Search

For data-parallel frameworks, their communication is highly structured. Coflow is a networking abstraction proposed for their all-or-nothing job-specific semantics. Minimizing coflow completion time (CCT) decreases the completion time of corresponding jobs. However, state-of-the-art coflow scheduling approaches suffer from several drawbacks. On the one hand, both sender-driven and receiver-driven scheduling approaches fail to achieve optimal especially when the bandwidth bottleneck exists. On the other hand, they fail to optimize the number of concurrent connections since the CCT can be prolonged due to too many or too few concurrent connections. In this paper, we propose Django, a bilateral coflow scheduling framework. We first use Support Vector Machine (SVM) as the machine learning model to automatically identify the optimal number of concurrent connections, i.e. , the queue limitation in the switch. Based on the predicted results, we further develop a set of distributed coflow scheduling algorithms in a scalable manner. Testbed experiments and trace-driven simulations show that Django can estimate the number of concurrent connections with an accuracy of 98%, reduce the average CCT and 95th percentile CCT by 15% and 40%, respectively. • State-of-the-art coflow scheduling approaches suffer from several drawbacks. • On the one hand, both sender-driven and receiver-driven scheduling approaches fail to achieve optimal especially when the bandwidth bottleneck exists. • On the other hand, they fail to optimize the number of concurrent connections since the CCT can be prolonged due to too many or too few concurrent connections. • In this paper, we propose Django, a bilateral coflow scheduling framework, which can automatically identify the optimal number of concurrent connections. • Testbed experiments and trace-driven simulations show that Django can estimate the number of concurrent connections with an accuracy of 98%, reduce the average CCT and tail CCT. [ABSTRACT FROM AUTHOR]

In recent years, frequent occurrences of algal blooms due to environmental changes have posed significant threats to the environment and human health. This paper analyzes the reasons of algal bloom from the perspective of environmental factors such as nutrients, temperature, light, hydrodynamics factors and others. Various commonly used algal bloom monitoring methods are discussed, including traditional field monitoring methods, remote sensing techniques, molecular biology-based monitoring techniques, and sensor-based real-time monitoring techniques. The advantages and limitations of each method are summarized. Existing algal bloom prediction models, including traditional models and machine learning (ML) models, are introduced. Support Vector Machine (SVM), deep learning (DL), and other ML models are discussed in detail, along with their strengths and weaknesses. Finally, this paper provides an outlook on the future development of algal bloom warning techniques, proposing to combine various monitoring methods and prediction models to establish a multi-level and multi-perspective algal bloom monitoring system, further improving the accuracy and timeliness of early warning, and providing more effective safeguards for environmental protection and human health. • Reviewed the environmental factors of algal blooms. • Reviewed machine learning models for algal bloom early warning. • Reviewed the basic process of machine learning early warning. • Perspectived the future directions for algal bloom early warning. [ABSTRACT FROM AUTHOR]

Predicting the performance of applications executed on GPUs is a great challenge and is essential for efficient job schedulers. There are different approaches to do this, namely analytical modeling and machine learning (ML) techniques. Machine learning requires large training sets and reliable features, nevertheless it can capture the interactions between architecture and software without manual intervention. In this paper, we compared a BSP-based analytical model to predict the time of execution of kernels executed over GPUs. The comparison was made using three different ML techniques. The analytical model is based on the number of computations and memory accesses of the GPU, with additional information on cache usage obtained from profiling. The ML techniques Linear Regression, Support Vector Machine, and Random Forest were evaluated over two scenarios: first, data input or features for ML techniques were the same as the analytical model and, second, using a process of feature extraction, which used correlation analysis and hierarchical clustering. Our experiments were conducted with 20 CUDA kernels, 11 of which belonged to 6 real-world applications of the Rodinia benchmark suite, and the other were classical matrix-vector applications commonly used for benchmarking. We collected data over 9 NVIDIA GPUs in different machines. We show that the analytical model performs better at predicting when applications scale regularly. For the analytical model a single parameter λ is capable of adjusting the predictions, minimizing the complex analysis in the applications. We show also that ML techniques obtained high accuracy when a process of feature extraction is implemented. Sets of 5 and 10 features were tested in two different ways, for unknown GPUs and for unknown Kernels. For ML experiments with a process of feature extractions, we got errors around 1.54% and 2.71%, for unknown GPUs and for unknown Kernels, respectively. [ABSTRACT FROM AUTHOR]

• A deep-learning framework to localize and characterize acoustic emissions. • A single sensor approach for metallic panels with rivet-connected doublers. • A procedure to determine the optimal number of deep learning layers and nodes. • Sensitivity analyses to the number of sensors, input patterns, and output layers. • Experimental validations, comprehensive comparisons, and generalization assessments. This paper introduces a deep learning-based framework to localize and characterize acoustic emission (AE) sources in plate-like structures that have complex geometric features, such as doublers and rivet connections. Specifically, stacked autoencoders are pre-trained and utilized in a two-step approach that first localizes AE sources and then characterizes them. To achieve these tasks with only one AE sensor, the paper leverages the reverberation patterns, multimodal characteristics, and dispersive behavior of AE waveforms. The considered waveforms include AE sources near rivet connections, on the surface of the plate-like structure, and on its edges. After identifying AE sources that occur near rivet connections, the proposed framework classifies them into four source-to-rivet distance categories. In addition, the paper investigates the sensitivity of localization results to the number of sensors and compares their localization accuracy with the triangulation method as well as machine learning algorithms, including support vector machine (SVM) and shallow neural network. Moreover, the generalization of the deep learning approach is evaluated for typical scenarios in which the training and testing conditions are not identical. To train and test the performance of the proposed approach, Hsu-Nielsen pencil lead break tests were carried out on two identical aluminum panels with a riveted stiffener. The results demonstrate the effectiveness of the deep learning-based framework for single-sensor, AE-based structural health monitoring of plate-like structures. [ABSTRACT FROM AUTHOR]

Coastal sites have been recognized to be among the most productive and dynamic regions. Nevertheless, these areas have been disrupted due to the rapid expansion of human interventions and climate change. Coastal ecosystems are becoming more endangered by the probable adverse effects of climate change as revealed by IPCC panel. The main objective of this paper is to integrate Machine Learning (ML) models to spatially predict coastal vulnerability along the Mediterranean coast of (Tangier-Tetouan), Morocco. Different models including Artificial neural network (ANN), Decision Tree (DT), logistic regression (LR), random forest (RF), and Support vector machine (SVM) are used on 10033 points to represent the vulnerability in terms of (high, moderate, and low), determined by the coastal vulnerability index (CVI). Eight coastal vulnerability predictors (Slope, Elevation, Tidal range, Sea level rise, maximum wave height, Natural habitat, Geomorphology, and Shoreline change rate) were considered as input for each model. The obtained dataset was split into two parts: 30% for testing and 70% for training. The presented models were examined on various parameters, such as accuracy, recall, F1 score, root mean square error (RMSE), and kappa index. The findings show that accuracy and Kappa statistics for RF and DT and SVM ranged from (92–99%) which indicates a better performance than other proposed models. The approach is considered an appropriate tool for predicting and mapping coastal vulnerability and can make significant contributions to the successful management of coastal zones. Additionally, it can aid in the development of public interest and promote the planning, creation, and execution of adaption measures. The prediction was applied over (∼124 km) of the shoreline. Results showed that about 19.95% (∼27.73 km) of the shoreline is highly vulnerable, while 20.92% (∼25.95 km) is moderately vulnerable and 59.1% (∼73.32 km) presents a low vulnerability. • This paper aims to integrate Machine Learning (ML) models. • We spatially predict coastal vulnerability along Mediterranean coast of Morocco. • Different models including Artificial neural network (ANN) are used. • The models are used on 10033 points to represent the vulnerability coastal vulnerability index (CVI). • Different predictors were considered for each model. [ABSTRACT FROM AUTHOR]

