detected by SCAO. Finally, we applied SCAO to Spitzer Enhanced Imaging Products (SEIP), the most complete catalog of Spitzer observations, and found 129219 YSO candidates. The website from SCAO is available at http://scao.astr.nthu.edu.tw. [ABSTRACT FROM AUTHOR]

In an Internet of Things (IoT) environment, a large volume of potentially confidential data might be leaked from sensors installed everywhere. To ensure the authenticity of such sensitive data, it is important to initially verify the source of data and its identity. Practically, IoT device identification is the primary step toward a secure IoT system. An appropriate device identification approach can counteract malicious activities such as sending false data that trigger irreparable security issues in vital or emergency situations. Recent research indicates that primary identity metrics such as Internet Protocol (IP) or Media Access Control (MAC) addresses are insufficient due to their instability or easy accessibility. Thus, to identify an IoT device, analysis of the header information of packets by the sensors is of imperative consideration. This paper proposes a combination of sensor measurement and statistical feature sets in addition to a header feature set using a classification-based device identification framework. Various machine Learning algorithms have been adopted to identify different combinations of these feature sets to provide enhanced security in IoT devices. The proposed method has been evaluated through normal and under-attack circumstances by collecting real-time data from IoT devices connected in a lab setting to show the system robustness. [ABSTRACT FROM AUTHOR]

Low-cost air quality sensors can help increase spatial and temporal resolution of air pollution exposure measurements. These sensors, however, most often produce data of lower accuracy than higher-end instruments. In this study, we investigated linear and random forest models to correct PM 2.5 measurements from the Denver Department of Public Health and Environment (DDPHE)'s network of low-cost sensors against measurements from co-located U.S. Environmental Protection Agency Federal Equivalence Method (FEM) monitors. Our training set included data from five DDPHE sensors from August 2018 through May 2019. Our testing set included data from two newly deployed DDPHE sensors from September 2019 through mid-December 2019. In addition to PM 2.5 , temperature, and relative humidity from the low-cost sensors, we explored using additional temporal and spatial variables to capture unexplained variability in sensor measurements. We evaluated results using spatial and temporal cross-validation techniques. For the long-term dataset, a random forest model with all time-varying covariates and length of arterial roads within 500 m was the most accurate (testing RMSE = 2.9 μg/m3 and R2 = 0.75; leave-one-location-out (LOLO)-validation metrics on the training set: RMSE = 2.2 μg/m3 and R2 = 0.93). For on-the-fly correction, we found that a multiple linear regression model using the past eight weeks of low-cost sensor PM 2.5 , temperature, and humidity data plus a near-highway indicator predicted each new week of data best (testing RMSE = 3.1 μg/m3 and R2 = 0.78; LOLO-validation metrics on the training set: RMSE = 2.3 μg/m3 and R2 = 0.90). The statistical methods detailed here will be used to correct low-cost sensor measurements to better understand PM 2.5 pollution within the city of Denver. This work can also guide similar implementations in other municipalities by highlighting the improved accuracy from inclusion of variables other than temperature and relative humidity to improve accuracy of low-cost sensor PM 2.5 data. Image 1 • Random forest regression worked best for calibrating/correcting archived PM 2.5 sensor data. • Timespan-optimized linear regression worked best for correcting data on-the-fly. • Road/traffic variables improved both types of calibration/correction. • Spatial cross-validation techniques help evaluate models. • Our models will be used for real-time air pollution communication in Denver. [ABSTRACT FROM AUTHOR]

This paper presents a new Quantitative Structure-Activity Relationship (QSAR) model based on Extreme Learning Machine (ELM) to predict the biological activity of the benchmark Escape-Data sets compounds in order to provide an effective learning solution for regression analysis. The pre-processing phase of this model has been performed for the chemo-genomics datasets using the k-Nearest Neighbours (k-NN) algorithm to predict missing values of the dataset. In the second phase, the Genetic algorithm hybrid with Binary Whale Optimization algorithm (GBWOA) is adapted to determine the significance and the optimized features in feature selection phase. The min–max method is used in the third phase to transform all features to binary form in order to increases the efficiency of the proposed model by smoothing the data points and reducing fluctuation among features. ELM is used in the final phase as regression algorithm to predict chemo-genomics chemical compound. Different experiments have been performed in this paper on datasetwhich has been collected from ExCAPE chemo-genomics database project composed of 43509 compounds, 1134 targets besides biological activity and 40 chemical descriptors. The experimental results show that the proposed model is efficient in improving the level of prediction based on some statistical measurements. Also, ELM produced satisfactory results when the number of hidden nodes is greater than or equal to 1000 L. Moreover, the proposed model achieved high accuracy using R 2 measure (≈ 0.971) which outperforms the other algorithms in literature such as (WOA, ALO, BAT and CSA) with accuracies (≈ 0.673, ≈ 0.753, ≈ 0.680, and ≈ 0.897) respectively. In addition, the docking results succeeded in validating the current QSAR model. In the current research, 41686 (95.81%) compounds are lead compound and 36965 (84.95%) compounds are a candidate for multi-target genes. • QSAR model applied for activity prediction of novel drugs and validated by docking. • QSAR model based on Extreme Learning Machine for (PXC 50) is presented. • ELM proved less training time consumed and it gives high prediction performance. • The GBWOA-ELM aims to improve the WOA by using Genetic Algorithm. • The results show that the (GBWOA-ELM) model is accurate to predict the PXC 50. [ABSTRACT FROM AUTHOR]