Linearly Separable Research Articles

Relative abundance, habitat selection, and diet of the coyote in northern México

The coyote (Canis latrans) is a widespread predator with a high degree of adaptation to different ecosystems. The objective of this study was to estimate the relative abundance index, habitat selection, and diet of C. latrans through scent stations, in two types of vegetation located in northwestern Chihuahua. From April 2018 to March 2019, ten fixed scent stations (SS) were placed in pine-oak forest and other ten in semi-open grassland, with a linear separation of 500 meters between each station to achieve a five km transect in each vegetation during 12 samplings (two sampling night per month) for totaling 420 SS after discarding inactive SS. The relative abundance index of coyote showed that both types of vegetation, pine-oak forest (0.30) and semi-open grassland (0.23) were used in a similar way. The habitat selection test (Chi2) showed that coyote abundance and type of vegetation were independent (χ2 = 2.96, P > 0.05), not showing statistically significant differences in annual relative abundance index of coyotes between the two vegetation types. The food items detected in thirty-four scats collected belonged to mammals (55.9 %), fruits (35.3 %) and arthropods (8.8 %). Rodents and lagomorphs were the main source of food. Throughout the sampling period, the pine-oak forest showed the greatest relative abundance index of C. latrans. This may be because the forest provides them with shelter from climatic situations and a greater variety of seeds, fruits and insects. In this study the two seasons with the highest relative abundance were spring and autumn in both ecosystems, coinciding with an increase in rainfall, resulting in an abundance of potential prey (rodents and lagomorphs), and other food items like fruits. Regarding annual diet the most consumed prey were mammals. We found differences in diet between seasons, that can be explained by the variation in food availability among seasons.

Driving Fatigue Monitoring via Kernel Sparse Representation Regression With GMC Penalty

Automatic monitoring technology for fatigued driving can greatly reduce traffic accidents caused by drivers due to decreased attention. The use of monitoring algorithms based on physiological signal sensors is of great significance for objective monitoring. In this paper, a novel algorithm with excellent performance and strong generalization ability, kernel sparse representation regression based on generalized minimax-concave (GMC-KSRR), is proposed to identify fatigue physiological signals. Specifically, in the proposed algorithm, the training samples of physiological features are mapped in reproducing kernel Hilbert space (RKHS) for linear separability. Then, in RKHS, the sparse coefficients of the test samples are obtained by generalized minimax-concave (GMC) penalty-based sparse representation (SR). Different from the traditional <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\ell _{{1}}$ </tex-math></inline-formula> -norm, the GMC penalty we used does not underestimate the high-amplitude component of sparsity coefficients, resulting in a more accurate regression result. Finally, the regression decision is performed in the label subspace according to the obtained sparse coefficients. Meanwhile, a multimodality algorithm based on GMC-KSRR has also been proposed that weighs each modality according to quality for robust fatigue monitoring. Eventually, competitive experimental results on the well-known SEED-VIG dataset against the state-of-the-art methods demonstrate the efficacy of the proposed algorithms in identifying fatigue physiological signals, indicating its powerful application potential in driving fatigue monitoring.

A novel autoencoder approach to feature extraction with linear separability for high-dimensional data.

Feature extraction often needs to rely on sufficient information of the input data, however, the distribution of the data upon a high-dimensional space is too sparse to provide sufficient information for feature extraction. Furthermore, high dimensionality of the data also creates trouble for the searching of those features scattered in subspaces. As such, it is a tricky task for feature extraction from the data upon a high-dimensional space. To address this issue, this article proposes a novel autoencoder method using Mahalanobis distance metric of rescaling transformation. The key idea of the method is that by implementing Mahalanobis distance metric of rescaling transformation, the difference between the reconstructed distribution and the original distribution can be reduced, so as to improve the ability of feature extraction to the autoencoder. Results show that the proposed approach wins the state-of-the-art methods in terms of both the accuracy of feature extraction and the linear separabilities of the extracted features. We indicate that distance metric-based methods are more suitable for extracting those features with linear separabilities from high-dimensional data than feature selection-based methods. In a high-dimensional space, evaluating feature similarity is relatively easier than evaluating feature importance, so that distance metric methods by evaluating feature similarity gain advantages over feature selection methods by assessing feature importance for feature extraction, while evaluating feature importance is more computationally efficient than evaluating feature similarity.

