Search

Cardiotocography (CTG) is a clinical procedure that is used to track and gauge the severity of fetal distress. Although CTG is the most often used equipment to monitor and assess the health of the fetus, the high rate of false positive results due to visual interpretation significantly contributes to needless surgical delivery or delayed intervention. In this study, a novel approach is introduced where both printing CTG paper is digitized and a machine learning approach is employed to detect the abnormality in the digitized CTG signal. Image processing-based preprocessing steps are employed to make the printing of CTG paper more convenient to extract the CTG signal. Various signal-processing techniques are used to calibrate the extracted CTG signal. Then, Empirical Mode Decomposition (EMD) is used to decompose the CTG signal into its frequency components and instantaneous frequency and spectral entropy features are extracted. After feature normalization and feature selection with ReliefF algorithm, support vector machines (SVM) is used for the classification of the normal and abnormal classes. A novel dataset is used in the experimental works and various performance evaluation metrics are used for the evaluation of the achievement of the proposed method. 10-fold cross-validation-based experiments show that the proposed method is quite efficient in abnormality detection in printing CTG papers where an average accuracy score of around 90.0% is produced.

Traditional algorithm lacks the ability to generalize across different application scenarios and how to set the stopping conditions of network iteration in Stochastic Configuration Network (SCN) is a problem. This paper proposes an event recognition system based on minimum redundancy, maximum relevance (mRMR) and a modified Stochastic Configuration Network (CWCs-SCN) for the fiber Bragg grating (FBG) sensor system. mRMR is used to analyze the importance of 14th time-frequency domain features in different application scenarios. After that, a modified SCN (CWCs-SCN) neural network is used to solve the stopping condition of the network iteration. The experiments demonstrated that the average recognition accuracy of four kind of human intrusion signal can reach to 96.7%, which is better than others traditional methods. In addition, a more compact neural network is obtained compared to SCN. To verify the generalization ability of the proposed model, it is applied to gear fault diagnosis with an accuracy of 96.3%. This imply that the proposed method in this paper has a wide application prospect. [ABSTRACT FROM AUTHOR]

Vehicle classification based on magnetic sensors can be effectively applied to intelligent transportation and realize intelligent management of traffic. The use of representative vehicle signal features is a prerequisite and guarantee for accurate vehicle classification. This paper uses a single 3-axis magnetic sensor to acquire vehicle signals and extract a large number of features from the vehicle signals. In order to obtain a set of features that are simple and without loss of classification accuracy, we propose a Filtering algorithm based on Feature Pairing Elimination (FPE-Filter). In addition, choosing the right classifier is also crucial for models with high classification accuracy. In this paper, we compare four common classification models: SVM, RF, KNN, and C4.5. The experimental results show that the SVM performs best and the classification accuracy reaches 95%. [ABSTRACT FROM AUTHOR]

Objective: Resting-state functional magnetic resonance imaging (rs-fMRI)-derived functional connectivity (FC) patterns have been extensively used to delineate global functional organization of the human brain in healthy development and neuropsychiatric disorders. In this paper, we investigate how FC in males and females differs in an age prediction framework. Methods: We first estimate FC between regions-of-interest (ROIs) using distance correlation instead of Pearson’s correlation. Distance correlation, as a multivariate statistical method, explores spatial relations of voxel-wise time courses within individual ROIs and measures both linear and nonlinear dependence, capturing more complex between-ROI interactions. Then, we propose a novel non-convex multi-task learning (NC-MTL) model to study age-related gender differences in FC, where age prediction for each gender group is viewed as one task, and a composite regularizer with a combination of the non-convex $\ell _{2,1-2}$ and $\ell _{1-2}$ terms is introduced for selecting both common and task-specific features. Results and Conclusion: We validate the effectiveness of our NC-MTL model with distance correlation-based FC derived from rs-fMRI for predicting ages of both genders. The experimental results on the Philadelphia Neurodevelopmental Cohort demonstrate that our NC-MTL model outperforms several other competing MTL models in age prediction. We also compare the age prediction performance of our NC-MTL model using FC estimated by Pearson’s correlation and distance correlation, which shows that distance correlation-based FC is more discriminative for age prediction than Pearson’s correlation-based FC. Significance: This paper presents a novel framework for functional connectome developmental studies, characterizing developmental gender differences in FC patterns. [ABSTRACT FROM AUTHOR]

To monitor process and identify the deviation as early as possible, data-driven methods have been applied for process monitoring and fault detection in semiconductor manufacturing. Although various fault detection and classification models had been discussed in the literature, however, little research has been devoted to feature selection from trace data that is important for process monitoring of natural variation. Additionally, the high-mix production mode with different recipes leads to process dynamic of wafer-to-wafer (W2W) variation which should also be identified for safeguarding false alarms and serving as a warning indicator. Therefore, this paper proposes a data-driven framework to identify the key features with respect to the W2W variation. In particular, the self-organizing map is used to annotate the grade of wafer variation among the in-line metrology data. Subsequently, the adaptive boosting (AdaBoost) is adopted to examine the effectiveness of every feature and its processing times, respectively. To validate the proposed framework, an empirical study from a semiconductor fabrication plant is conducted. The experimental results demonstrate that the key feature identification is of critical importance to build highly capable models for process monitoring. Through the dimensionality reduction technique, it has been illustrated that a smaller set of the identified key features are able to pinpoint the W2W variation of different wafer grades more clearly than the whole set of process features. Note to Practitioners— Process monitoring has become more difficult with the shrinking linewidth in semiconductor manufacturing. The challenges of analyzing equipment sensor or raw trace data for process monitoring in high-mix manufacturing processes are to incorporate subject-matter expert knowledge for setting control limit meticulously, to detect the subtle changes by analyzing the whole trace data profile, and to identify W2W variation for reducing false alarms. This paper proposes a data-driven framework for process monitoring by adopting data-driven approaches without recourse to domain judgement. Experimental results demonstrate that the proposed data-driven framework can effectively identify the key features via sensor readings and corresponding processing times, respectively. The engineers can make use of the extracted features to perform a predictive monitoring on metrology data for detection of potential process deterioration. [ABSTRACT FROM AUTHOR]

