Value Of Kernel Research Articles

Comparison of Support Vector Machine Performance with Oversampling and Outlier Handling in Diabetic Disease Detection Classification

Diabetes mellitus is a disease that attacks chronic metabolism, characterized by the body’s inability to process carbohydrates, fats so that glucose levels are high. Diabetes mellitus is the sixth cause of death in the world. Classifying data about diabetes mellitus makes it easier to predict the disease. As technology develops, diabetes mellitus can be detected using machine learning methods. The method that can be done is the support vector machine. The advantage of SVM is that it is very effective in completing classification, so it can quickly separate each positive and negative point. This study aimed to obtain the best SVM classification model based on accuracy, sensitivity, and precision values in detecting diabetes by adding Synthetic Minority Over-Sampling Technique (SMOTE) and handling outliers. The SMOTE method was applied to handle class imbalance. The Support Vector Machine (SVM) method aimed to produce a function as a dividing line or what can be called a hyperplane that matches all input data with the smallest possible error. The data studied were indications of diabetes, consisting of 8-factor variables and 1 class variable. The test results show that the SVM-SMOTE scenario produces the best accuracy. The SVM SMOTE scenario produced an accuracy value of the RBF kernel of 88% with an error of 12%, and this is obtained from the division of test data and training data of 90:10. This SVM-SMOTE scenario produced a precision value of 0.880 and a sensitivity value of 0.880. The research results showed that factor classification was more accurate if it is carried out using the support vector machine (SVM) method with imbalance data handling (SMOTE), and it can be concluded that the distribution of test data and training data influences a test scenario.

Genetic dissection of embryo size and weight related traits for enhancement of kernel oil in maize

Embryo is a key determinant of kernel-oil in maize. Higher calorific value of maize kernel is attributed to increment in kernel-oil and it stores in specialised structure called embryo. Understanding the genetic behaviour of embryo size and weight related-traits is inevitable task for genetic improvement of kernel-oil. Here, the six-basic generations (P1, P2, F1, F2, BC1P1 and BC1P2) of three crosses (CRPBIO-962 × EC932601, CRPBIO-973 × CRPBIO-966 and CRPBIO-966 × CRPBIO-979) between contrasting embryo-sized maize inbreds were field evaluated at three locations to decipher the genetics of twenty embryo, kernel and embryo-to-kernel related-traits through generation-mean-analysis (GMA). Combined ANOVA revealed the significance of all the traits among generations; however, location and generation × location were found to be non-significant (P > 0.05) for most of the traits. Significance (P < 0.05) of scaling and joint-scaling tests revealed the presence of non-allelic interactions. Elucidation of six-parameters disclosed the predominance of dominance main-effect (h) and dominance × dominance interaction-effect (l) for most of traits. The signs of (h) and (l) indicated the prevalence of duplicate-epistasis type across crosses and locations. Thus, the population improvement approaches along with heterosis breeding method could be effective for improvement of these traits. Quantitative inheritance pattern was observed for all the traits with high broad-sense heritability and better-stability across locations. The study also predicted one to three major-gene blocks/QTLs for embryo-traits and up to 11 major-gene blocks/QTLs for embryo-to-kernel traits. These findings could provide deep insights to strategize extensive breeding methods to improve embryo traits for enhancing kernel-oil in sustainable manner.

동적 장애물 환경에서 Heat Kernel 및 Costmap 레이어를 이용한 AMR 장애물 회피 방법

At the logistics site, interest and demand for AMR(Autonomous Mobile Robot) is increasing. A representative function required in this situation is obstacle avoidance. In particular, research on avoiding difficult or dynamic obstacles is more important. Research on this is continuously being conducted, but there is a lack of research on the case where multiple AMRs mutually avoid each other. In this study, we create a navigation costmap layer using heat kernel and propose an obstacle avoidance method based on it. The proposed method makes it difficult to approach the obstacle by adding a cost equal to the heat by applying Heat Kernel operation around it after detecting the obstacle. Since the resultant value of Heat Kernel increases rapidly as it approaches an obstacle, AMR can safely avoid obstacles or mutual avoidance.