Applying the FN-corrector to improve the quality of audio event classification

The paper deals with the problem of acoustic events classification which is actively applied to the problems of a safe city, smart home, IoT devices, and for the detection of industrial accidence. A solution to improve the accuracy of classifiers without changing their structure and collecting additional data is proposed. The main data source for the experiments was the TUT Urban Acoustic Scenes 2018, Development Dataset. The paper presents the way to increase the accuracy of audio event classification by using the FN-corrector. The FN-corrector is a linear two-stage classifier performing the transformation of the feature space into a linearly separable space and the linear separation of one class from another. If a corrector is applied, the responses of the original classifier generate four classes: positive (P), negative (N), false positive (FP), and false negative (FN). As a result, it becomes possible to train two types of correctors: the FP-corrector separating positive and false positive classifier responses, and the FN-corrector separating negative and false negative classifier responses. In the experiments, the VGGish convolutional neural network was used as the initial classifier. The audio signal is converted into a spectrogram and is fed to the input of the neural network which forms the spectrogram feature description and performs a classification. As an example, two ”confused“ classes are selected to demonstrate the increase in classification accuracy. Using the feature description of audio recordings of these classes, an FN-corrector was built, trained and connected to the original classifier. The response from the classifier, as well as the feature description, has been passed to the corrector input. Next, the corrector translated the feature space into a new basis (into a linearly separable space) and classified the classifier answer responding to the question whether the original classifier makes a mistake on such a feature vector or not. If the original classifier made a mistake, then his answer is changed by the corrector to the opposite, otherwise the answer remains the same. The results of the experiments demonstrated a decrease in the level of class confusion and, accordingly, an increase in the accuracy of the original classifier without changing its structure and without collecting an additional data set. The results obtained can be used on IoT devices that have significant limitations on the size of the models used, as well as in solving the problems of domain adaptation which is relevant in audio analytics

To Estimate Performance of Artificial Neural Network Model Based on Terahertz Spectrum: Gelatin Identification as an Example.

It is a necessity to determine significant food or traditional Chinese medicine (TCM) with low cost, which is more likely to achieve high accurate identification by THz-TDS. In this study, feedforward neural networks based on terahertz spectra are employed to predict the animal origin of gelatins, whose adaption to the mission is examined by parallel models built by random sample partition and initialization. It is found that the generalization performance of feedforward ANNs in original data is not satisfactory although prediction on trained samples can be accurate. A multivariate scattering correction is conducted to enhance prediction accuracy, and 20 additional models verify the effectiveness of such dispose. A special partition of total dataset is conducted based on statistics of parallel models, whose influence on ANN performance is investigated with another 20 models. The performance of the models is unsatisfactory because of notable differences in training and test sets according to principal component analysis. By comparing the distribution of the first two principal components before and after multivariate scattering correction, we found that the reciprocal of the minimum number of line segments required for error-free classification in 2-D feature space can be viewed as an index to describe linear separability of data. The rise of proposed linear separability would have a lower requirement for harsh parameter tuning of ANN models and tolerate random initialization. The difference in principal components of samples between a training set and a data set determines whether partition is acceptable or whether a model would have generality. A rapid way to estimate the performance of an ANN before sufficient tuning on a classification mission is to compare differences between groups and differences within groups. Given that a representative peak missing curve is discussed in this article, an analysis based on gelatin THz spectra may be helpful for studies on some other feature-less species.