Unsupervised spectral feature selection (USFS) methods could output interpretable and discriminative results by embedding a Laplacian regularizer in the framework of sparse feature selection to keep the local similarity of the training samples. To do this, USFS methods usually construct the Laplacian matrix using either a general-graph or a hyper-graph on the original data. Usually, a general-graph could measure the relationship between two samples while a hyper-graph could measure the relationship among no less than two samples. Obviously, the general-graph is a special case of the hyper-graph and the hyper-graph may capture more complex structure of samples than the general graph. However, in previous USFS methods, the construction of the Laplacian matrix is separated from the process of feature selection. Moreover, the original data usually contain noise. Each of them makes difficult to output reliable feature selection models. In this paper, we propose a novel feature selection method by dynamically constructing a hyper-graph based Laplacian matrix in the framework of sparse feature selection. Experimental results on real datasets showed that our proposed method outperformed the state-of-the-art methods in terms of both clustering and segmentation tasks. [ABSTRACT FROM AUTHOR]

Artificial intelligence evolved as a powerful tool in condition monitoring of the induction motor for early diagnosis of bearing faults. This paper attempted to identify the features to improve the accuracy of diagnosis using the benchmark database of the Case Western Reserve University (CWRU) and the Machinery Fault Prevention Technology (MFPT). In this paper, 18 time-domain features are extracted constituting six proposed time-dependent spectral features (TDSFs) and 12 statistical time-domain features. For the first time, a set of six TDSFs are extracted to diagnose the bearing faults. This involves extracting frequency-domain features using the relationship between Parseval’s theorem and the Fourier transform directly from the time domain. Particle swarm optimization (PSO) and wheel-based differential evolution (WBDE) feature selection algorithms are also implemented to identify four prominent features. It is found that three of the four selected features are TDSF features, and results of selected features revealed the attainment of 90.92% for 48 class identifications of CWRU database and 100% for 17 class identifications of the MFPT database, respectively. [ABSTRACT FROM AUTHOR]

In the last two decades, a variety of different types of multiobjective optimization problems (MOPs) have been extensively investigated in the evolutionary computation community. However, most existing evolutionary algorithms encounter difficulties in dealing with MOPs whose Pareto optimal solutions are sparse (i.e., most decision variables of the optimal solutions are zero), especially when the number of decision variables is large. Such large-scale sparse MOPs exist in a wide range of applications, for example, feature selection that aims to find a small subset of features from a large number of candidate features, or structure optimization of neural networks whose connections are sparse to alleviate overfitting. This paper proposes an evolutionary algorithm for solving large-scale sparse MOPs. The proposed algorithm suggests a new population initialization strategy and genetic operators by taking the sparse nature of the Pareto optimal solutions into consideration, to ensure the sparsity of the generated solutions. Moreover, this paper also designs a test suite to assess the performance of the proposed algorithm for large-scale sparse MOPs. The experimental results on the proposed test suite and four application examples demonstrate the superiority of the proposed algorithm over seven existing algorithms in solving large-scale sparse MOPs. [ABSTRACT FROM AUTHOR]

Multivariate time series (MTS) arise due to multisensor data collection in manufacturing. These data are complex in the sense that attributes have a varying scale, volitivity, continuity, and so on, and interattribute dependence also appears, which can mask the inherent information about system health status. Conventional machine learning-based process monitoring techniques are inefficient in predicting faults with MTS—their detection capability heavily relies on the input features, yet the classification power of MTS attributes is weakened by the complexity of multisensor data. Effective feature extraction is, therefore, necessary to facilitate fault prediction with MTS. This study proposes a fault prediction framework for MTS based on bag-of-words (BOW) feature extraction, statistical feature selection, and classification analysis. BOW models are for the first time adopted in a manufacturing context. Their superior capability in information preservation, local pattern recognition, and temporal effect accommodation has overcome the major limitations in current manufacturing practices with MTS. A comprehensive case study demonstrates the desirable performance of this framework on two MTS data sets from paper manufacturing and automotive manufacturing, as well as its superiority over conventional machine learning-based fault prediction. Note to Practitioners—Process monitoring in a multisensor environment has been a vital interest in manufacturing. The difficulty lies in the lack of detection power in many conventional techniques, e.g., control chart and logistic regression. A critical reason for such failure is the neglect of time effect in MTS—patterns associated with system faults tend to stretch a period, but a conventional control chart or classifier inspects each time stamp separately. Fault detection based on time series sequences is, therefore, essential. However, how to effectively extract features from MTS becomes a challenge. The framework proposed in this study adopts BOW models, specifically symbolic aggregate approximation (SAX), to extract features from MTS sequences, thus substantially improves the detection power against local patterns over time. Many manufacturing multisensor data are subject to such local patterns that point to the root cause of fault, so the proposed framework has a wide application in manufacturing. [ABSTRACT FROM AUTHOR]

Although ultra wideband (UWB) positioning is considered as one of the most promising solutions for indoor positioning due to its high positioning accuracy, the accuracy in situations with a large number of users will be reduced because the time between two UWB position updates can be very high. To obtain a position estimate in between these updates, we can combine the UWB positioning with a different technology, e.g., an inertial measurement unit (IMU) that captures data from an accelerometer, gyroscope, and magnetometer. Previous research using the IMU outputs for location-based services employs the periodic behaviour of the accelerometer signal to count steps. However, most of these algorithms require extensive manual tuning of multiple parameters to obtain satisfactory accuracy. To overcome these practical issues, step counting algorithms using machine learning (ML) principles can be developed. In this paper, we consider accelerometer-based step counters using ML. As the performance and complexity of such algorithms depend on the features used in the training and inference phase, proper selection of the employed features is important. Therefore, in this paper, we propose a novel feature selection algorithm, where we first rank the features based on their Bhattacharyya coefficients and then systematically construct a subset of these ranked features. In this paper, we compare three ranking approaches and apply our algorithm to different ML algorithms employing an experimental set. Although the performance of the evaluated combinations slightly varies for different ML algorithms, their performance is comparable to state-of-the-art step counters, without the need to tune parameters manually. [ABSTRACT FROM PUBLISHER]

Knowledge distillation (KD) is a popular method to train efficient networks (“student”) with the help of high-capacity networks (“teacher”). Traditional methods use the teacher’s soft logits as extra supervision to train the student network. In this paper, we argue that it is more advantageous to make the student mimic the teacher’s features in the penultimate layer. Not only the student can directly learn more effective information from the teacher feature, feature mimicking can also be applied for teachers trained without a softmax layer. Experiments show that it can achieve higher accuracy than traditional KD. To further facilitate feature mimicking, we decompose a feature vector into the magnitude and the direction. We argue that the teacher should give more freedom to the student feature’s magnitude, and let the student pay more attention on mimicking the feature direction. To meet this requirement, we propose a loss term based on locality-sensitive hashing (LSH). With the help of this new loss, our method indeed mimics feature directions more accurately, relaxes constraints on feature magnitudes, and achieves state-of-the-art distillation accuracy. We provide theoretical analyses of how LSH facilitates feature direction mimicking, and further extend feature mimicking to multi-label recognition and object detection. [ABSTRACT FROM AUTHOR]