Genotype and Environment Effects on Phytosterol and Tocopherol Contents in Almond Kernel Oil

The phytosterol and tocopherol contents of almond kernel oil were evaluated in three almond cultivars, including the Spanish cultivar ‘Marcona’ and the French cultivars ‘Ferragnès’ and ‘Ferraduel’, grown under two different agro-climatic conditions, Zaragoza (Spain) and Meknès (Morocco). The analysis of variance showed significant effects of the genotype on kernel phytosterol content, Δ5-Avenasterol, Δ7-Campesterol, Stigmasterol and on the three tocopherol isomers. The location effect was significant on oil and kernel phytosterol contents, β-sitosterol,Δ5-avenasterol, Δ7-stigmastenol, Δ7-campesterol and α- and γ-tocopherol. The highest value of kernel and oil phytosterol contents were observed at Meknès (1.48 g kg−1 and 2.54 g kg−1, respectively), as compared to those obtained at Zaragoza (1.31 g kg−1 and 2.54 g kg−1, respectively).The highest values ofβ-sitosterol and Δ5-avenasterol were obtained at the Zaragoza location (81.93% and 10.55% of total phytosterols, respectively).The highest value of α-tocopherol was observed in the Morocco location (496 mg kg−1 oil). These results indicate that under warm climate conditions in Morocco, the almond cultivars tend to accumulate more phytosterol and tocopherol. However, the significance of the genotype × location interaction on the phytosterol and tocopherol contentindicates that the magnitude of variation in these traits mainly depends on the genotype.

On the value of apricot kernel in modern medicine and its future development

On the value of apricot kernel in modern medicine and its future development

Research on a Convolution Kernel Initialization Method for Speeding Up the Convergence of CNN

This paper presents a convolution kernel initialization method based on the local binary patterns (LBP) algorithm and sparse autoencoder. This method can be applied to the initialization of the convolution kernel in the convolutional neural network (CNN). The main function of the convolution kernel is to extract the local pattern of the image by template matching as the target feature of subsequent image recognition. In general, the Xavier initialization method and the He initialization method are used to initialize the convolution kernel. In this paper, firstly, some typical sample images were selected from the training set, and the LBP algorithm was applied to extract the texture information of the typical sample images. Then, the texture information was divided into several small blocks, and these blocks were input into the sparse autoencoder (SAE) for pre-training. After finishing the training, the weight values of the sparse autoencoder that met the statistical features of the data set were used as the initial value of the convolution kernel in the CNN. The experimental result indicates that the method proposed in this paper can speed up the convergence of the network in the network training process and improve the recognition rate of the network to an extent.

تاثیر پرتو الکترون و بخار آب با فشار پایین بر بهبود ارزش تغذیه‌ای پودر هسته خرما برای نشخوارکنندگان

The present experiment was conducted to investigate the effect of the electron beam (beam) and low-pressure steam (steam) treatments on the nutritional value of date kernel powder for ruminants. At first, date kernel powders were irradiated with the beam at doses of 0, 25, 50, 75, 100, and 125 KGy. A part of the irradiated date kernels was then treated with steam (134 ᵒC, 40 min.). Digestion and fermentation of date kernel powder were measured using gas production and two-step digestion techniques. Compared to the control, beam or steam resulted in a decrease in NDF and ADF percentages (P<0.05), a significant increase in potential and rate of gas production, partitioning factor, truly degradable organic matter, and in vitro digestibility of dry matter, NDF, and ADF of date kernel (P<0.05), and the 100 KGy had the best performance (P<0.05). Overall, the treatment with steam or electron beam alone improved the nutritional value of the date kernel, but treatment with steam after beam did not have a positive additive effect. Therefore, the results of the present experiment indicate that beam alone has a great capacity for improving the nutritional value of the date kernel, and the use of steam after the beam has failed to make more improvement.