Klasifikasi Citra Jenis Tanaman Jamur Layak Konsumsi Menggunakan Algoritma Multiclass Support Vector Machine

Mushrooms are plants that have high nutritional content and have various benefits for the health of the human body. However, not everyone knows the types of mushrooms that are suitable for consumption. The types of mushrooms have their own characteristics when viewed from the image. For this reason, a system is needed by utilizing digital image processing to classify types of mushrooms suitable for consumption, so that people can find out which types of mushrooms are suitable for consumption. This research is to classify types of mushrooms suitable for consumption using the Multiclass SVM algorithm with first-order feature extraction, which performs feature extraction based on the characteristics of the image histogram. The result of feature extraction is used as input for classification in Multiclass SVM. Multiclass SVM can map data points to dimensionless space to obtain hyperplane linear separation between each class. The developed method is implemented in Matlab, in order to produce a system in the form of a GUI so that it can be used by general users easily. Based on the test results, the average accuracy is 83%.

Modelling of the intergranular fracture of TWIP steels working at high temperature by using CZM–CPFE method

Grain boundaries (GBs) of metallic materials will become the weakest and most vulnerable place when the materials deform at high temperature. This could directly affect deformation flow, plastic strengthening, the initiation and propagation of microcracks, and finally the fracture failure of materials. In this research, to reveal the high temperature deformation mechanisms of twinning–induced plasticity (TWIP) steels, the crystal plasticity finite element method (CPFEM) was employed to model the mechanical response of grain interiors. The cohesive zone model (CZM) of the bi–linear traction separation law (TSL) was developed to describe the damage and failure of GBs. By combining CZM and CPFEM, an analysis method was proposed to investigate the effect of the microstructure morphology, orientation and size of grains, and temperature. Furthermore, the damage initiation, accumulation and fracture at GBs were represented by the representative volume element model based on the CZM–CPFE method to explore the influences of GB angle, grain size and local microvoids (including the location and size) on the GB cracking. The stress damage cloud maps, the distribution of quadratic stress damage initiation criterion (QUADSCRT) and the scalar stiffness degradation (SDEG) describing the damage and fracture along the GBs were used to articulate the relationships of GB angle, size effect, local microvoids and the expected failure strain. The results show that low angle grain boundaries (LAGBs), the small average grain size (<21.2 μm), microvoids at the horizontal GBs and microvoids with a diameter below 1.5 μm can delay the initiation of microcracks and fracture failure along the GBs. The predicted damage and failure of GBs and the intergranular cracking characteristics in plastic deformation of TWIP steels at high temperature were corroborated by microscopic in–situ SEM experiments, in which the deformation characteristics inside the grains, and microcrack initiation and propagation near GBs were observed at 500 and 750°C. This research enhances the understanding of intergranular fracture, mechanical response and microstructure evolution of TWIP steels worked at different temperatures, and also provides a new way and strategy for studying the deformation behaviours of TWIP steels considering GB properties.

Deep Incomplete Multi-View Clustering via Mining Cluster Complementarity

Incomplete multi-view clustering (IMVC) is an important unsupervised approach to group the multi-view data containing missing data in some views. Previous IMVC methods suffer from the following issues: (1) the inaccurate imputation or padding for missing data negatively affects the clustering performance, (2) the quality of features after fusion might be interfered by the low-quality views, especially the inaccurate imputed views. To avoid these issues, this work presents an imputation-free and fusion-free deep IMVC framework. First, the proposed method builds a deep embedding feature learning and clustering model for each view individually. Our method then nonlinearly maps the embedding features of complete data into a high-dimensional space to discover linear separability. Concretely, this paper provides an implementation of the high-dimensional mapping as well as shows the mechanism to mine the multi-view cluster complementarity. This complementary information is then transformed to the supervised information with high confidence, aiming to achieve the multi-view clustering consistency for the complete data and incomplete data. Furthermore, we design an EM-like optimization strategy to alternately promote feature learning and clustering. Extensive experiments on real-world multi-view datasets demonstrate that our method achieves superior clustering performance over state-of-the-art methods.

Experimental Quantum Embedding for Machine Learning

AbstractThe classification of big data usually requires a mapping onto new data clusters which can then be processed by machine learning algorithms by means of more efficient and feasible linear separators. Recently, Lloyd et al. have advanced the proposal to embed classical data into quantum ones: these live in the more complex Hilbert space where they can get split into linearly separable clusters. Here, these ideas are implemented by engineering two different experimental platforms, based on quantum optics and ultra‐cold atoms, respectively, where we adapt and numerically optimize the quantum embedding protocol by deep learning methods, and test it for some trial classical data. A similar analysis is also performed on the Rigetti superconducting quantum computer. Therefore, it is found that the quantum embedding approach successfully works also at the experimental level and, in particular, we show how different platforms could work in a complementary fashion to achieve this task. These studies might pave the way for future investigations on quantum machine learning techniques especially based on hybrid quantum technologies.