Domain adaptation aims at extracting knowledge from auxiliary source domains to assist the learning task in a target domain. In classification problems, since the distributions of the source and target domains are different, directly using source data to build a classifier for the target domain may hamper the classification performance on the target data. Fortunately, in many tasks, there can be some features that are transferable, i.e., the source and target domains share similar properties. On the other hand, it is common that the source data contain noisy features which may degrade the learning performance in the target domain. This issue, however, is barely studied in existing works. In this paper, we propose to find a feature subset that is transferable across the source and target domains. As a result, the domain discrepancy measured on the selected features can be reduced. Moreover, we seek to find the most discriminative features for classification. To achieve the above goals, we formulate a new sparse learning model that is able to jointly reduce the domain discrepancy and select informative features for classification. We develop two optimization algorithms to address the derived learning problem. Extensive experiments on real-world data sets demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper, classification of mental task-root brain–computer interfaces (BCIs) is being investigated. The mental tasks are dominant area of investigations in BCI, which utmost interest as these system can be augmented life of people having severe disabilities. The performance of BCI model primarily depends on the construction of features from brain, electroencephalography (EEG), signal, and the size of feature vector, which are obtained through multiple channels. The availability of training samples to features are minimal for mental task classification. The feature selection is used to increase the ratio for the mental task classification by getting rid of irrelevant and superfluous features. This paper suggests an approach to augment the performance of a learning algorithm for the mental task classification on the utility of power spectral density (PSD) using feature selection. This paper also deals a comparative analysis of multivariate and univariate feature selection for mental task classification. After applying the above stated method, the findings demonstrate substantial improvements in the performance of learning model for mental task classification. Moreover, the efficacy of the proposed approach is endorsed by carrying out a robust ranking algorithm and Friedman’s statistical test for finding the best combinations and compare various combinations of PSD and feature selection methods. [ABSTRACT FROM AUTHOR]

With the widespread deployment of phasor measurement units (PMUs), synchronized measurements of the power system has opened opportunities for data-driven short-term voltage stability (STVS) assessment. The existing intelligent system-based methods for data-driven stability assessment assume full and complete data input is always available. However, in practice, after a fault occurs in the system, some PMU data may not be fully available due to PMU loss and/or fault-induced topology change, which deteriorates the stability assessment performance. To address this issue, this paper proposes a missing-data tolerant method for post-fault STVS assessment. The buses in the system are strategically grouped to maintain a high level of grid observability for the stability assessment model under any PMU loss and/or topology change scenario, and a structure-adaptive ensemble learning model is designed to adapt its structure to only use available feature inputs for real-time STVS assessment. By marked contrast to existing methods, the proposed method demonstrates much stronger missing-data tolerance and can maintain a high STVS assessment accuracy even when a large portion of measurements are missing. [ABSTRACT FROM AUTHOR]

As a popular method for image and video retargeting, seam carving has been used for image/video forgery manipulation. Although significant advances have been made in detecting seam-carving forgery, there are very few contributions in exposing the forgery from recompressed JPEG images, especially the doctored images that are recompressed at the same or a lower quality. The detection is generally challenging because the recompression after tampering compromises the existing forgery traces. Aiming to address this problem, we propose a hybrid large-feature mining-based approach that contains multiple types of large features. Ensemble learning is used to deal with the high-feature dimensionality. This paper shows that the proposed approach effectively distinguishes the seam-carved JPEG images from untouched JPEG images and improves the detection accuracy. In our proposed multiple types of features, directional derivative-based feature set and Gabor residual-based feature set generally perform the best. This paper also indicates that feature selection may play an important role to greatly reduce the feature number while maintaining a better or comparable detection accuracy. [ABSTRACT FROM AUTHOR]

The camera provides rich visual information regarding objects and becomes one of the most mainstream sensors in the automation community. However, it is often difficult to be applicable when the objects are not visually distinguished. On the other hand, tactile sensors can be used to capture multiple object properties, such as textures, roughness, spatial features, compliance, and friction, and therefore provide another important modality for the perception. Nevertheless, effective combination of the visual and tactile modalities is still a challenging problem. In this paper, we develop a visual–tactile fusion framework for object recognition tasks. This paper uses the multivariate-time-series model to represent the tactile sequence and the covariance descriptor to characterize the image. Further, we design a joint group kernel sparse coding (JGKSC) method to tackle the intrinsically weak pairing problem in visual–tactile data samples. Finally, we develop a visual–tactile data set, composed of 18 household objects for validation. The experimental results show that considering both visual and tactile inputs is beneficial and the proposed method indeed provides an effective strategy for fusion. [ABSTRACT FROM AUTHOR]

Feature selection is an optional preprocessing procedure and is frequently used to improve the classification accuracy of a machine learning algorithm by removing irrelevant and/or redundant features. However, in many real-world applications, the test cost is also required for making optimal decisions, in addition to the classification accuracy. To the best of our knowledge, thus far, few studies have been conducted on test-cost-sensitive feature selection (TCSFS). In TCSFS, the objectives are twofold: 1) to improve the classification accuracy and 2) to decrease the test cost. Therefore, in fact, it constitutes a multiobjective optimization problem. In this paper, we transformed this multiobjective optimization problem into a single-objective optimization problem by utilizing a new evaluation function and in this paper, we propose a new general wrapper framework for TCSFS. Specifically, in our proposed framework, we add a new term to the evaluation function of a wrapper feature selection method so that the test cost of measuring features is taken into account. We experimentally tested our proposed framework, using 36 classification problems from the University of California at Irvine (UCI) repository, and compared it to some other state-of-the-art feature selection frameworks. The experimental results showed that our framework allows users to select an optimal feature subset with the minimal test cost, while simultaneously maintaining a high classification accuracy. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) networks are obstructed by security vulnerabilities that hackers can leverage to operate intrusions in many environments, such as smart homes, smart factories, and smart healthcare systems. To overcome this obstruction, researchers have come up with different intrusion detection and prevention systems (IDPSs). Out of all the implemented technologies, Machine Learning (ML) has emerged as the most promising approach. Therefore, to improve the detection accuracy, most ML-based intrusion detection solutions focus only on investigating appropriate ML algorithms. Yet, the limitations in terms of detection accuracy in various attacks are often caused by lack of appropriate detection features. Moreover, the majority of the previous works lack intrusion prevention mechanisms and deployment architectures. Thus, in this research, we study the properties of different smart home security attacks and the quality of the features that can be brought out and employed in ML algorithms to detect each of these attacks efficiently. Furthermore, this research proposes effective intrusion prevention mechanisms and a Software-Defined Networking (SDN) based deployment architecture of the IDPSs within home networks. Experimental evaluations of the proposed solution are provided using different feature sets and various ML models. The contributions and advancements discussed in this paper will upgrade future research and engineering works on IDPSs for IoT. [ABSTRACT FROM AUTHOR]

