Fast SVM Research Articles

Physicochemical parameters combined flash GC e-nose and NIR for quality and volatile characterization of Zhenjiang vinegar with different aging time

In order to clarify the scientific connotation of the processing adjuvant vinegar of “the aging the better”, the physicochemical indexes, antioxidant activity in vitro, odor characteristics and NIR spectrum characteristics of Zhenjiang vinegar with different aging time were comprehensively studied, and the compositional characterization by UPLC-MS. The results showed that the physicochemical quality of Zhenjiang vinegar increased rapidly during the 2–3 years, entered a rising stable period during the 5–8 years and entered a sudden increase period during the 8–10 years of fermentation. Overall, it was in line with the trend that “the aging the better”. The study further characterized the odor changes of different aging time samples, established a fast and non-damage SVM discriminant model based on NIR spectral characteristics (accuracy 100 %), and 753 components were characterized by UPLC-MS, in order to provide a reference for the quality evaluation and standard improvement of Zhenjiang vinegar and processing adjuvant.

Multimodal artificial intelligence-based pathogenomics improves survival prediction in oral squamous cell carcinoma

In this study, we aimed to develop a novel prognostic algorithm for oral squamous cell carcinoma (OSCC) using a combination of pathogenomics and AI-based techniques. We collected comprehensive clinical, genomic, and pathology data from a cohort of OSCC patients in the TCGA dataset and used machine learning and deep learning algorithms to identify relevant features that are predictive of survival outcomes. Our analyses included 406 OSCC patients. Initial analyses involved gene expression analyses, principal component analyses, gene enrichment analyses, and feature importance analyses. These insights were foundational for subsequent model development. Furthermore, we applied five machine learning/deep learning algorithms (Random Survival Forest, Gradient Boosting Survival Analysis, Cox PH, Fast Survival SVM, and DeepSurv) for survival prediction. Our initial analyses revealed relevant gene expression variations and biological pathways, laying the groundwork for robust feature selection in model building. The results showed that the multimodal model outperformed the unimodal models across all methods, with c-index values of 0.722 for RSF, 0.633 for GBSA, 0.625 for FastSVM, 0.633 for CoxPH, and 0.515 for DeepSurv. When considering only important features, the multimodal model continued to outperform the unimodal models, with c-index values of 0.834 for RSF, 0.747 for GBSA, 0.718 for FastSVM, 0.742 for CoxPH, and 0.635 for DeepSurv. Our results demonstrate the potential of pathogenomics and AI-based techniques in improving the accuracy of prognostic prediction in OSCC, which may ultimately aid in the development of personalized treatment strategies for patients with this devastating disease.

Fast homomorphic SVM inference on encrypted data

Fast homomorphic SVM inference on encrypted data

Fast multi-label SVM training based on approximate extreme points

Fast multi-label SVM training based on approximate extreme points

Automation of image categorization with most relevant negatives

Image categorization requires the algorithm to be learned in order to obtain the efficient categorization. The algorithm used for image categorization may misclassify images that are visually similar to the positive ones. Generally, sampling negatives is done at random. In this paper, we have improved Negative Bootstrap in an efficient way to obtain most relevant negatives. To obtain most misclassified visually similar images in a faster way, fast intersection kernel SVM is generalized and used for classification. The accuracy of classified visual concepts is obtained by using the performance metrics. Several different metrics have been used to show the accuracy of relevant negatives. Manual labeling of negatives could be avoided by using the efficient negative bootstrap algorithm.

Fast Extended One-Versus-Rest Multi-Label Support Vector Machine Using Approximate Extreme Points

Existing extended one-versus-rest multi-label support vector machine (OVR-ESVM) adopting non-linear kernel is seriously restricted by excessive training time when it is applied to large-scale data set. In order to overcome this problem, we improve the OVR-ESVM by introducing the principle of approximate extreme points and new approximate ranking loss to construct a novel extended OVR-ESVM using approximate extreme points (AEML-ESVM). By optimizing only on the representative set which can be acquired via adopting the approximate extreme points method, the AEML-ESVM classification algorithm can substantially shorten the training time and its classification performance is comparable to that of the OVR-ESVM classification algorithm. And it uses the new approximate ranking loss as empirical loss term to exploit label correlation of individual instance directly. Experimental study on three benchmark large-scale data sets illustrates that AEML-ESVM classification algorithm can reduce training time greatly and achieve comparable classification performance with OVR-ESVM classification algorithm. And it is also superior to the existing fast multi-label SVM classification algorithms in terms of classification performance and training time.

Sleeping Laboratory: Intelligent Glass And Sensor Technology In Real Time Medical Environment.