Epileptic electroencephalography classification using embedded dynamic mode decomposition

Objective. Seizure prediction devices for drug-resistant epileptic patients could lead to improved quality of life and new treatment options, but current approaches to classification of electroencephalography (EEG) segments for early identification of the pre-seizure state typically require many features and complex classifiers. We therefore propose a novel spatio-temporal EEG feature set that significantly aids in separation and easy classification of the interictal and preictal states. Approach. We derive key spectral features from the embedded dynamic mode decomposition (EmDMD) of the brain state system. This method linearizes the complex spatio-temporal dynamics of the system, describing the dynamics in terms of a spectral basis of modes and eigenvalues. The relative subband spectral power and mean phase locking values of these modes prove to be good indicators of the preictal state that precedes seizure onset. Main results. We analyze the linear separability and classification of preictal and interictal states based on our proposed features using seizure data extracted from the CHB-MIT scalp EEG and Kaggle American Epilepsy Society Seizure Prediction Challenge intracranial EEG databases. With a light-weight support vector machine or random forest classifier trained on these features, we classify the preictal state with a sensitivity of up to 92% and specificity of up to 89%. Significance. The EmDMD-derived features separate the preictal and interictal states, improving classification accuracy and motivating further work to incorporate them into seizure prediction algorithms.

The coupled bending-torsional dynamic behavior in the rotating machinery: modeling, simulation and experiment validation

The nonlinear coupling between bending and torsional vibrations would greatly affect the dynamic behavior of the rotor system. Different from the traditional way, in this paper the unbalance mass that causes the internal and external couplings is treated as an independent nonlinear element when establishing a finite element model of the coupled bending-torsional (CBT) rotor system with six degrees of freedom for each node. To improve the efficiency of transient numerical simulation, a general algorithm named with linear and nonlinear nodes separation method is proposed, which decreases the dimension of nonlinear iteration matrix, resulting in a great calculation efficiency of solving transient vibration responses of the high-dimension nonlinear motion equation. With the presented nonlinear CBT model, simulations for steady and transient torsional excitation processes are carried out, resulting that the torsional excitation would generate sideband frequency components in the spectra of lateral vibrations and vice versa, and the effect of gravity is considerable and non-negligible in CBT modeling. Additionally, results of experiments with two different scales of setups, a laboratory test-rig and an industrial centrifugal compressor platform, are well agreed with the numerical ones. The investigation provides a complete modeling and high efficient solution method not only for CBT rotordynamic problems, but also for other problems with nonlinear forces.

Simulation of the crystallization processes by population balance model using a linear separation method

Abstract Crystallization problem is one of the popular problems in wide area of science. The first principles are not used to design a crystallizer in which complicated processes include nucleation, crystal growth, attrition and agglomeration of crystals. It is modeled by the population balance model, which is one of the important models of mathematical biology and engineering, is a nonlinear partial integro-differential equation and examines the exchange of particles and the production of new particles in a system of particles. For the crystallization problem, one-dimensional and multi-dimensional models are considered and semi-analytical solutions are obtained via the linear separation method.

Emotion Recognition from Speech using SVM and Random Forest Classifier

Speech is the most natural way of people to communicate with one another. It is a vital medium for communicating a person's thoughts, feelings, and mental condition to others. The process of identifying the intellectual state is the recognition of basic emotion through speech. In human life, emotions are incredibly significant. In this project, the emotion is recognized from speech using Support Vector Machine (SVM) and Random Forest classifiers. These are supervised machine learning algorithms used for both classification and regression problems. SVM classifies data by creating N-dimensional hyper planes that divide the input into different categories. The classification is accomplished using a linear and non-linear separation surface in the dataset's input feature. Random Forest is a classifier that combines a number of decision trees on different subsets of a dataset and averages the results to increase the dataset's predicted accuracy. These classifiers are used to categorize emotions like happiness, rage, sadness and neutral for a certain incoming voice signal. Here, the system is trained and developed to recognize emotion in real-time speech. The result demonstrates that the Random Forest classifier is significantly better, when compared to the SVM classifier.