One of the main challenges in multi-source domain adaptation is how to reduce the domain discrepancy between each source domain and a target domain, and then evaluate the domain relevance to determine how much knowledge should be transferred from different source domains to the target domain. However, most prior approaches barely consider both discrepancies and relevance among domains. In this paper, we propose an algorithm, called Iterative Refinement based on Feature Selection and the Wasserstein distance (IRFSW), to solve semi-supervised domain adaptation with multiple sources. Specifically, IRFSW aims to explore both the discrepancies and relevance among domains in an iterative learning procedure, which gradually refines the learning performance until the algorithm stops. In each iteration, for each source domain and the target domain, we develop a sparse model to select features in which the domain discrepancy and training loss are reduced simultaneously. Then a classifier is constructed with the selected features of the source and labeled target data. After that, we exploit optimal transport over the selected features to calculate the transferred weights. The weight values are taken as the ensemble weights to combine the learned classifiers to control the amount of knowledge transferred from source domains to the target domain. Experimental results validate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper, we focus on co-selection of instances and features in the semi-supervised learning scenario. In this context, co-selection becomes a more challenging problem as data contain labeled and unlabeled examples sampled from the same population. To carry out such semi-supervised co-selection, we propose a unified framework, called sCOs, which efficiently integrates labeled and unlabeled parts into the co-selection process. The framework is based on introducing both a sparse regularization term and a similarity preserving approach. It evaluates the usefulness of features and instances in order to select the most relevant ones, simultaneously. We propose two efficient algorithms that work for both convex and nonconvex functions. To the best of our knowledge, this paper offers, for the first time ever, a study utilizing nonconvex penalties for the co-selection of semi-supervised learning tasks. Experimental results on some known benchmark datasets are provided for validating sCOs and comparing it with some representative methods in the state-of-the art. [ABSTRACT FROM AUTHOR]

The matrix-based Rényi's α-order entropy functional was recently introduced using the normalized eigenspectrum of a Hermitian matrix of the projected data in a reproducing kernel Hilbert space (RKHS). However, the current theory in the matrix-based Rényi's α-order entropy functional only defines the entropy of a single variable or mutual information between two random variables. In information theory and machine learning communities, one is also frequently interested in multivariate information quantities, such as the multivariate joint entropy and different interactive quantities among multiple variables. In this paper, we first define the matrix-based Rényi's α-order joint entropy among multiple variables. We then show how this definition can ease the estimation of various information quantities that measure the interactions among multiple variables, such as interactive information and total correlation. We finally present an application to feature selection to show how our definition provides a simple yet powerful way to estimate a widely-acknowledged intractable quantity from data. A real example on hyperspectral image (HSI) band selection is also provided. [ABSTRACT FROM AUTHOR]

Wearable electroencephalography sensor integrated with seizure detection algorithm has become an emerging research topic in health management for patients with refractory epilepsy. To enhance seizure detection accuracy and reduce power consumption, this paper has proposed a selection method of optimal features customized for each patient. A feature pool consisting of several popular features is constructed. Then, the customized multi-domain features for each patient are selected from feature pool based on minimum redundancy maximum relevance. By removing redundant features, the computational complexity of on-line seizure detection can be greatly reduced, which is desired in wearable device where low-power and real-time operation are required. Four test datasets are used to evaluate the performance of proposed method, with detection accuracy calculated. Experiment results show that different datasets vary in optimal features. By using the proposed method, the detection accuracy and the value of area under receiver operating characteristic curve can be improved by 11.49% and 0.13 respectively, while feature dimension being reduced to 1. [ABSTRACT FROM AUTHOR]

Metrology, which plays an important role in ensuring production quality in modern manufacturing industries, incurs substantial costs both in terms of the infrastructure required and the time needed to perform measurements. In particular, in the semiconductor manufacturing industry, measuring fundamental quantities on different sites of a wafer surface is associated with increased production time. To increase metrology efficiency, a typical strategy is to limit the number of sites measured and to exploit statistical models (soft sensing) to reconstruct the wafer profile. Moreover, for quality reasons, spatial dynamic sampling strategies may be employed to ensure that all regions of a wafer surface are checked periodically during production. In this paper, we propose a new sampling strategy, called induced start dynamic sampling (ISDS), which adapts greedy feature selection algorithms to the spatial dynamic sampling problem, such that the number of measured sites at each process run is minimized while achieving good wafer profile reconstruction accuracy and process visibility. The superiority of the proposed strategy with respect to the state of the art is demonstrated using both simulated data and an industrial chemical vapor deposition case study. Note to Practitioners—In this paper, we tackle a practical metrology problem encountered in semiconductor manufacturing, namely, the design of a dynamic wafer measurement plan to monitor the accuracy of a process across the whole wafer surface. The measurement plan is called “dynamic,” since the measurement locations, which are drawn from a candidate set that provides coverage of the whole wafer, change at each process iteration. Our methodology addresses the challenge of finding an optimized tradeoff between the number of measurements performed on each wafer and the reconstruction accuracy that can be achieved for the unmeasured areas on the wafer while, at the same time, for quality assurance purposes, ensuring that all locations on a wafer are visited in a finite number of process runs. The major benefit of the methodology is that it can significantly reduce the number of sites that need to be measured on each wafer enabling greater throughput on metrology tools without sacrificing process monitoring and anomaly detection capability. [ABSTRACT FROM AUTHOR]

Microarray datasets are usually imbalanced with a huge number of features and few samples. Most of the existing methods consider classification accuracy as the performance measure while converting the imbalanced datasets into a balanced one and selecting the optimal feature subset, resulting in overfitting. To address the above-mentioned issue, in this paper, a multi-objective genetic algorithm approach is proposed that utilizes a combination of evaluation metrics for both imbalanced data processing and feature selection. The proposed methodology is implemented in seven real datasets taken from public repositories. The cross-validation results prove that the proposed methodology outperforms existing methods in all seven datasets. The proposed methodology serves as a simple and efficient method for preprocessing microarray datasets and helps clinical practitioners in making strategic decisions with the help of a few features. It will also greatly reduce the time spent on processing unnecessary features. [ABSTRACT FROM AUTHOR]