Insights into the selection mechanism of Widmanstätten growth by phase-field calculations

We investigate the growth of Widmanstätten structures in metallic alloys with phase field calculations. We show that elastic energy anisotropy associated with shear dominated transformations is alone sufficient to give rise to plate or needle like precipitates with constant lengthening rates at constant supersaturations. We show that Ivantsov analytical solutions still provide good estimates of the diffusion flux around tips even when elasticity is accounted for. A careful analysis demonstrates that, for shear dominated transformations, (i) the lengthening direction lies in the habit plane determined by one of the minima of the elastic kernel; (ii) the tip shape and size are equilibrium features, driven by elastic forces, rather than dynamically determined as usually assumed; (iii) the tip size of acicular precipitates can be rationalized by the value of the elastic kernel in the lengthening direction.

Dietary implications of toasted Jatropha curcas kernel on the growth, haematology, and organ histology of Clarias gariepinus fingerlings.

The nutritional value of Jatropha curcas kernel (JCK) can be improved through different processing methods; however, when using a thermal treatment, optimization of the process is needed to prevent denaturation of nutrients. In this study, JCK was toasted for varying durations (0, 10, 20, and 30 min) and nutritionally evaluated. The implication of feeding Clarias gariepinus with dietary inclusions (35% CP; 315 kcal g-1) of the toasted JCK was also reported. The results obtained suggest that prolonged duration of toasting improved the nutritional characteristics of the JCK until the 20th min. Beyond this time, the protein content and essential amino acids are reduced. However, the antinutrients continuously decreased with prolonged processing. The growth, carcass analysis, and haematology of the fish groups fed toasted JCK at varying duration also did better than those fed raw JCK. Importantly, the performance tends to reduce for those fed JCK toasted beyond 20 min. The estimated cost of producing 1 kg of the fish also substantially reduced with feeding the processed JCK than feeding raw JCK. Histological examination of the intestine and liver tissues further revealed fewer signs of histopathological degeneration for fish-fed processed JCK compared to the control. It was concluded that the processing of JCK by toasting should not exceed 20 min to improve the nutritional composition of the feed ingredients and their dietary utilization by fish.

Evaluation of the ultrasonically accelerated debitterizing with citric acid solutions of different p H: On the basis of amino acids changes in apricot kernels during debitterizing

In this study, an investigation was conducted into the effects of different pH values of citric acid solutions combined with ultrasound irradiation on the amino acids of the apricot kernels during debitterizing. The concentrations and varieties of amino acids in the samples were determined using the automatic amino acid analyzer. In addition, the amino acid reference model proposed by WHO/FAO was adopted as the evaluation criteria, and the ratio of essential amino acids in debitterized samples was compared to assess the effect of ultrasonic debitterizing of different pH solutions on the nutritional value carried by apricot kernels amino acids. Finally, with the amino acid content as a parameter, principal component analysis (PCA) and cluster analysis (HCA) were carried out to thoroughly evaluate the nutritional value of amino acids contained in the debitterized apricot kernels. The results indicate that the citric acid solution with a pH value of 5 could be taken as the optimal ultrasonic solution to remove the bitterness of apricot kernels. Specifically, the total amino acid content and the nutritional value of the apricot kernel are high after debitterizing under the condition, which could shorten the debitterizing of apricot kernels and provide highly nutritious amino acids. Novelty impact statement This work can maintain the properties of apricot kernels amino acids to a better extent based on rapid debitterizing. Ultrasound coupled with the citric acid solution as a new method was used to debitterizing apricot kernels. Finally, the debitterizing conditions could be optimized to obtain apricot kernels amino acids of good quality.

Research on News Text Classification Based on Deep Learning Convolutional Neural Network

Aiming at the problems of low classification accuracy and low efficiency of existing news text classification methods, a new method of news text classification based on deep learning convolutional neural network is proposed. Determine the weight of the news text data through the VSM (Viable System Model) vector space model, calculate the information gain of mutual information, and determine the characteristics of the news text data; on this basis, use the hash algorithm to encode the news text data to calculate any news. The spacing between the text data realizes the feature preprocessing of the news text data; this article analyzes the basic structure of the deep learning convolutional neural network, uses the convolutional layer in the convolutional neural network to determine the change value of the convolution kernel, trains the news text data, builds a news text classifier of deep learning convolutional neural network, and completes news text classification. The experimental results show that the deep learning convolutional neural network can improve the accuracy and speed of news text classification, which is feasible.