Due to issues such as limited variability in monitoring data across different tool wear conditions and interference during the machining process, data-driven monitoring models are susceptible to misclassification. Therefore, this paper proposes a pioneering approach that takes into account the tool wear law and the characteristic distribution of tool wear monitoring data. Specifically, the paper proposes an automatic feature extraction and tool condition monitoring method based on Siamese Long Short-term Memory Networks (SLSTMs) to transform the original data distribution, enhance the differentiation of different tool wear condition monitoring data, and achieve accurate prediction of tool wear condition. Additionally, a hybrid data and mechanism-driven tool wear monitoring method is proposed that takes advantage of the fact that tool wear is continuous and irreversible to enable reliable adjustment of the monitoring results of the data-driven model without human intervention. The study conducted milling experiments on a three-axis vertical machining center, using TA2 titanium alloy as the workpiece material. Vibration signals from the spindle in three directions were collected as input to the network, with tool wear state labeled as "0″ or "1″ as the output. Classical machine learning algorithms such as Decision Trees, K-Nearest Neighbors (KNN) and Support Vector Machines (SVM), as well as classical deep learning algorithms such as Convolutional Neural Networks (CNN), Sparse Stacked Autoencoders (SSAE) and Long Short-Term Memory Neural Networks (LSTM), were used to construct and compare feature extraction and state monitoring models. The proposed model based on Siamese Long Short-Term Memory Neural Networks (SLSTMs) achieved testing average accuracy of 98.2% (σ2 = 0.44), outperforming typical deep learning algorithms such as CNN, LSTM and SSAE as well as traditional machine learning algorithms such as Decision Trees, KNN and SVM in terms of accuracy and robustness. Moreover, the proposed data and mechanism hybrid-driven tool wear monitoring method can effectively improve the monitoring accuracy and robustness of data-driven models such as CNN and SSAE. The monitoring accuracy of CNN and SSAE increased from 95.9% to 97.6% and 91.0% to 94.3%, respectively. [ABSTRACT FROM AUTHOR]

Contaminated recyclables, which are frequently discarded as waste, pose a significant challenge to the implementation of a circular economy. These contaminated recyclables impede the circulation of resources, resulting in higher processing costs at material recovery facilities (MRFs). Over the past few decades, machine learning (ML) models such as linear regression (LR), support vector machine (SVM), and random forest (RF) have evolved to provide new methods for predicting inbound contamination rates in addition to traditional statistical models. In this study, we applied ML models to predict inbound contamination rates using demographic features from 15 counties in the U.S. with different curbside collection strategies. In general, we found that ML models outperformed linear mixed models. Specifically, SVM models had the highest performance (R2 = 0.75; mean absolute error (MAE) = 0.06), which may be due to their ability to model nonlinear relationships between features and inbound contamination rates. The key predictor was population, with poverty rate being positively correlated and median age negatively correlated with inbound contamination rates. To improve the management of contamination and enhance the implementation of a circular economy, better models are needed to understand and estimate inbound contamination rates as well as identify critical factors in the present and future. • We examine the factors influencing inbound paper contamination in MRFs. • Importance of the factors are determined by three machine learning models. • The contamination levels in single stream were found to be higher than that of commingled. • SVM regressor obtained the best performance. • Population size is determined to be the most significant factor by all three ML methods. [ABSTRACT FROM AUTHOR]

[Display omitted] Boundary conditions are crucial for the construction of physical models of concrete-filled steel tube (CFST) arches, but the current model updating methods seldom consider boundary conditions and usually necessitate repeated modifications and recalculations of finite element models, which are not efficient enough. Moreover, the current physics-informed neural network (PINN) provides a new approach for solving physical equations through physical constraint merging into the neural network, but it is still limited by complex structures, multiple scales, and unknown physical model engineering problems. Therefore, this paper proposed a boundary recognition method for physically-enhanced machine learning. First, the basic variables of the physical equation and the variables to be optimized were taken as inputs to enhance the neural network's ability to predict the target response. Then, similar to the handling of physical loss constraints in the PINN, the physical loss between the physical model (FE) and the neural network predictions was used to update the parameters. Then, a physical feedback-enhanced neural network (PFENN) meta -model was established, and modified mutation particle swarm optimization (MMPSO) was adopted to optimize the identification of boundary parameters. The identification process involves validation via mutual verification using a finite element model, meta -model, and experimental model. The prediction accuracy is higher than that of classical Kriging, response surface (RS), support vector machine regression (SVR), and neural network (NN) meta -models; additionally, the proposed model has good optimization stability, and reliable optimization results, and its computational efficiency is increased by 4.57 times compared with that of the traditional non-metamodel method. Moreover, the proposed method enhances the accuracy and interpretability of surrogate models by enhancing physical parameters and reducing physical losses, which is suitable for boundary recognition and physical model construction of different structures. [ABSTRACT FROM AUTHOR]

Electrical activity in plants undergoes potential changes in interpreting the plant's physiological state. These electrical signals show arbitrary and probabilistic dynamics in which stress can be detected and analysed in the early stages of symptom appearance in plants. Evaluation of uncertainty in plant signals by signal strength enhancement is stress-specific. In this paper, we considered physiological signals from 15 days old soybean plants that endured three types of stress (cold, osmotic and lowlight). A novel enhancement methodology is proposed to pre-process the stress information present in the electrical signals. Sixteen spectral features were computed from spectrograms of enhanced signals to attain a better hypothesis. To verify the performance metrics of pre and post-enhancement methods, classification is done in two levels (Multi-class, Binary-class) using machine learning algorithms (Decision Trees (DT), K- Nearest Neighbors (KNN), Support Vector Machines (SVM), Ensemble Boosted Tree (EBT), Ensemble Bagged Tree (EBGT), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA)). Superlative feature pairs showing classification accuracy greater than 95% after 5-fold cross-validation were nominated. The results obtained by the SVM classifier in cold stress and KNN in lowlight stress are accurate to 98% after enhancement. The successful premises taken from this work assure early-stage detection and classification of plant stress, thereby suggesting future requirements to develop enhancement & classification techniques according to the type of stress. • Plant electrical signals play a significant role in comprehending plant stresses. • Accuracies improved after processing signals by enhancement methods. • Spectral feature extraction and feature pairing structured the classification model. • Classification performance is greatly improved after enhancement. • Various ML classifiers obtained results greater than 95% in classifying abiotic stress. [ABSTRACT FROM AUTHOR]