Cardiovascular diseases are the primary cause of deaths in the world. Atrial fibrillation (AF) is the most common type of cardiac arrhythmia. Due to its high prevalence and associated risks, early detection of AF is an important objective for healthcare systems worldwide. The growing demand for medical assistance implies increased expenses, which could be limited by implementing ambulatory monitoring techniques based on wearable devices, thus, reducing the number of people requiring observation in hospitals. One of the main challenges in this context is related to the large amount of data from patients to be analyzed, which points to the suitability of using computational intelligence techniques for it. The selection of the features to be extracted from data plays a key role in order for any classifier of heart rhythm to provide good results in this regard. This paper demonstrates that it is possible to achieve an accurate detection of AF using a very low number of relatively simple features extracted from photoplethysmographic signals, enabling the use of affordable wearable devices (with scarce processing and data storage resources) with this purpose over long periods of time. This fact has been validated in experiments using data from real patients under medical supervision. [ABSTRACT FROM AUTHOR]

It has been. As an automated validation method recommended by the IEEE Standard 1597.1/2, the FSV method has gained broad attention in the practice of computational electromagnetics modeling and simulations. This paper reviews the motivation, evolution, enhancements, and criticisms of the method over this time. The aims of this paper are to give a detailed contextualization of the development of the FSV method itself and to discuss the open questions and possible strategies of the next generation of automated validation methods. [ABSTRACT FROM PUBLISHER]

The extreme learning machine (ELM), which was originally proposed for “generalized” single-hidden layer feedforward neural networks, provides efficient unified learning solutions for the applications of regression and classification. Although, it provides promising performance and robustness and has been used for various applications, the single-layer architecture possibly lacks the effectiveness when applied for natural signals. In order to over come this shortcoming, the following work indicates a new architecture based on multilayer network framework. The significant contribution of this paper are as follows: 1) unlike existing multilayer ELM, in which hidden nodes are obtained randomly, in this paper all hidden layers with invertible functions are calculated by pulling the network output back and putting it into hidden layers. Thus, the feature learning is enriched by additional information, which results in better performance; 2) in contrast to the existing multilayer network methods, which are usually efficient for classification applications, the proposed architecture is implemented for dimension reduction and image reconstruction; and 3) unlike other iterative learning-based deep networks (DL), the hidden layers of the proposed method are obtained via four steps. Therefore, it has much better learning efficiency than DL. Experimental results on 33 datasets indicate that, in comparison to the other existing dimension reduction techniques, the proposed method performs competitively better with fast training speeds. [ABSTRACT FROM AUTHOR]

To realize accurate fault diagnosis of rolling bearing under random noise, a novel fault diagnosis method based on ensemble empirical mode decomposition (EEMD) and optimized Elman_AdaBoost is proposed in this paper. First, the EEMD method is used to decompose the original vibration signal into several intrinsic mode functions (IMFs). Then, the correlation coefficient and kurtosis of IMFs are used to remove any excess ingredients, and the signal is reconstructed by using the rest of IMFs. In addition, based on the analysis of the time–frequency characteristics of the reconstructed signal, the root-mean-square value and the power spectrum center are extracted as a 2-D feature vector from each reconstructed signal and regarded as the input characteristic vector of the Elman neural network. Combined with the AdaBoost classification algorithm, the Elman_AdaBoost algorithm is improved and proposed to establish the strong classifier. Finally, based on the vibration data recorded in the bearing center of the Case Western Reserve University, three kinds of bearings with different faults and normal bearings are selected as the sample data and the training samples and test samples are constructed. Experimental results show that the proposed EEMD method and optimized Elman_AdaBoost model can effectively diagnose the rolling bearing under random noise, and has the advantages of fast speed, small error, and stable performance. Comparing with the traditional Elman_AdaBoost and single Elman neural network algorithm, this paper with fewer characteristics can get better accuracy and real-time processing performance, and presents a simple and practical resolution for the fault diagnosis of rolling bearings. [ABSTRACT FROM PUBLISHER]

This paper presents a novel approach to action recognition using synthetic multi-view data from depth maps. Specifically, multiple views are first generated by rotating 3D point clouds from depth maps. A pyramid multi-view depth motion template is then adopted for multi-view action representation, characterizing the multi-scale motion and shape patterns in 3D. Empirically, despite the view-specific information, the latent information between multiple views often provides important cues for action recognition. Concentrating on this observation and motivated by the success of the dictionary learning framework, this paper proposes to explicitly learn a view-specific dictionary (called specificity) for each view, and simultaneously learn a latent dictionary (called latent correlation) across multiple views. Thus, a novel method, specificity and latent correlation learning, is put forward to learn the specificity that captures the most discriminative features of each view, and learn the latent correlation that contributes the inherent 3D information to multiple views. In this way, a compact and discriminative dictionary is constructed by specificity and latent correlation for feature representation of actions. The proposed method is evaluated on the MSR Action3D, the MSR Gesture3D, the MSR Action Pairs, and the ChaLearn multi-modal data sets, consistently achieving promising results compared with the state-of-the-art methods based on depth data. [ABSTRACT FROM AUTHOR]

The task of feature selection is to find the most representative features from the original high-dimensional data. Because of the absence of the information of class labels, selecting the appropriate features in unsupervised learning scenarios is much harder than that in supervised scenarios. In this paper, we investigate the potential of locally linear embedding (LLE), which is a popular manifold learning method, in feature selection task. It is straightforward to apply the idea of LLE to the graph-preserving feature selection framework. However, we find that this straightforward application suffers from some problems. For example, it fails when the elements in the feature are all equal; it does not enjoy the property of scaling invariance and cannot capture the change of the graph efficiently. To solve these problems, we propose a new filter-based feature selection method based on LLE in this paper, which is named as LLE score. The proposed criterion measures the difference between the local structure of each feature and that of the original data. Our experiments of classification task on two face image data sets, an object image data set, and a handwriting digits data set show that LLE score outperforms state-of-the-art methods, including data variance, Laplacian score, and sparsity score. [ABSTRACT FROM PUBLISHER]