SVM Classifier: Identify Linear Separability Of NAND And NOR Logic Gates

Abstract-Artificial Neural Network plays vital role to resolve several real-life problems. Those problems would be impossible or difficult to solve by human or statistical principles. ANN produces better results on large data set as it has self-learning capabilities. The most frequently used family of ANN is pattern classification. Generally, patterns are obtained from real world and get together based on definite attributes in specific regions. Data is classified based on knowledge by recognizing patterns. This study underlines linear separability of Boolean logic gates according to classification. We considered Boolean functions NAND and NOR for classification task. Our problem statement related to this study is, “There is significant difference between the performance of classifier regarding NAND, NOR Boolean logic gates using Support Vector Machine”. Zoo data set is used to carry out the experiment. Accuracy score is measured using Support Vector Machine (SVM). For actual and predicted data set, classification report and confusion matrix is produced. For classification SVM is emerged as a promising technique. Usually, SVM gives remarkable performance as compare to other Machine Learning classifiers methods. As a result, SVM produces outstanding performance on classification problem. Keywords- Boolean Functions, NAND, NOR, Linear Separability, Support Vector Machine (SVM), Confusion Matrix.

A Novel Method for Fast Kernel Minimum Noise Fraction Transformation in Hyperspectral Image Dimensionality Reduction

Feature extraction, aiming to simplify and optimize data features, is a typical hyperspectral image dimensionality reduction technique. As a kernel-based method, kernel minimum noise fraction (KMNF) transformation is excellent at handling the nonlinear features within HSIs. It adopts the kernel function to ensure data linear separability by transforming the original data to a higher feature space, following which a linear analysis can be performed in this space. However, KMNF transformation has the problem of high computational complexity and low execution efficiency. It is not suitable for the processing of large-scale datasets. In terms of this problem, this paper proposes a graphics processing unit (GPU) and Nyström method-based algorithm for Fast KMNF transformation (GNKMNF). First, the Nyström method estimates the eigenvector of the entire kernel matrix in KMNF transformation by the decomposition and extrapolation of the sub-kernel matrix to reduce the computational complexity. Then, the sample size in the Nyström method is determined utilizing a proportional gradient selection strategy. Finally, GPU parallel computing is employed to further improve the execution efficiency. Experimental results show that compared with KMNF transformation, improvements of up to 1.94% and 2.04% are achieved by GNKMNF in overall classification accuracy and Kappa, respectively. Moreover, with a data size of 64 × 64 × 250, the execution efficiency of GNKMNF speeds up by about 80×. The outcome demonstrates the significant performance of GNKMNF in feature extraction and execution efficiency.

Wavefield separation using irreversible-migration filtering

Wavefield separation is important for eliminating unwanted components of seismic data while retaining preferred components therein; however, there have been difficulties with using traditional methods to select a proper muting window in the transformed domain. A narrow muting window can separate different components well, but it causes visible artifacts due to sudden truncations (e.g., the Gibbs phenomenon), whereas a wide muting window would, on the other hand, leave unwanted components. We have adopted separating the wavefield based on its irreversibility after migration and demigration. One-way wave-equation migration would automatically damp high-slope components in the evanescent region while accurately handling the other components. Thus, the migration velocity can be tuned to naturally trap a given range of high-slope components into the evanescent region in the migrated domain, which would be irreversible after demigration. In contrast, the other components (i.e., those outside the evanescent region) can be recovered after demigration. Our method achieves perfect muting by taking advantage of migration and demigration, and thus it avoids the manual operation of setting a muting window. As a result, our method is free of muting artifacts. We conduct numerical experiments with synthetic and field data, and the results verify the excellent performance of our method for several different kinds of wavefield separation, including linear event separation, structural noise elimination, diffraction-reflection separation, and vertical seismic profile wavefield separation. Our method integrates noise reduction and the wavefield separation, and thus it can reduce the computational cost of using different data processing schemes and avoid the related potential error accumulation.