Pedestrian Reidentification Algorithm Based on Deconvolution Network Feature Extraction‐Multilayer Attention Mechanism Convolutional Neural Network

Pedestrian reidentification is a key technology in large‐scale distributed camera systems. It can quickly and efficiently detect and track target people in large‐scale distributed surveillance networks. The existing traditional pedestrian reidentification methods have problems such as low recognition accuracy, low calculation efficiency, and weak adaptive ability. Pedestrian reidentification algorithms based on deep learning have been widely used in the field of pedestrian reidentification due to their strong adaptive ability and high recognition accuracy. However, the pedestrian recognition method based on deep learning has the following problems: first, during the learning process of the deep learning model, the initial value of the convolution kernel is usually randomly assigned, which makes the model learning process easily fall into a local optimum. The second is that the model parameter learning method based on the gradient descent method exhibits gradient dispersion. The third is that the information transfer of pedestrian reidentification sequence images is not considered. In view of these issues, this paper first examines the feature map matrix from the original image through a deconvolution neural network, uses it as a convolution kernel, and then performs layer‐by‐layer convolution and pooling operations. Then, the second derivative information of the error function is directly obtained without calculating the Hessian matrix, and the momentum coefficient is used to improve the convergence of the backpropagation, thereby suppressing the gradient dispersion phenomenon. At the same time, to solve the problem of information transfer of pedestrian reidentification sequence images, this paper proposes a memory network model based on a multilayer attention mechanism, which uses the network to effectively store image visual information and pedestrian behavior information, respectively. It can solve the problem of information transmission. Based on the above ideas, this paper proposes a pedestrian reidentification algorithm based on deconvolution network feature extraction‐multilayer attention mechanism convolutional neural network. Experiments are performed on the related data sets using this algorithm and other major popular human reidentification algorithms. The results show that the pedestrian reidentification method proposed in this paper not only has strong adaptive ability but also has significantly improved average recognition accuracy and rank‐1 matching rate compared with other mainstream methods.

Generating Cartoon Images from Face Photos with Cycle-Consistent Adversarial Networks

The generative adversarial network (GAN) is first proposed in 2014, and this kind of network model is machine learning systems that can learn to measure a given distribution of data, one of the most important applications is style transfer. Style transfer is a class of vision and graphics problems where the goal is to learn the mapping between an input image and an output image. CYCLE-GAN is a classic GAN model, which has a wide range of scenarios in style transfer. Considering its unsupervised learning characteristics, the mapping is easy to be learned between an input image and an output image. However, it is difficult for CYCLE-GAN to converge and generate high-quality images. In order to solve this problem, spectral normalization is introduced into each convolutional kernel of the discriminator. Every convolutional kernel reaches Lipschitz stability constraint with adding spectral normalization and the value of the convolutional kernel is limited to [0, 1], which promotes the training process of the proposed model. Besides, we use pretrained model (VGG16) to control the loss of image content in the position of l1 regularization. To avoid overfitting, l1 regularization term and l2 regularization term are both used in the object loss function. In terms of Frechet Inception Distance (FID) score evaluation, our proposed model achieves outstanding performance and preserves more discriminative features. Experimental results show that the proposed model converges faster and achieves better FID scores than the state of the art.