Coastal areas are at a higher risk of flooding, and novel changes in the climate are induced to raise the sea level. Flood acceleration and frequency have increased recently because of unplanned infrastructural conveniences and anthropogenic activities. Therefore, the assessment of flood susceptibility mapping is considered the most significant flood management model. In this paper, flood susceptibility identification is performed by applying the innovative Multi-criteria decision-making model (MCDM) called Analytical Hierarchy Process (AHP) by ensembles with Support vector machine (AHP-SVM) and Decision Tree (AHP-DT). This model combines two Representation concentration pathway (RCP) scenarios such as RCP 2.6 & RCP 8.5. The factors influencing the coastal flooding in Bandar Abbas, Iran, identified through Flood susceptibility mapping. Multi-criteria decision-making (MCDM) has been applied to evaluate the Coastal flood conditioning factors, and ensemble machine learning (ML) approaches are employed for Coastal risk factor (CRF) prediction and classification. The statistical variances are measured through Friedman and Wilcoxon signed rank tests and statistical metrics such as Accuracy, sensitivity, and specificity. Among the models, AHP-DT obtained an improved AUC value of ROC as 0.95. After applying the ML models, the northern and western park of Raidak Basin River recognises very low and low flood susceptibility because of their topographic characteristics. The eastern part of the middle section fell very high and high CFSM. Observed from this result analysis, the people living nearer to the coastline are distributed by the low to medium exposure in the region of the west and middle of the considered study area. The results of this study can help decision-makers take necessary risk reduction approaches in the high-risk flooding zones of the coastal system. • Coastal areas are facing an increased risk of flooding due to rising sea levels and the impact of climate change. • Floods increased due to unplanned infrastructure and human activities. • The AHP-SVM-DT combination for flood susceptibility mapping is a notable flood management model. • The study uses flood susceptibility mapping to examine Bandar Abbas, Iran, to identify factors impacting coastal flooding. • AHP-DT model achieved AUC 0.95, effectively predicting and classifying coastal flood risk factors. [ABSTRACT FROM AUTHOR]

This paper proposes a novel targeted blend of machine learning (ML) based approaches for controlling wastewater treatment plant (WWTP) operation by predicting distributions of key effluent parameters of a biological nutrient removal (BNR) process. Two years of data were collected from Plajyolu wastewater treatment plant in Kocaeli, Türkiye and the effluent parameters were predicted using six machine learning algorithms to compare their performances. Based on mean absolute percentage error (MAPE) metric only, support vector regression machine (SVRM) with linear kernel method showed a good agreement for COD and BOD 5 , with the MAPE values of about 9% and 0.9%, respectively. Random Forest (RF) and EXtreme Gradient Boosting (XGBoost) regression were found to be the best algorithms for TN and TP effluent parameters, with the MAPE values of about 34% and 27%, respectively. Further, when the results were evaluated together according to all the performance metrics, RF, SVRM (with both linear kernel and RBF kernel), and Hybrid Regression algorithms generally made more successful predictions than Light GBM and XGBoost algorithms for all the parameters. Through this case study we demonstrated selective application of ML algorithms can be used to predict different effluent parameters more effectively. Wider implementation of this approach can potentially reduce the resource demands for active monitoring the environmental performance of WWTPs. • A novel targeted blend of machine learning (ML) based approaches for controlling wastewater treatment plant. • It is used to predict effluent parameters of a biological nutrient removal process. • ML provides opportunity for early intervention to problems in wastewater treatment. • Five different ML algorithms are tested for their performance. • The mean absolute percent errors of models were 9% for COD and 0.9% for BOD5. [ABSTRACT FROM AUTHOR]

Cancer, the second largest human disease, has become a major public health problem. The prediction of chemicals' carcinogenicity before their synthesis is crucial. In this paper, seven machine learning algorithms (i.e., Random Forest (RF), Logistic Regression (LR), Support Vector Machines (SVM), Complement Naive Bayes (CNB), K-Nearest Neighbor (KNN), XGBoost, and Multilayer Perceptron (MLP)) were used to construct the carcinogenicity triple classification prediction (TCP) model (i.e., 1A, 1B, Category 2). A total of 1444 descriptors of 118 hazardous organic chemicals were calculated by Discovery Studio 2020, Sybyl X-2.0 and PaDEL-Descriptor software. The constructed carcinogenicity TCP model was evaluated through five model evaluation indicators (i.e., Accuracy, Precision, Recall, F1 Score and AUC). The model evaluation results show that Accuracy, Precision, Recall, F1 Score and AUC evaluation indicators meet requirements (greater than 0.6). The accuracy of RF, LR, XGBoost, and MLP models for predicting carcinogenicity of Category 2 is 91.67%, 79.17%, 100%, and 100%, respectively. In addition, the constructed machine learning model in this study has potential for error correction. Taking XGBoost model as an example, the predicted carcinogenicity level of 1,2,3-Trichloropropane (96−18−4) is Category 2, but the actual carcinogenicity level is 1B. But the difference between Category 2 and 1B is only 0.004, indicating that the XGBoost is one optimum model of the seven constructed machine learning models. Besides, results showed that functional groups like chlorine and benzene ring might influence the prediction of carcinogenic classification. Therefore, considering functional group characteristics of chemicals before constructing the carcinogenicity prediction model of organic chemicals is recommended. The predicted carcinogenicity of the organic chemicals using the optimum machine leaning model (i.e., XGBoost) was also evaluated and verified by the toxicokinetics. The RF and XGBoost TCP models constructed in this paper can be used for carcinogenicity detection before synthesizing new organic substances. It also provides technical support for the subsequent management of organic chemicals. • Development of a triple-classification model using 7 machine learning algorithms. • First triple-classification model for carcinogenicity classification prediction. • Verification of triple-classification model by US, EU, and WHO databases. • RF and XGBoost models as wide applications of carcinogenicity prediction. [ABSTRACT FROM AUTHOR]

• Classified vibration signals to detect leakages in real-life pipe networks. • Implemented automated detection using accelerometers and machine learning methods. • Compared applied models based on multiple criteria against SqueezeNet. In addition to the issues of water quality and economic losses, leakages in pipe networks are a potential danger to public safety. Early detection of water leaks in pipe networks is crucial for implementing countermeasures to reduce structural damage and water losses. The method of data collection based on sensors installed in pipeline networks has gained a lot of attention due to its potential for application in real-time monitoring systems for leak detection. However, many models developed on data collected from simulations, engineered tests or field experiments, and few have been validated using real network data. To make further investigations in real water distribution systems (WDSs), machine learning (ML)-based leak detection models were developed and validated using signal data collected by piezoelectric accelerometers installed in WDSs over several cities of China. Nonlinear features describing the non-stationary characteristics contained in leak signals are extracted from the signals to build feature sets as inputs to the common classifiers, i.e., support vector machine (SVM), decision tree (DT), and K-nearest neighbor (KNN). In addition, the spectrograms obtained by the short-time Fourier transform processing are used for the convolutional neural network SqueezeNet. The effects of different features, different dataset compositions, and different models for leak detection are compared and analyzed. SqueezeNet has the best performance with 95.15% accuracy in identifying leaks, and KNN is the best of the three classifiers with superior sensitivity and 88.17% accuracy in identification. This paper demonstrates the effectiveness and practicability of using machine learning models to identify leaks in real pipe networks. The application of machine learning models can perform effective leak detection with minimal human intervention, thus providing a meaningful contribution to the study of leak detection in real WDSs. The proposed model aims to identify leakages under operational conditions in real-life pipe networks with higher efficiency and reduced manual involvement. This paper demonstrates the applicability of accelerometers based on microelectromechanical systems for leak detection and establishes real network-based machine learning models; thereby, providing meaningful contribution to the literature regarding leak detection in a real WDS. [ABSTRACT FROM AUTHOR]