Moving cast shadow identification and extraction are the commonly encountered problems in visual surveillance applications. In the video object detection problem, a shadow is a category of the dynamic entity which needs to be removed for accurate detection of moving objects. In this paper, we have proposed a new scheme to detect the moving object in video frames while removing the shadow. The proposed scheme is based on the notion of background modeling and model learning. The background is modeled by the proposed Spatio-Temporal Kernel Density Estimation (ST-KDE) based model, but the model learning takes place in online mode in the fused feature space. In the fusion process, the Local Binary Pattern (LBP) features of the ST-KDE model of a frame are determined and are fused probabilistically with the Gabor features of the corresponding original frame. To achieve accurate background modeling, weights in the probabilistic fusion are adaptively determined from the scene dynamics. Background model learning is pixel-based where model histograms learn the background with every temporal frame. Learning and classification of the fused feature frames happen simultaneously with frame by frame. The classified frame thus obtained contains the object with some residual shadow which is eliminated by entropy map and thresholding. The main novelties of our paper are: (i) ST-KDE based background modeling, (ii) Determination of the weights of the fusion process in an adaptive framework and, (iii) online background model learning in feature space. The proposed modeling approach has been tested successfully with different video sequences considered from CDnet, Lasiesta, ATON-CVRR, SRD, and ISTD databases. The performance metrics of the proposed scheme have been compared with many existing methods and it has been found that the proposed method exhibits improved performance. [ABSTRACT FROM AUTHOR]

Emotion analysis based on peripheral physiological signals has attracted increasing attention recently in affective computing. Previous works usually predict emotional states using a single emotion label for each discrete time. However, in real-world scenarios, it is not sufficient due to the fact that the real-world emotional state is usually a mixture of basic emotions. In this paper, we formulate the emotion analysis as an emotion distribution learning (EDL) problem and make two contributions. First, we establish a standardised dataset containing four negative emotions (anger, disgust, sadness, fear) and three positive emotions (tenderness, joy, amusement), which could be a useful benchmark for the EDL task. Second, we propose an emotion distribution prediction system which has the following distinct characteristics: (1) after processing raw peripheral physiological signals, we compute totally 89 representative features from four channels, i.e., GSR, SKT, ECG and HR, (2) an adaptive feature selection strategy based on recursive feature elimination (RFE) is used to select the most significant features in our EDL task, and (3) we design a dedicated EDL model based on convolution neural networks that takes information from both the feature correlation and the time domain into consideration. Experiments were conducted to validate our proposed system, and the results indicated that (1) the proposed feature selection strategy effectively selects significant features and improves algorithmic performance, and (2) the proposed EDL model can obtain good results in terms of six evaluation measures and outperform existing methods. [ABSTRACT FROM AUTHOR]

For Intrusion Detection Systems (IDSs) in Vehicular Ad Hoc Networks (VANETs), single-objective optimization algorithm has inherited limitations for the feature selection problem with the multiple objectives. Moreover, the imbalanced problem commonly exists in various datasets. Thus, in this paper, a feature selection algorithm based on a many-objective optimization algorithm (FS-MOEA) is proposed for IDSs in VANETs, in which Adaptive Non-dominant Sorting Genetic (A-NSGA-III) serves as the many-objective optimization algorithm. Two improvements, called Bias and Weighted (B&W) niche-preservation and Information Gain (IG)-Analytic Hierarchy Process (AHP) prioritizing, are further designed in FS-MOEA. The former is used to counterbalance the imbalanced problem in datasets by assigning rare classes higher priorities, while the latter is employed to search the optimal feature subset for FS-MOEA. In IG-AHP prioritizing, a more distinct measurement, i.e. average IG, is used as the dominant factor to guide the decision analysis of AHP. Experimental results show that the proposed FS-MOEA can not only improve the performance of IDSs in VANETs but also alleviate the negative impact of the imbalanced problem. [ABSTRACT FROM AUTHOR]

This paper reviews the recent literature on technologies and methodologies for quantitative human gait analysis in the context of neurodegenerative diseases. The use of technological instruments can be of great support in both clinical diagnosis and severity assessment of these pathologies. In this paper, sensors, features and processing methodologies have been reviewed in order to provide a highly consistent work that explores the issues related to gait analysis. First, the phases of the human gait cycle are briefly explained, along with some non-normal gait patterns (gait abnormalities) typical of some neurodegenerative diseases. Then the paper reports the most common processing techniques for both feature selection and extraction and for classification and clustering. Finally, a conclusive discussion on current open problems and future directions is outlined. [ABSTRACT FROM AUTHOR]

With advances in digital stethoscopes, internet of things, signal processing and machine learning, chest sounds can be easily collected and transmitted to the cloud for remote monitoring and diagnosis. However, low quality of recordings complicates remote monitoring and diagnosis, particularly for neonatal care. This paper proposes a new method to objectively and automatically assess the signal quality to improve the accuracy and reliability of heart rate (HR) and breathing rate (BR) estimation from noisy neonatal chest sounds. A total of 88 10-second long chest sounds were taken from 76 preterm and full-term babies. Six annotators independently assessed the signal quality, number of detectable beats, and breathing periods from these recordings. For quality classification, 187 and 182 features were extracted from heart and lung sounds, respectively. After feature selection, class balancing, and hyperparameter optimization, a dynamic binary classification model was trained. Then HR and BR were automatically estimated from the chest sound and several approaches were compared.The results of subject-wise leave-one-out cross-validation, showed that the model distinguished high and low quality recordings in the test set with 96% specificity, 81% sensitivity and 93% accuracy for heart sounds, and 86% specificity, 69% sensitivity and 82% accuracy for lung sounds. The HR and BR estimated from high quality sounds resulted in significantly less median absolute error (4 bpm and 12 bpm difference, respectively) compared to those from low quality sounds. The methods presented in this work, facilitates automated neonatal chest sound auscultation for future telehealth applications. [ABSTRACT FROM AUTHOR]

This paper proposes a novel methodology for run-time fluid characterization through the application of machine learning techniques. It aims to integrate sophisticated multi-dimensional photonic sensors inside the chemical processes, following the Industry 4.0 paradigm. Currently, this analysis is done offline in laboratory environments, which increases the decision-making times. As an alternative, the proposed method tunes the spectral-based machine learning solutions to the requirements of each case enabling the integration of compound detection systems at the computing edge. It includes a novel feature selection strategy that combines filters and wrappers, namely Wavelength-based Hybrid Feature Selection, to select the relevant information of the spectrum (i.e., the relevant wavelengths). This technique allows providing different trade-offs involving the spectrum dimensionality, complexity, and detection quality. In terms of execution time, the provided solutions outperform the state-of-the-art up to 61.78 times using less than 99% of the wavelengths while maintaining the same detection accuracy. Also, these solutions were tested in a real-world edge platform, decreasing up to 68.57 times the energy consumption for an ethanol detection use case. [ABSTRACT FROM AUTHOR]