Quantitative identification of coal texture using the support vector machine with geophysical logging data: A case study using medium-rank coal from the Panjiang, Guizhou, China

Coal texture is important for predicting coal seam permeability and selecting favorable blocks for coalbed methane (CBM) exploration. Drilled cores and mining seam observations are the most direct and effective methods of identifying coal texture; however, they are expensive and cannot be used in unexplored coal seams. Geophysical logging has become a common method of coal texture identification, particularly during the CBM mining stage. However, quantitative methods for identifying coal texture based on geophysical logging data require further study. The support vector machine (SVM), a machine-learning method, has received great interest due to its remarkable generalization performance, and it has been used to quantitatively identify hard and soft coal using geophysical logging data. In this study, four well-logging curves, the acoustic time difference (AC), caliper log (CAL), density (DEN), and natural gamma (GR), were used for coal texture analysis. Hard coal (undeformed and cataclastic coal) exhibited higher DEN, GR, lower CAL, and lower AC than soft coal. The accuracy rate of coal texture identification was highest (97%) when the linear kernel function was applied, and the maximum training accuracy rate was achieved when the penalty parameter value of the linear kernel increased to 1. The results of verification with a newly cored CBM exploration well indicated that the SVM-based identification method was effective for coal texture analysis. With the increasing availability of data, this method can be used to distinguish hard and soft coal in a coal-bearing basin under numerous sample learning conditions.

‘SMART’ Manufacturing Method of Oleochemical Plants by Determining Fatty Acid Composition of C12 of Palm Kernel Stearin using Fuzzy Logic System

Industry 4.0 is the name given to the latest evolution in the digitalisation and automation of manufacturing processes. Transitioning to Industry 4.0 is an emerging imperative for chemical processing based manufacturers throughout the world. “Smart” equipments, products and factories are transforming how companies do business, both internally and with suppliers and customers. At the present moment palm oil products particularly, Palm Kernel Oil have become a major feedstock for the oleochemical industries in Malaysia. Since Malaysia accounts for 42% of the global production of palm oil products, it is inevitable that the oleochemical industries has been flourishing and blooming in the last few years. The majority fatty acid found in palm kernel stearin is C12 which can be found in range of 54.8 to 64.0 wt%. But this value is determined by the iodine and saponification value of the palm kernel. The higher the iodine and saponification value, hence the higher the fatty acid composition can be found for higher extraction for oleochemical plant. The iodine value (Wjis) is ranged from 5.8 to 8.1 whereas the saponification value is ranged from 245 to 254 mgKOHg&lt;sup&gt;-1&lt;/sup&gt;oil. Fuzzy Logic system is developed to determine the fatty acid composition for the different range of iodine value and saponification value to ensure kernel with different composition of fatty acid to be separated and produce higher purity of oleochemical substances. This system will ensure ‘SMART’ manufacturing replacing the ‘NON-SMART’ methods where all kernels are grinded randomly and extracting altogether.

Effects of Pulsed Magnetic Fields of Different Intensities on Dislocation Density, Residual Stress, and Hardness of Cr4Mo4V Steel

To study the effects of pulsed magnetic fields of different intensities on the dislocation density, residual stress, and hardness of Cr4Mo4V steel, magnetic treatment is conducted at 0, 1.0, 1.3, 1.5, 2.0, and 2.5 T. The dislocation density and residual stress are measured using Electron Backscatter Diffraction (EBSD) and X-ray technique, respectively. The results reveal the dislocation density and compressive residual stress decrease at lower magnetic fields such as 1.0 T and 1.3 T, while they increase at higher magnetic fields such as 2.0 T and 2.5 T. The average value of kernel averaged misorientation (KAM) and compressive residual stress decrease about 10.4% and 15.8%, respectively, at 1.0 T, while they increase about 5.88% and 18.2%, respectively, at 2.5 T. The average value of hardness decreases about 3.5% at 1.0 T, from 817 HV to 787 HV. With the increments of intensities, the hardness of the treated samples increases. The hardness essentially remains unchanged at 2.0 T and 2.5 T. The reason for the dislocation motion under the action of pulsed magnetic fields is discussed.