This paper proposes a method of compensating for brake disc balance error by using a machine learning algorithm. Automobile brake discs in the production process will inevitably produce unbalance. The unevenness produced by the uneven mass distribution will produce high-frequency vibration in the process of high-speed rotation, which seriously affects the safety of the vehicle and the personal safety of the occupants. The key measure to solve this problem is to correct the unbalance with higher precision before the brake disc leaves the factory. The traditional correction method is to improve the detection accuracy of unbalance to achieve balance accuracy. In this paper, we hope to improve the balance accuracy by compensating for the errors generated in the correction process. We use the random forest model, decision tree model, and support vector machine model to predict the errors of the balancing machine during the calibration process. The main idea is to take the parameters of the brake disc and the features in the milling process as input and the error amplitude as output. The results show that the stochastic forest model has higher prediction accuracy than the decision tree model and support vector machine model. This method can also predict errors from other sources, such as thermal errors. [ABSTRACT FROM AUTHOR]

Regularized empirical risk minimization including support vector machines plays an important role in machine learning theory. In this paper regularized pairwise learning (RPL) methods based on kernels will be investigated. One example is regularized minimization of the error entropy loss which has recently attracted quite some interest from the viewpoint of consistency and learning rates. This paper shows that such RPL methods and also their empirical bootstrap have additionally good statistical robustness properties, if the loss function and the kernel are chosen appropriately. We treat two cases of particular interest: (i) a bounded and non-convex loss function and (ii) an unbounded convex loss function satisfying a certain Lipschitz type condition. [ABSTRACT FROM AUTHOR]

• Brain decoding using a dense individual fMRI dataset (∼7 h fMRI data per subject). • Challenging benchmark: decoding 21 conditions from single fMRI brain volumes. • The best individual model reaches 58–67 % accuracy, close to state-of-the-art group decoding (76 %). • Individual decoding models generalized poorly to other subjects. Brain decoding aims to infer cognitive states from patterns of brain activity. Substantial inter-individual variations in functional brain organization challenge accurate decoding performed at the group level. In this paper, we tested whether accurate brain decoding models can be trained entirely at the individual level. We trained several classifiers on a dense individual functional magnetic resonance imaging (fMRI) dataset for which six participants completed the entire Human Connectome Project (HCP) task battery >13 times over ten separate fMRI sessions. We evaluated nine decoding methods, from Support Vector Machines (SVM) and Multi-Layer Perceptron (MLP) to Graph Convolutional Neural Networks (GCN). All decoders were trained to classify single fMRI volumes into 21 experimental conditions simultaneously, using ∼7 h of fMRI data per participant. The best prediction accuracies were achieved with GCN and MLP models, whose performance (57–67 % accuracy) approached state-of-the-art accuracy (76 %) with models trained at the group level on >1 K hours of data from the original HCP sample. Our SVM model also performed very well (54–62 % accuracy). Feature importance maps derived from MLP —our best-performing model— revealed informative features in regions relevant to particular cognitive domains, notably in the motor cortex. We also observed that inter-subject classification achieved substantially lower accuracy than subject-specific models, indicating that our decoders learned individual-specific features. This work demonstrates that densely-sampled neuroimaging datasets can be used to train accurate brain decoding models at the individual level. We expect this work to become a useful benchmark for techniques that improve model generalization across multiple subjects and acquisition conditions. [Display omitted] [ABSTRACT FROM AUTHOR]

The technological evolution of the smart grids is going to take its shape in the form of a new paradigm called the Internet of Energy (IoE); which is considered to be the convergence of internet, communication, and energy. Like other evolved technologies, the IoE inherits security vulnerabilities from its constituents that need to be addressed. Intrusion Detection Systems (IDS) have been used to counteract malicious attacks. Among the types of IDS, anomaly-based IDS that employ mostly machine learning algorithms are considered to be the promising one, owing to their capability of detecting zero-day attacks. However, using complex algorithms to detect attacks, the existing anomaly-based IDS designed for IoE require considerable amount of time. It is tempting to reduce the training and testing time in order to make the IDS feasible for the IoE architecture. In this paper, we propose a hybrid anomaly-based IDS that can be installed at any networked site of the IoE architecture, such as Advanced Metering Infrastructure (AMI), to counteract security attacks. Our proposed system reduces the overall classification time of detection compared to the existing hybrid methods. The proposed solution uses a combination of K-means and Support Vector Machine, where the K-means centroids are used in a unique training method that reduces the training and testing times of the Support Vector Machine without compromising classification performance. We choose the best value of "k" and fine-tuned the SVM for best anomaly detection. Our approach achieves the highest accuracy of 99.9% in comparison with the existing approaches. • A hybrid anomaly-based IDS is proposed for the Internet of Energy architecture. • Our approach achieves 99.9% accuracy as compared to the existing approaches. • Our approach performs 1367 times faster than the existing approaches. [ABSTRACT FROM AUTHOR]

Cell wall lytic enzymes play key roles in biochemical, morphological, genetic research and industry fields. To save time and labor costs， bioinformatic methods are usually adopted to narrow the scope of in vitro experimentation. In this paper, we established a novel machine learning (support vector machine) based identifier called CWLy-pred to identify cell wall lytic enzymes. An improved MRMD feature selection method is also proposed to select the optimal training set to avoid data redundancy. CWLy-pred obtains an accuracy of 93.067%, a sensitivity of 85.3%, a specificity of 94.8%, an MCC of 0.775 and an AUC of 0.900. It outperforms the state-of-the-art identifier in terms of accuracy, sensitivity, specificity and MCC. Our proposed model is based on a feature set of only 6 dimensions; therefore, it not only can overcome overfitting problems but can also supervise biological experiments effectively. CWLy-pred is embedded in a web application at http://server.malab.cn/CWLy-pred/index.jsp , which is accessible for free. Cell wall lytic enzymes, as an important biotechnical tool in drug development, agriculture and the food industry. In this research, the accurate identification of cell wall lytic enzymes is one of the key and fundamental tasks. In this study, in order to eliminate the inefficiency of in vitro experiments, a support vector machine-based cell wall lytic enzyme identification model was constructed using bioinformatics. This machine learning process includes data collection, feature extraction, feature selection, model training. In prediction phase, the optimal feature set is extracted from the query sequence. Then, it is fed to the proposed model to predict whether it is cell wall lytic enzyme or not. In training phase, the 473-dimensional feature set is extracted, and the optimal feature set is selected by an improved MRMD selection method (MRMD + IFS). Then, the SVM-based model is trained and optimized using the above optimal feature set. The model obtains an accuracy of 93.067%, a sensitivity of 85.3%, a specificity of 94.8%, an MCC of 0.775 and an AUC of 0.900. It outperforms the state-of-the-art identifier in terms of accuracy, sensitivity, specificity and MCC. Our proposed model is based on a feature set of only 6 dimensions; therefore, it not only can overcome overfitting problems but can also supervise biological experiments effectively. Unlabelled Image • We established a novel machine learning (support vector machine) based cell wall lytic enzymes identifier. • The six most important recognition features are located. • A new improved MRMD feature selection method is proposed. [ABSTRACT FROM AUTHOR]