With the growing needs of practical applications such as security monitoring, partial face recognition is a challenging but important issue, because the captured faces in real-world surveillance videos may be occluded or with variations. Though current face recognition methods perform well in relatively constrained scenes, they may suffer from degradation for partial faces. In this paper, we propose a framework of model-based transfer learning and sparse coding (MTLSC) for partial face recognition. First, due to less information in partial face image, we exploit the mirrored image of an original probe sample as sample augment to provide further information. Considering the inadequacy of training face samples, we obtain face features based on model-based transfer learning VGGNet that is pre-trained on VGGFace dataset. Then we reconstruct face features by sliding window in view of different sizes of partial face hard to extract the same feature dimension. Finally we carry out sparse coding with rectification and calculate the minimum score of the probe and mirrored samples among all classes to get the results. Thus, by model-based transfer learning, sliding window for feature reconstruction and sparse coding with rectification, the proposed framework improves partial face recognition performance. Experimental results on three face databases (LFW, AR and NIR), and two person re-identification databases (iLIDS-VID and PKU-Reid) demonstrate our method is effective for partial face recognition. [ABSTRACT FROM AUTHOR]

When learning from high-dimensional data for symbolic regression (SR), genetic programming (GP) typically could not generalize well. Feature selection, as a data preprocessing method, can potentially contribute not only to improving the efficiency of learning algorithms but also to enhancing the generalization ability. However, in GP for high-dimensional SR, feature selection before learning is seldom considered. In this paper, we propose a new feature selection method based on permutation to select features for high-dimensional SR using GP. A set of experiments has been conducted to investigate the performance of the proposed method on the generalization of GP for high-dimensional SR. The regression results confirm the superior performance of the proposed method over the other examined feature selection methods. Further analysis indicates that the models evolved by the proposed method are more likely to contain only the truly relevant features and have better interpretability. [ABSTRACT FROM AUTHOR]

In many real-world scenarios, data from multiple modalities (sources) are collected during a development phase. Such data are referred to as multiview data. While additional information from multiple views often improves the performance, collecting data from such additional views during the testing phase may not be desired due to the high costs associated with measuring such views or, unavailability of such additional views. Therefore, in many applications, despite having a multiview training data set, it is desired to do performance testing using data from only one view. In this paper, we present a multiview feature selection method that leverages the knowledge of all views and use it to guide the feature selection process in an individual view. We realize this via a multiview feature weighting scheme such that the local margins of samples in each view are maximized and similarities of samples to some reference points in different views are preserved. Also, the proposed formulation can be used for cross-view matching when the view-specific feature weights are pre-computed on an auxiliary data set. Promising results have been achieved on nine real-world data sets as well as three biometric recognition applications. On average, the proposed feature selection method has improved the classification error rate by 31 percent of the error rate of the state-of-the-art. [ABSTRACT FROM AUTHOR]

With the growth of IoT applications, sensor data quality has become increasingly important to ensure the success of these data-driven applications. Sensor data riddled with errors are redundant and affects the accuracy of decision-making results. Moreover, several constraints are inherent in anomaly detection for IoT applications such as limited manpower, time, bandwidth, computational resources and the lack of labelled datasets. Most machine learning algorithms also require a large training set to fit a satisfactory model. In this paper, we propose a fully automated anomaly detection framework, which combines systematic time series feature engineering with unsupervised feature selection. The unsupervised feature selection approach automatically selects time series features based on their predictive power with respect to the statistics of near future measurements. The proposed framework also only needs a short calibration phase with respect to the deployment phase to detect anomalies despite not having seen any in training, which is critical for real-world applications. Its feature selection reduces the amount of data required in the deployment phase on an IoT application. The selected features are suitable for building a reliable anomaly detection model while achieving a similar or better anomaly detection performance than established methods, which are operating on the raw data. Results of the evaluation on two publicly available environmental monitoring datasets show that our proposed unsupervised feature selection approach is a crucial step for having a more accurate anomaly detection while providing complex application-specific time series features, which are safeguarding the sensor system against unseen sensor anomalies. [ABSTRACT FROM AUTHOR]

Although cluttered indoor scenes have a lot of useful high-level semantic information which can be used for mapping and localization, most visual odometry (VO) algorithms rely on the usage of geometric features such as points, lines, and planes. Lately, driven by this idea, the joint optimization of semantic labels and estimating odometry has gained popularity in the robotics community. This joint optimization method is accurate but is generally very slow. At the same time, in the vision community, direct and sparse approaches for VO have stricken the right balance between speed and accuracy. We merge the successes of these two communities and present a preprocessing method to incorporate semantic information in the form of visual saliency to direct sparse odometry (DSO)—a highly successful direct sparse VO algorithm. We also present a framework to filter the visual saliency based on scene parsing. Our framework SalientDSO relies on the widely successful deep learning-based approaches for visual saliency and scene parsing, which drives the feature selection for obtaining highly accurate and robust VO even in the presence of as few as 40 point features per frame. We provide an extensive quantitative evaluation of SalientDSO on the ICL-NUIM and the TUM monoVO data sets and show that we outperform DSO and ORB-simultaneous localization and mapping—two very popular state-of-the-art approaches in the literature. We also collect and publicly release a CVL-UMD data set which contains two indoor cluttered sequences on which we show qualitative evaluations. To the best of our knowledge, this is the first paper to use visual saliency and scene parsing to drive the feature selection in direct VO. Note to Practitioners—The algorithm of estimating the camera motion from a set of moving camera frames/images is commonly called VO. This problem has many applications ranging from building a 3-D map of the scene for the robot to navigate, grasp, and so on. Any VO algorithm must be fast, robust, and with low drift (low accumulation in error). These desired functions are generally obtained by selecting “good” features in an image, which, in the computer vision sense, turns out to be “corners.” However, when we constrain the setting to an indoor scene with a lot of clutter, we have a lot of objects which can be used to obtain “good” features from both a computer vision sense and a conceptual sense. We use this philosophy and present a preprocessing method to select better features as compared to a traditional VO pipeline using only geometric features and improve the robustness of the state-of-the-art VO method: direct sparse odometry, obtaining more accurate and robust results even with the lesser number of features. We evaluate our methods on three different data sets: ICL-NUIM, TUM monoVO, and CVL-UMD. We collected a custom dataset we call CVL-UMD to demonstrate the robustness of our approach, namely, SalientDSO in cluttered indoor scenes. [ABSTRACT FROM AUTHOR]

Forecasting spot prices of electricity is challenging because it not only contains seasonal variations, but also random, abrupt spikes, which depend on market conditions and network contingencies. In this paper, a hybrid model has been developed to forecast the spot prices of electricity in two main stages. In the first stage, the prices are forecasted using autoregressive time varying (ARXTV) model with exogenous variables. To improve the forecasting ability of the ARXTV model, the price variations in the training process have been smoothened using the wavelet technique. In the second stage, a kernel regression is used to estimate the price spikes, which are detected using support vector machine based model. In addition, mutual information technique is employed to select appropriate input variables for the model. A case study is carried out with the aid of price data obtained from the Australian energy market operator. It is demonstrated that the proposed hybrid method can accurately forecast electricity prices containing spikes. [ABSTRACT FROM AUTHOR]