Generalized RBF kernel for incomplete data

We construct genRBF kernel, which generalizes standard Gaussian RBF kernel to the case of incomplete data. Instead of using typical imputation techniques, which fill missing attributes by single values, we model possible outcomes at missing coordinates using data distribution. This allows to derive analytical formula for the expected value of RBF kernel taken over all possible imputations, which is a basic idea behind our method. In particular, for complete observations genRBF reduces to standard RBF kernel. Experiments show that introduced kernel applied to SVM classifier and regressor gives better results than state-of-the-art methods, especially in the case when large number of features is missing. Moreover, genRBF is easy to implement and can be used together with any kernel approach without any additional modifications.

Gaussian kernels for incomplete data

This paper discusses a method to estimate the expected value of the Gaussian kernel in the presence of incomplete data. We show how, under the general assumption of a missing-at-random mechanism, the expected value of the Gaussian kernel function has a simple closed-form solution. Such a solution depends only on the parameters of the Gamma distribution which is assumed to represent squared distances. Furthermore, we show how the parameters governing the Gamma distribution depend only on the non-central moments of the kernel arguments, via the second-order moments of their squared distance, and can be estimated by making use of any parametric density estimation model of the data distribution. We approximate the data distribution with the maximum likelihood estimate of a Gaussian mixture distribution. The validity of the method is empirically assessed, under a range of conditions, on synthetic and real problems and the results compared to existing methods. For comparison, we consider methods that indirectly estimate a Gaussian kernel function by either estimating squared distances or by imputing missing values and then computing distances. Based on the experimental results, the proposed method consistently proves itself an accurate technique that further extends the use of Gaussian kernels with incomplete data.

Small-scale modification to the lensing kernel

Calculations of the cosmic microwave background (CMB) lensing power implemented into the standard cosmological codes such as camb and class usually treat the surface of last scatter as an infinitely thin screen. However, since the CMB anisotropies are smoothed out on scales smaller than the diffusion length due to the effect of Silk damping, the photons which carry information about the small-scale density distribution come from slightly earlier times than the standard recombination time. The dominant effect is the scale dependence of the mean redshift associated with the fluctuations during recombination. We find that fluctuations at $k=0.01\text{ }\text{ }{\mathrm{Mpc}}^{\ensuremath{-}1}$ come from a characteristic redshift of $z\ensuremath{\approx}1090$, while fluctuations at $k=0.3\text{ }\text{ }{\mathrm{Mpc}}^{\ensuremath{-}1}$ come from a characteristic redshift of $z\ensuremath{\approx}1130$. We then estimate the corrections to the lensing kernel and the related power spectra due to this effect. We conclude that neglecting it would result in a deviation from the true value of the lensing kernel at the half percent level at small CMB scales. For an all-sky, noise-free experiment, this corresponds to a $\ensuremath{\sim}0.1\ensuremath{\sigma}$ shift in the observed temperature power spectrum on small scales ($2500\ensuremath{\lesssim}l\ensuremath{\lesssim}4000$).

Multiple kernels learning-based biological entity relationship extraction method

BackgroundAutomatic extracting protein entity interaction information from biomedical literature can help to build protein relation network and design new drugs. There are more than 20 million literature abstracts included in MEDLINE, which is the most authoritative textual database in the field of biomedicine, and follow an exponential growth over time. This frantic expansion of the biomedical literature can often be difficult to absorb or manually analyze. Thus efficient and automated search engines are necessary to efficiently explore the biomedical literature using text mining techniques.ResultsThe P, R, and F value of tag graph method in Aimed corpus are 50.82, 69.76, and 58.61%, respectively. The P, R, and F value of tag graph kernel method in other four evaluation corpuses are 2–5% higher than that of all-paths graph kernel. And The P, R and F value of feature kernel and tag graph kernel fuse methods is 53.43, 71.62 and 61.30%, respectively. The P, R and F value of feature kernel and tag graph kernel fuse methods is 55.47, 70.29 and 60.37%, respectively. It indicated that the performance of the two kinds of kernel fusion methods is better than that of simple kernel.ConclusionIn comparison with the all-paths graph kernel method, the tag graph kernel method is superior in terms of overall performance. Experiments show that the performance of the multi-kernels method is better than that of the three separate single-kernel method and the dual-mutually fused kernel method used hereof in five corpus sets.