Knowledge of the sub-cellular localization of the most diverse class of transcribed RNA, long non-coding RNAs (lncRNAs) will lead us to identify different types of cancers and other diseases as lncRNAs play key role in related cellular functions. In recent days with the exponential growth of known records, it becomes essential to establish new machine learning based techniques to identify the new one due to faster and cheaper solutions provided compared to laboratory methods. In this paper, we propose Locate-R, a novel method for predicting the sub-cellular location of lncRNAs. We have used only n -gapped l -mer composition and l -mer composition as features and select best 655 features to build the model. This model is based locally deep support vector machines which significantly enhance the prediction accuracy with respect to exiting state-of-the-art methods. Our predictor is readily available for use as a stand-alone web application from: http://locate-r.azurewebsites.net/. • An efficient feature extraction and selection procedure. • An extensible framework/methodology for binary and multi-class classification problems. • A prediction tool available for RNA location prediction. [ABSTRACT FROM AUTHOR]

Abstract RNA editing process like Adenosine to Intosine (A-to-I) often influences basic functions like splicing stability and most importantly the translation. Thus knowledge about editing sites is of great importance in molecular biology. With the growth of known editing sites, machine learning or data centric approaches are now being applied to solve this problem of prediction of RNA editing sites. In this paper, we propose EPAI-NC, a novel method for prediction of RNA editing sites. We have used l -mer composition and n -gapped l -mer composition as features and used Pearson Correlation Coefficient to select features according to Pareto Principle. Locally deep support vector machines were used to train the classification model of EPAI-NC. EPAI-NC significantly enhances the prediction accuracy compared to the previous state-of-the-art methods when tested on standard benchmark and independent dataset. Highlights • Locally Deep Support Vector Machine for prediction of Adenosine to Inosine RNA Editing sites. • Easy to generate nucleotide sequence based features with effective feature selection method. • A robust predictor for A-to-I RNA Editing sites with an web based application ready to use. [ABSTRACT FROM AUTHOR]

Highlights • We have polished the result analysis in the experimental section for clarification. • We have added missing references to make the references more comprehensive. • We have carefully gone through the paper and did a thorough revision, including grammatical errors, typos and the formats. Abstract Intelligence technology is an important research area. As a very special yet important case of object recognition, hand-held object recognition plays an important role in intelligence technology for its many applications such as visual question-answering and reasoning. In real-world scenarios, the datasets are open-ended and dynamic: new object samples and new object classes increase continuously. This requires the intelligence technology to enable hybrid incremental learning, which supports both data-incremental and class-incremental learning to efficiently learn the new information. However, existing work mainly focuses on one side of incremental learning, either data-incremental or class-incremental learning while do not handle two sides of incremental learning in a unified framework. To solve the problem, we present a Hybrid Incremental Learning (HIL) method based on Support Vector Machine (SVM), which can incrementally improve its recognition ability by learning new object samples and new object concepts during the interaction with humans. In order to integrate data-incremental and class-incremental learning into one unified framework, HIL adds the new classification-planes and adjusts existing classification-planes under the setting of SVM. As a result, our system can simultaneously improve the recognition quality of known concepts by minimizing the prediction error and transfer the previous model to recognize unknown objects. We apply the proposed method into hand-held object recognition and the experimental results demonstrated its advantage of HIL. In addition, we conducted extensive experiments on the subset of ImageNet and the experimental results further validated the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper, we present a full end-to-end implementation of big data analytics applications in a heterogeneous CPU+FPGA architecture. Selecting the optimal architecture that results in the highest acceleration for big data applications requires an in-depth of each application. Thus, we develop the MapReduce implementation of K -means, K nearest neighbor, support vector machine and naive Bayes in a Hadoop Streaming environment that allows developing mapper functions in a non-Java based language suited for interfacing with FPGA-based hardware accelerating environment. We further profile various components of Hadoop MapReduce to identify candidates for hardware acceleration. We accelerate the mapper functions through hardware+software (HW+SW) co-design. Moreover, we study how various parameters at the application (size of input data), system (number of mappers running simultaneously per node and data split size), and architecture (choice of CPU core such as big vs little, e.g., Xeon vs Atom) levels affect the performance and power-efficiency benefits of Hadoop streaming hardware acceleration and the overall performance and energy-efficiency of the system. A promising speedup as well as energy-efficiency gains of up to 8.3 × and 15 × is achieved, respectively, in an end-to-end Hadoop implementation. Our results show that HW+SW acceleration yields significantly higher speedup on Atom server, reducing the performance gap between little and big cores after the acceleration. On the other hand, HW+SW acceleration reduces the power consumption of Xeon server more significantly, reducing the power gap between little and big cores. Our cost Analysis shows that the FPGA-accelerated Atom server yields execution times that are close to or even lower than stand-alone Xeon server for the studied applications, while reducing the server cost by more than 3 × . We confirm the scalability of FPGA acceleration of MapReduce by increasing the data size on 12-node Xeon cluster and show that FPGA acceleration maintains its benefit for larger data sizes on a cluster. [ABSTRACT FROM AUTHOR]

In recent years, machine learning approaches have been successfully applied to the field of neuroimaging for classification and regression tasks. However, many approaches do not give an intuitive relation between the raw features and the diagnosis. Therefore, they are difficult for clinicians to interpret. Moreover, most approaches treat the features extracted from the brain (for example, voxelwise gray matter concentration maps from brain MRI) as independent variables and ignore their spatial and anatomical relations. In this paper, we present a new Support Vector Machine (SVM)-based learning method for the classification of Alzheimer's disease (AD), which integrates spatial-anatomical information. In this way, spatial-neighbor features in the same anatomical region are encouraged to have similar weights in the SVM model. Secondly, we introduce a group lasso penalty to induce structure sparsity, which may help clinicians to assess the key regions involved in the disease. For solving this learning problem, we use an accelerated proximal gradient descent approach. We tested our method on the subset of ADNI data selected by Cuingnet et al. (2011) for Alzheimer's disease classification, as well as on an independent larger dataset from ADNI. Good classification performance is obtained for distinguishing cognitive normals (CN) vs. AD, as well as on distinguishing between various sub-types (e.g. CN vs. Mild Cognitive Impairment). The model trained on Cuignet's dataset for AD vs. CN classification was directly used without re-training to the independent larger dataset. Good performance was achieved, demonstrating the generalizability of the proposed methods. For all experiments, the classification results are comparable or better than the state-of-the-art, while the weight map more clearly indicates the key regions related to Alzheimer's disease. [ABSTRACT FROM AUTHOR]

[Display omitted] • New version of USDA-BFPD released. • Nutrients prediction from ingredients statement achieved by machine learning models. • MLP achieved the highest accuracy for carbohydrates prediction up to 0.900. Automatic, accurate, and robust prediction of food attributes using emerging machine learning techniques including artificial intelligence may be helpful to rapid food analysis, and personal dietary record. In this paper we have, for the first time, evaluated 5 machine learning models for quantitatively predicting nutrients from foods based on their ingredients. Based on the USDA Global Branded Food Products Database (BFPD), we prepared a machine-readable dataset for two domains - nutrient and ingredient. Based on these datasets, we investigated five prominent models, including AdaBoost, Bayesians, Linear Support Vector Machine (SVM), Multi-Layer Perceptron (MLP) and one tailored-improved MLP called MLPcr to predict 13 label nutrients included in BFPD based on their ingredients. We report on 3 varied nutrients in this paper – carbohydrates, protein, and sodium. The prediction based on the neural network model, MLP, and MLPcr achieved the most accuracy, as high as 0.900 for carbohydrates. A detailed evaluation of the prediction results found that the data distribution and multi-factor complexity have an essential impact on the accuracy of the final prediction. Our research shows the possibility of using neural networks for prediction of nutrients using food ingredients, as well as potential use of neural network applications to the broader scope of food research. [ABSTRACT FROM AUTHOR]