On Secure Distributed Linearly Separable Computation

Distributed linearly separable computation, where a user asks some distributed servers to compute a linearly separable function, was recently formulated by the same authors and aims to alleviate the bottlenecks of stragglers and communication cost in distributed computation. The data center assigns a subset of input datasets to each server in an uncoded manner, and each server computes some coded packets on the assigned datasets, which are then sent to the user. The user should recover the task function from the answers of a subset of servers, such that the effect of stragglers could be tolerated. In this paper, we formulate a novel secure framework for this distributed linearly separable computation, where we aim to let the user only retrieve the desired task function without obtaining any other information about the input datasets, even if it receives the answers of all servers. In order to preserve the security of the input datasets, some common randomness variable independent of the datasets should be introduced into the transmission. We show that any non-secure linear-coding based computing scheme for the original distributed linearly separable computation problem, can be made secure without increasing the communication cost (number of symbols the user should receive). Then we focus on the case where the computation cost of each server (number of datasets assigned to each server) is minimum and aim to minimize the size of the randomness variable (i.e., randomness size) introduced in the system while achieving the optimal communication cost. We first propose an information theoretic converse bound on the randomness size. We then propose secure computing schemes based on two well-known data assignments, namely <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">fractional repetition assignment</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">cyclic assignment</i> . These schemes are optimal subject to using these assignments. Motivated by the observation of the general limitation of these two schemes on the randomness size, we propose a computing scheme with novel assignment, which strictly outperforms the above two schemes. Some additional optimality results are also obtained.

Linear separability, irrelevant variability, and categorization difficulty.

In rule-based (RB) category-learning tasks, the optimal strategy is a simple explicit rule, whereas in information-integration (II) tasks, the optimal strategy is impossible to describe verbally. This study investigates the effects of two different category properties on learning difficulty in category learning tasks-namely, linear separability and variability on stimulus dimensions that are irrelevant to the categorization decision. Previous research had reported that linearly separable II categories are easier to learn than nonlinearly separable categories, but Experiment 1, which compared performance on linearly and nonlinearly separable categories that were equated as closely as possible on all other factors that might affect difficulty, found that linear separability had no effect on learning. Experiments 1 and 2 together also established a novel dissociation between RB and II category learning: increasing variability on irrelevant stimulus dimensions impaired II learning but not RB learning. These results are all predicted by the best available measures of difficulty in RB and II tasks. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

A heuristic approach for multiple instance learning by linear separation

We present a fast heuristic approach for solving a binary multiple instance learning (MIL) problem, which consists in discriminating between two kinds of item sets: the sets are called bags and the items inside them are called instances. Assuming that only two classes of instances are allowed, a common standard hypothesis states that a bag is positive if it contains at least a positive instance and it is negative when all its instances are negative. Our approach constructs a MIL separating hyperplane by preliminary fixing the normal and reducing the learning phase to a univariate nonsmooth optimization problem, which can be quickly solved by simply exploring the kink points. Numerical results are presented on a set of test problems drawn from the literature.

The Signed Cumulative Distribution Transform for 1-D signal analysis and classification

&lt;p style='text-indent:20px;'&gt;This paper presents a new mathematical signal transform that is especially suitable for decoding information related to non-rigid signal displacements. We provide a measure theoretic framework to extend the existing Cumulative Distribution Transform [&lt;xref ref-type="bibr" rid="b29"&gt;29&lt;/xref&gt;] to arbitrary (signed) signals on &lt;inline-formula&gt;&lt;tex-math id="M1"&gt;\begin{document}$ \overline {\mathbb{R}} $\end{document}&lt;/tex-math&gt;&lt;/inline-formula&gt;. We present both forward (analysis) and inverse (synthesis) formulas for the transform, and describe several of its properties including translation, scaling, convexity, linear separability and others. Finally, we describe a metric in transform space, and demonstrate the application of the transform in classifying (detecting) signals under random displacements.&lt;/p&gt;