Optimal selection of features of Partial Discharge (PD) signals recorded from defects in High Voltage (HV) cables will contribute not only to the improvement of PD pattern recognition accuracy and efficiency but also to PD parameter visualization in HV cable condition monitoring and diagnostics. This paper presents a novel Random Forest (RF)-based feature-selection algorithm for PD pattern recognition of HV cables. The algorithm is applied to feature selection of both PD signals and interference signals with the aim of obtaining the optimal features for data processing. First, the experimental data acquisition and feature extraction processes are introduced. PD signals were captured from faults created in a cable to obtain the raw PD data, and then, a set of 3500 transient PD pulses and a set of 3500 typical interference pulses were extracted, based on which 34 PD features were extracted for further processing. Furthermore, 119 two-dimensional features and 1082 three-dimensional features were generated. The paper then discusses the basic principle of the RF algorithm. Finally, RF-based feature selection was implemented to determine the optimal features for PD pattern recognition. The results were obtained and evaluated with the Back Propagation Neural Network and Support Vector Machine. Results show that the proposed RF-based method is effective for PD feature selection of HV cables with the potential for application to additional HV power apparatus. [ABSTRACT FROM AUTHOR]

This paper is concerned with the network selection for heterogeneous networks. Although many network selection approaches have been developed for heterogeneous networks, few of them explore the network selection problem for integrated cellular and drone-cell networks. Furthermore, most of the existing designs may be unsuitable for the integrated cellular and drone-cell networks owing to the high dynamic drone-cell-user association and the relatively limited drone-cell network capacity. In this paper, we investigate a network selection problem for the integrated cellular and drone-cell networks with a goal of maximizing a proportional fairness function of time average utilities across users under coarse correlated equilibrium constraints and minimum time average utility constraints. To mitigate this challenging problem, we first convert it into a non-linear integer programming (NLIP) problem based on our derived theoretical results. Next, we propose a repeated-stochastic-game-based efficient and fair network selection (RSG-EF) algorithm to alleviate the NLIP problem by leveraging a linear approximation mechanism. Simulation results show that the RSG-EF algorithm can achieve the highest total utility and a high level of fairness across users compared with three benchmark algorithms. [ABSTRACT FROM AUTHOR]

Locality preserving projections (LPP) has been widely studied and extended in recent years, because of its promising performance in feature extraction. In this paper, we propose a modified version of the LPP by constructing a novel regression model. To improve the performance of the model, we impose a low-rank constraint on the regression matrix to discover the latent relations between different neighbors. By using the $L_{2,1}$ -norm as a metric for the loss function, we can further minimize the reconstruction error and derive a robust model. Furthermore, the $L_{2,1}$ -norm regularization term is added to obtain a jointly sparse regression matrix for feature selection. An iterative algorithm with guaranteed convergence is designed to solve the optimization problem. To validate the recognition efficiency, we apply the algorithm to a series of benchmark datasets containing face and character images for feature extraction. The experimental results show that the proposed method is better than some existing methods. The code of this paper can be downloaded from http://www.scholat.com/laizhihui. [ABSTRACT FROM AUTHOR]

Identifying statistical dependence between the features and the label is a fundamental problem in supervised learning. This paper presents a framework for estimating dependence between numerical features and a categorical label using generalized Gini distance, an energy distance in reproducing kernel Hilbert spaces (RKHS). Two Gini distance based dependence measures are explored: Gini distance covariance and Gini distance correlation. Unlike Pearson covariance and correlation, which do not characterize independence, the above Gini distance based measures define dependence as well as independence of random variables. The test statistics are simple to calculate and do not require probability density estimation. Uniform convergence bounds and asymptotic bounds are derived for the test statistics. Comparisons with distance covariance statistics are provided. It is shown that Gini distance statistics converge faster than distance covariance statistics in the uniform convergence bounds, hence tighter upper bounds on both Type I and Type II errors. Moreover, the probability of Gini distance covariance statistic under-performing the distance covariance statistic in Type II error decreases to 0 exponentially with the increase of the sample size. Extensive experimental results are presented to demonstrate the performance of the proposed method. [ABSTRACT FROM AUTHOR]

Adverse drug-drug interaction (ADDI) is a significant life-threatening issue, posing a leading cause of hospitalizations and deaths in healthcare systems. This paper proposes a unified Multi-Attribute Discriminative Representation Learning (MADRL) model for ADDI prediction. Unlike the existing works that equally treat features of each attribute without discrimination and do not consider the underlying relationship among drugs, we first develop a regularized optimization problem based on CUR matrix decomposition for joint representative drug and discriminative feature selection such that the selected drugs and features can well approximate the original feature spaces and the critical factors discriminative to ADDIs can be properly explored. Different from the existing models that ignore the consistent and unique properties among attributes, a Generative Adversarial Network (GAN) framework is then designed to capture the inter-attribute shared and intra-attribute specific representations of adverse drug pairs for exploiting their consensus and complementary information in ADDI prediction. Meanwhile, MADRL is compatible with any kind of attributes and capable of exploring their respective effects on ADDI prediction. An iterative algorithm based on the alternating direction method of multipliers is developed for optimization. Experiments on publicly available dataset demonstrate the effectiveness of MADRL when compared with eleven baselines and its six variants. [ABSTRACT FROM AUTHOR]

Click-through rate (CTR) prediction, whose goal is to predict the probability of the user to click on an item, has become increasingly significant in the recommender systems. Recently, some deep learning models with the ability to automatically extract the user interest from his/her behaviors have achieved great success. In these work, the attention mechanism is used to select the user interested items in historical behaviors, improving the performance of the CTR predictor. Normally, these attentive modules can be jointly trained with the base predictor by using gradient descents. In this paper, we regard user interest modeling as a feature selection problem, which we call user interest selection. For such a problem, we propose a novel approach under the framework of the wrapper method, which is named Meta-Wrapper. More specifically, we use a differentiable module as our wrapping operator and then recast its learning problem as a continuous bilevel optimization. Moreover, we use a meta-learning algorithm to solve the optimization and theoretically prove its convergence. Meanwhile, we also provide theoretical analysis to show that our proposed method 1) efficiencies the wrapper-based feature selection, and 2) achieves better resistance to overfitting. Finally, extensive experiments on three public datasets manifest the superiority of our method in boosting the performance of CTR prediction. [ABSTRACT FROM AUTHOR]