Machine learning algorithms often suffer from good generalization in testing domains especially when the training (source) and test (target) domains do not have similar distributions. To address this problem, several domain adaptation techniques have been proposed to improve the performance of the learning algorithms when they face accuracy degradation caused by the domain shift problem. In this paper, we focus on the non-homogeneous distributed target domains and propose a new latent subdomain discovery model to divide the target domain into subdomains while adapting them. It is expected that applying adaptation on subdomains increase the rate of detection in comparing with the situation that the target domain is seen as one single domain. The proposed division method considers each subdomain as a cluster which has the definite ratio of positive to negative samples, linear discriminability and conditional distribution similarity to the source domain. This method divides the target domain into subdomains while adapting the trained target classifier for each subdomain using Adapt-SVM adaptation method. It also has a simple solution for selecting the appropriate number of subdomains. We call our proposed method Cluster-based Adaptive SVM or CA-SVM in short. We test CA-SVM on two different computer vision problems, pedestrian detection and image classification. The experimental results show the advantage in accuracy rate for our approach in comparison to several baselines. [ABSTRACT FROM AUTHOR]

Tires are the only components of vehicles in contact with the road surface. The tire–road interaction yields many dynamic parameters that have an impact on the final behavior of the vehicle, such as the forces in the tire–road interaction, the length of the contact patch, the velocity in the contact patch, and the effective radius of the tire. Previous studies have shown the feasibility of estimating these parameters through the strain curves measured with a tire instrumented with strain gauges, denoted as Strain-based Intelligent Tire. These parameters are required to characterize the loss of grip in the tire–road interaction. Nonetheless, the time and computer resources required for estimating the level of adherence is not compatible with the need of current active control systems, and the instant data retrieval about the tire–road surface. The objective of this paper is to present a novel methodology in order to develop an Automatic Full Slip Detection System implemented on the Strain-based Intelligent Tire, while the specific developments for real time will be studied in the further steps of this research. This system operates with two conditions or states in the tire, namely, full sliding situation or non-full sliding situation. The inputs required to provide the tire condition are the strain curves measured when the tire is rolling. Therefore, the algorithms implemented in order to estimate the limit of adherence are presented. To delimit the states, the technique Support Vector Machines (SVM) is used to generate a separation hyperplane between these states. Support Vector Machines (SVM) is one of the most widely used supervised learning algorithms in the area of image recognition and, until now, had not been implemented in the automatic recognition of tire full slip detection. [Display omitted] • The tire as an active sensor able to detect its full slip working condition. • New capabilities to the concept of the Strain-based Intelligent Tire. • Monitoring the tire limit grip in the tire–road interaction. • Implementation of the classification algorithm Support Vectors Machines. • Development of the Automatic Full Slip Detection System. [ABSTRACT FROM AUTHOR]

This paper surveys main principles of feature selection and their recent applications in big data bioinformatics. Instead of the commonly used categorization into filter, wrapper, and embedded approaches to feature selection, we formulate feature selection as a combinatorial optimization or search problem and categorize feature selection methods into exhaustive search, heuristic search, and hybrid methods, where heuristic search methods may further be categorized into those with or without data-distilled feature ranking measures. [ABSTRACT FROM AUTHOR]

Identification of splice sites is important due to their key role in predicting the exon-intron structure of protein coding genes. Though several approaches have been developed for the prediction of splice sites, further improvement in the prediction accuracy will help predict gene structure more accurately. This paper presents a computational approach for prediction of donor splice sites with higher accuracy. In this approach, true and false splice sites were first encoded into numeric vectors and then used as input in artificial neural network (ANN), support vector machine (SVM) and random forest (RF) for prediction. ANN and SVM were found to perform equally and better than RF, while tested on HS3D and NN269 datasets. Further, the performance of ANN, SVM and RF were analyzed by using an independent test set of 50 genes and found that the prediction accuracy of ANN was higher than that of SVM and RF. All the predictors achieved higher accuracy while compared with the existing methods like NNsplice , MEM , MDD , WMM , MM1 , FSPLICE , GeneID and ASSP , using the independent test set. We have also developed an online prediction server (PreDOSS) available at http://cabgrid.res.in:8080/predoss , for prediction of donor splice sites using the proposed approach. [ABSTRACT FROM AUTHOR]

Accurate estimation of the tire slip ratio is critical for vehicle safety, as it is necessary for vehicle control purposes. In this paper, an intelligent tire system is presented to develop a novel slip ratio estimation model using machine learning (ML) algorithms. The accelerations, generated by a triaxial accelerometer installed onto the inner liner of the tire, are varied when the tire rotates to update the contact patch. Meanwhile, the reference value of slip ratio can be measured by the MTS Flat-Trac tire test platform. Then, by analyzing the variation between the accelerations and slip ratio, highly useful features are discovered, which are especially promising for assessing vertical acceleration. For these features, ML algorithms are trained to build the slip ratio estimation model, in which the ML algorithms include artificial neural networks (ANNs), gradient boosting machines (GBMs), random forests (RFs), and support vector machines (SVMs). Finally, the estimated NRMS errors are evaluated using 10-fold cross-validation (CV). The proposed estimation model is able to estimate the slip ratio continuously and stably using only the acceleration from the intelligent tire system, and the estimated slip ratio range can reach 30%. The estimation results have high robustness to vehicle velocity and load, where the best NRMS errors can reach 4.88%. In summary, the present study with the fusion of an intelligent tire system and machine learning paves the way for the accurate estimation of the tire slip ratio under different driving conditions, which creates new opportunities for autonomous vehicles, intelligent tires, and tire slip ratio estimation. • An intelligent tire system is presented to develop a novel slip ratio estimation method. • Important features are identified by analyzing the correlation between the accelerations and slip ratio. • Different machine learning algorithms are trained to build the slip ratio estimation models. • The proposed method can accurately estimate the slip ratio range up to ± 30% under different working conditions. [ABSTRACT FROM AUTHOR]

This paper presents four kernel-based algorithms for damage detection under varying operational and environmental conditions, namely based on one-class support vector machine, support vector data description, kernel principal component analysis and greedy kernel principal component analysis. Acceleration time-series from an array of accelerometers were obtained from a laboratory structure and used for performance comparison. The main contribution of this study is the applicability of the proposed algorithms for damage detection as well as the comparison of the classification performance between these algorithms and other four ones already considered as reliable approaches in the literature. All proposed algorithms revealed to have better classification performance than the previous ones. [ABSTRACT FROM AUTHOR]

The pain intensity detection approach proposed in this paper is based on the fact that facial features get deformed during pain. To model facial feature deformations, Thin Plate Spline is adopted that separates rigid and non-rigid deformations very well. For efficient pain level detection, we have mapped the deformation parameters to higher discriminative space using Distance Metric Learning (DML) method. In DML, we seek a common distance metric such that the features belonging to the same pain intensity are pulled close to each other and the features belonging to the different pain intensity are pushed as far as possible. The assessment of the proposed approach is carried out on the popularly accepted UNBC-McMaster Shoulder Pain Expression Archive Database by using Support Vector Machine as a classifier. To prove the efficacy of the proposed approach, it is compared with state-of-the-art approaches mentioned in literature. [ABSTRACT FROM AUTHOR]

Exemplar SVMs (E-SVMs, Malisiewicz et al., ICCV 2011), where an SVM is trained with only a single positive sample, have found applications in the areas of object detection and content-based image retrieval (CBIR), amongst others. In this paper we introduce a method of part based transfer regularization that boosts the performance of E-SVMs, with a negligible additional cost. This enhanced E-SVM (EE-SVM) improves the generalization ability of E-SVMs by softly forcing it to be constructed from existing classifier parts cropped from previously learned classifiers. In CBIR applications, where the aim is to retrieve instances of the same object class in a similar pose, the EE-SVM is able to tolerate increased levels of intra-class variation, including occlusions and truncations, over E-SVM, and thereby increases precision and recall. In addition to transferring parts, we introduce a method for transferring the statistics between the partsand also show that there is an equivalence between transfer regularization and feature augmentation for this problem and others, with the consequence that the new objective function can be optimized using standard libraries. EE-SVM is evaluated both quantitatively and qualitatively on the PASCAL VOC 2007 and ImageNet datasets for pose specific object retrieval. It achieves a significant performance improvement over E-SVMs, with greater suppression of negative detections and increased recall, whilst maintaining the same ease of training and testing. [ABSTRACT FROM AUTHOR]

Different species of bolete have different nutritional and medicinal value, which leads to the phenomenon of shoddy in the market from time to time. Therefore, consumers need a fast and effective detection method to identify their species. In this paper, different data pretreatment was carried out for the Fourier transform near infrared (FT-NIR) spectra, and the modeling results of partial least squares discrimination analysis (PLS-DA), support vector machines (SVM) and residual neural network (ResNet) were compared. The results show that PLS-DA and SVM models need a suitable combination of pretreatment for spectral data. The purpose is to improve the accuracy of the model and avoid over fitting. After spectral pretreatment, the accuracy of PLS-DA model were improved to 99.63% and 97.38% respectively. In order to ensure that the SVM model does not have the risk of over fitting, the accuracy of the SVM model after pretreatment were reduced to 98.5% and 93.63%. The ResNet model was established based on the original spectrum. The accuracy of the model was 100%, and there is no over fitting phenomenon, which is one of the advantages of the model. Comparing the above three models, ResNet is the best model for bolete species identification. • The spectral modeling results of different pretreatment were compared. • ResNet model was used to accurately identify 801 fruiting bodies of bolete. • Synchronous 2DCOS spectral model has good discrimination performance. • ResNet model has good performance and does not have the problem of over fitting. [ABSTRACT FROM AUTHOR]

A few decades ago, the classification of ships by means of their acoustics signature depended on the trained ears of SONAR system operators. With the advances of machine and deep learning techniques, new methods of classification were proposed in order to assist the operator to identify and classify the acoustic event. In this scenario, this paper presents a passive SONAR classification method based on Cyclostationary Analysis and Convolutional Neural Networks. The Spectral Coherence of the ship-radiated noise was extract, and served as input for a modified MobileNetV2 Convolutional Neural Network classifier. The method was benchmarked in two different dataset and was able to discriminate different classes. In parallel, another three classifiers based on Support Vector Machine were developed. Comparison with previous articles that used the same database was made, as well as, an evaluation of noise immunity was conducted. It was observed that the proposed method presented higher accuracy, even for low signal-to-noise ratios. • A ship classification method based on CNN and Cyclostationarity Analysis is proposed. • It uses the Spectral Coherence as a feature extraction for ship-radiated noise. • The MobileNetV2 CNN is used to classify vessels through the Spectral Coherence graph. • An evaluation of noise immunity of the classifier's accuracy is conduct. • The developed methodology is tested on two datasets obtaining good performance. [ABSTRACT FROM AUTHOR]

The ranking problem has received increasing attention in both the statistical and machine learning literature. This paper considers support vector machines for ranking. Under some mild conditions, a learning rate is established. [ABSTRACT FROM AUTHOR]

• A hybrid detection method is proposed for assessing delamination in composites. • Two-step method has combined the global frequency and local FBGs for damage detection. • FBGs were distributed near the damage area predicted by frequency index to further predict damage with better accuracy. • Support vector machine and the extreme learning machine were used as inverse algorithms. In this paper, a combined global-local strategy was developed for structural health monitoring (SHM) of fiber reinforced polymer (FRP) composites to assess the internal delamination damage, which has combined vibrational frequency shifts as the global index and the fiber Bragg grating (FBG) wavelength shifts as the local index. The delamination detection was carried out in two steps. The first step was based on the global damage index, which is the changes in multiple modes of frequencies measured by a non-contact scanning laser Doppler vibrometer (SLDV) system. Machine learning (ML) algorithms including support vector machine (SVM) and extreme learning machine (ELM) algorithms were used to initially predict the delamination interface, location, and size through frequency shifts. Numerical and experimental verification results show that SVM has a higher prediction accuracy than ELM when only a small number of samples exist, and the classification of SVM is better than its regression function to be used in the prediction of discrete delamination interfaces. To verify the damaged area predicted by frequency changes and further update the delamination edges with better accuracy, the second step has taken the changes in the wavelength of multiple FBG sensors as the local damage index. Multistage loads were applied at equidistant positions of the FRP specimens to induce deformations and the wavelength shifts in FBGs were used to determine the boundary of delamination. The results showed that delamination in FRP composites can be assessed more precisely using the global-local two-step SHM strategy, which can compensate for the shortcomings of only using vibration-based or FBG-based monitoring techniques. Besides, the SVM algorithm showed excellent predictive performance in both steps and had great potential in delamination damage monitoring. [ABSTRACT FROM AUTHOR]

• Recognition of local cortical folding patterns is hard due to their high variability. • Here, three new methods are proposed for their automatic classification. • They are tested on the Anterior Cingulate Cortex patterns and the Power Button Sign. • The CNN-based model is faster to use and has similar performance to the other models. • The two other models seem more effective in managing unbalanced problems. The study of local cortical folding patterns showed links with psychiatric illnesses as well as cognitive functions. Despite the tools now available to visualize cortical folds in 3D, manually classifying local sulcal patterns is a time-consuming and tedious task. In fact, 3D visualization of folds helps experts to identify different sulcal patterns but fold variability is so high that the distinction between these patterns sometimes requires the definition of complex criteria, making manual classification difficult and not reliable. However, the assessment of the impact of these patterns on the functional organization of the cortex could benefit from the study of large databases, especially when studying rare patterns. In this paper, several algorithms for the automatic classification of fold patterns are proposed to allow morphological studies to be extended and confirmed on such large databases. Three methods are proposed, the first based on a Support Vector Machine (SVM) classifier, the second on the Scoring by Non-local Image Patch Estimator (SNIPE) approach and the third based on a 3D Convolution Neural Network (CNN). These methods are generic enough to be applicable to a wide range of folding patterns. They are tested on two types of patterns for which there is currently no method to automatically identify them: the Anterior Cingulate Cortex (ACC) patterns and the Power Button Sign (PBS). The two ACC patterns are almost equally present whereas PBS is a particularly rare pattern in the general population. The three models proposed achieve balanced accuracies of approximately 80% for ACC patterns classification and 60% for PBS classification. The CNN-based model is more interesting for the classification of ACC patterns thanks to its rapid execution. However, SVM and SNIPE-based models are more effective in managing unbalanced problems such as PBS recognition. [ABSTRACT FROM AUTHOR]

• Supervised learning is used for chatter detection to avoid the threshold selection. • Combination of features by manually selected and extracted by SDAE are utilized. • Adaboost algorithm is utilized to obtain a strong classifier by a series of SVM weak classifiers. • Adaboost-SVM is improved to mitigate negative effects of wrong labels on classification accuracy. Unstable chatter vibration in the milling process significantly affect the machining quality and efficiency. In order to suppress or avoid the chatter vibration in the cutting operation, detection of chatter onset is highly needed. Until now, most of the existing chatter detection methods designed chatter indicators by extracting signal features, and the threshold of designed chatter indicator is usually needed, which is difficult to determine and might not be applicable in different cutting conditions. In fact, chatter detection is essentially a typical classification problem, hence milling chatter detection based on machine learning method is presented in this paper. In order to obtain the needed data set, milling experiments under different cutting conditions were performed. Multi-features are utilized for the chatter detection, including the dimensionless features in time domain and frequency domain, and the automatic features extracted by stacked-denoising autoencoder (SDAE). In order to improve the accuracy of chatter classification and avoid the negative effects of possible samples with wrong labels, adaptive boosting (Adaboost) algorithm that consists of a series of weak classifiers by support vector machine (SVM) is utilized and further improved. Experimental verification and performance analysis are also performed, and the results show that the presented method can detect the chatter with a high accuracy and is applicable in different milling conditions. [ABSTRACT FROM AUTHOR]

As one of the most important post-transcriptional modifications, the N7-methylguanosine (m7G) plays a key role in many RNA processing events. The accurate identification of m7G is crucial for elucidating its biological significance and future application in the medical field. In this study, a machine learning-based model was developed for the prediction of internal m7G sites, and five different feature extraction methods (Pseudo dinucleotide composition, Pseudo k-tuple composition, K monomeric units, Ksnpf frequency, and Nucleotide chemical property) were used in the feature extraction. The Random Forest algorithm was used to find the optimized feature subset and the SVM-based predictor achieved the best performance by taking the top 240 features for model training. With different performance assessment methods, 10-fold cross validation, Jackknife test, and independent test, m7GPredictor achieved competitive performance compared with the state-of-the-art predictor iRNA-m7G. The predictor developed in this study can offer useful information for elucidating the mechanism of internal m7G sites and related experimental validations. The dataset used in this study and the source code of m7Gpredictor is all available at https://github.com/NWAFU-LiuLab/m7Gpredictor. Image 1 • Five effective feature encoding methods were used in this paper. • A machine learning model was developed to predict internal m7G sites. • Our method is better than the state-of-the art predictor iRNA-m7G. [ABSTRACT FROM AUTHOR]

• Variational mode decomposition is utilized to process vibration signals. • Multi-scale dimensionless indicators are constructed as the fault features. • Fisher criterion is applied for feature evaluation and selection. • A random forest based fault classification method is proposed. • Experiments on centrifugal multi-level impeller blower validate the method. Fault diagnosis methods based on dimensionless indicators have long been studied for rotating machinery. However, traditional dimensionless indicators frequently suffer a low accuracy of fault diagnosis for nonlinear and non-stationary dynamic signals of rotating machinery. In this paper, we propose an effective fault diagnosis method based on multi-scale dimensionless indicator (MSDI) and random forests. In the proposed method, the real-time vibration signals are first processed by the variational mode decomposition and then six types of MSDI are constructed based on the decomposed signals. Through utilizing the Fisher criterion, several top ranked MSDIs are selected as fault features. Based on the selected MSDIs, the random forests model is applied to determine fault types. To verify the superiority of the proposed method, several experiments on fault diagnosis are conducted on a centrifugal multi-level impeller blower. The results demonstrate that the proposed method can successfully identify different fault types and the average accuracy can reach 95.58%. In contrast with traditional dimensionless indicators based methods, the proposed method can improve the fault diagnosis accuracy by 7.25% and outperforms other techniques such as back propagation neural network, support vector machine and extreme learning machine. These results indicate that the MSDI can effectively solve the deficiency of the traditional dimensionless indicator, and has stronger distinguishing ability for the fault types. [ABSTRACT FROM AUTHOR]

• A low-cost sensor system was used to differentiate organic apples from conventional ones. • Ten machine learning algorithms were evaluated using rainbow image data from the sensor system. • The classification results of rainbow image data were comparable to that of spectral data. Optical measuring technologies coupled with machine learning algorithms can be used to build a home-made sensor system. We built such a sensor system using a smartphone and a diffraction grating sheet. Diffraction images were captured under white light illumination and converted into a data matrix for data analysis. In this paper we present a systematic evaluation of this sensor system on the task of differentiating organic apples from conventional ones. We used the sensor system to measure 150 organic and conventional apples as rainbow images. We processed the rainbow images using computer vision techniques, built machine learning and chemometrics models, and used the resultant models to classify testing samples. Moreover, a comparative study was conducted where the same set of apples were scanned by a commercial spectrometer resulting in spectral data of the apple samples and classification was undertaken using partial least squares discriminant analysis (PLS-DA). Experimental results show that state of the art machine learning algorithms such as support vector machine (SVM) and locally weighted partial least squares classifier (LW-PLSC) are effective in handling low-quality image data with classification accuracies of 93 − 100%. These results suggest that the sensor system is convenient and low-cost, and provides a fast, effective, non-destructive and viable solution for in-line food authentication. [ABSTRACT FROM AUTHOR]

Earth fissures are the cracks on the surface of the earth mainly formed in the arid and the semi-arid basins. The excessive withdrawal of groundwater, as well as the other underground natural resources, has been introduced as the significant causing of land subsidence and potentially, the earth fissuring. Fissuring is rapidly turning into the nations' major disasters which are responsible for significant economic, social, and environmental damages with devastating consequences. Modeling the earth fissure hazard is particularly important for identifying the vulnerable groundwater areas for the informed water management, and effectively enforce the groundwater recharge policies toward the sustainable conservation plans to preserve existing groundwater resources. Modeling the formation of earth fissures and ultimately prediction of the hazardous areas has been greatly challenged due to the complexity, and the multidisciplinary involved to predict the earth fissures. This paper aims at proposing novel machine learning models for prediction of earth fissuring hazards. The Simulated annealing feature selection (SAFS) method was applied to identify key features, and the generalized linear model (GLM), multivariate adaptive regression splines (MARS), classification and regression tree (CART), random forest (RF), and support vector machine (SVM) have been used for the first time to build the prediction models. Results indicated that all the models had good accuracy (>86%) and precision (>81%) in the prediction of the earth fissure hazard. The GLM model (as a linear model) had the lowest performance, while the RF model was the best model in the modeling process. Sensitivity analysis indicated that the hazardous class in the study area was mainly related to low elevations with characteristics of high groundwater withdrawal, drop in groundwater level, high well density, high road density, low precipitation, and Quaternary sediments distribution. • Machine learning (ML) prediction of earth fissure hazard. • Key features selection using the simulated annealing (SA) method. • Good performance of the ML models (Accuracy >86%; Precision >81%). • The worst and best models respectively were GLM (as a linear model) and RF. [ABSTRACT FROM AUTHOR]