Sleep studies are used to evaluate and diagnose a variety of sleep problems. A sleep study is very potential in controlling neurological, breathing and movement disorders. Some of the problems that may lead doctor to order a sleep study are snoring, sleep apnea, sleep walking, bedwetting, sleep talking, the feeling of being tired all day, sleeping during a day. A sleeping laboratory can detect these interruptions, analyse and record detailed information to help doctor diagnose problem and determine the best treatment. Light therapies and simulation of day-night environment have been recognized that can stabilize the patterns of delayed sleep phase syndrome and advanced sleep phase syndrome. In this paper we present sleeping laboratory based on smart glass controlled automatically by sensors, ccd camera and real-time pattern recognition environment. We present sensor network and real-time multiparametric algorithms sending signals to intelligent glass controller. For classification we used fast SVM classifier with new kernel method. Historical medical database is used as training set for automated classifier.

FAST SVM BASED 3 PHASE CASCADED FIVE LEVEL INVERTER

Introduction of nearest three vector algorithm is a major achievement in the area of space vector technology. Complexity and severe computations are still the drawbacks of SVM methods mainly for multilevel inverter applications. A fast SVM technique is introduced in this project which allows the calculation of switch time duration and the efficient determination of switching times based on the two level inverter scheme. SVM modulating waves are generated based on the two level system and then this modulating waves are compared with required number of carrier signals in order to generate the switching pulses for the inverter. Four triangular carrier signals are needed for a five level system in order to generate the switching pulses. Coordinates of the nearest three voltage vectors is not needed, so the complexity of the SVM technique can be reduced and it is the major advantage of the proposed technique compared with conventional SVM techniques used for multilevel inverters. A three phase five level cascaded H bridge topology is used here to verify the effectiveness of the proposed technique. MATLAB simulation and hardware implementation of the proposed system is done. From the analysis of both simulation and hardware it is clear that proposed SVM technique have more fundamental output and less THD than sinusoidal PWM technique.

Successive overrelaxation for laplacian support vector machine.

Semisupervised learning (SSL) problem, which makes use of both a large amount of cheap unlabeled data and a few unlabeled data for training, in the last few years, has attracted amounts of attention in machine learning and data mining. Exploiting the manifold regularization (MR), Belkin et al. proposed a new semisupervised classification algorithm: Laplacian support vector machines (LapSVMs), and have shown the state-of-the-art performance in SSL field. To further improve the LapSVMs, we proposed a fast Laplacian SVM (FLapSVM) solver for classification. Compared with the standard LapSVM, our method has several improved advantages as follows: 1) FLapSVM does not need to deal with the extra matrix and burden the computations related to the variable switching, which make it more suitable for large scale problems; 2) FLapSVM’s dual problem has the same elegant formulation as that of standard SVMs. This means that the kernel trick can be applied directly into the optimization model; and 3) FLapSVM can be effectively solved by successive overrelaxation technology, which converges linearly to a solution and can process very large data sets that need not reside in memory. In practice, combining the strategies of random scheduling of subproblem and two stopping conditions, the computing speed of FLapSVM is rigidly quicker to that of LapSVM and it is a valid alternative to PLapSVM.

Efficient Divide-and-Conquer Classification Based on Parallel Feature-Space Decomposition for Distributed Systems

This study presents a divide-and-conquer (DC) approach based on feature space decomposition for classification. When large-scale datasets are present, typical approaches usually employed truncated kernel methods on the feature space or DC approaches on the sample space. However, this did not guarantee separability between classes, owing to overfitting. To overcome such problems, this work proposes a novel DC approach on feature spaces consisting of three steps. Firstly, we divide the feature space into several subspaces using the decomposition method proposed in this paper. Subsequently, these feature subspaces are sent into individual local classifiers for training. Finally, the outcomes of local classifiers are fused together to generate the final classification results. Experiments on large-scale datasets are carried out for performance evaluation. The results show that the error rates of the proposed DC method decreased comparing with the state-of-the-art fast SVM solvers, e.g., reducing error rates by 10.53% and 7.53% on RCV1 and covtype datasets respectively.

WSN Localization Algorithm Based on Fast SVM

Localization of sensor nodes is essential for wireless sensor network when it is applied to the special applications.This paper proposed a rang-free localization algorithm based on SVM. In this algorithm, multi-class SVMs are applied. So to improve the performance, a fast SVM algorithm is proposed in this paper.Finally, the experimental results demonstrate the algorithm proposed has small localization error, and it is robust and stable.

Research and Simulation on The Fast Incremental SVM Intrusion Detection Algorithm Based on Adjacent Boundary Area

Research and Simulation on The Fast Incremental SVM Intrusion Detection Algorithm Based on Adjacent Boundary Area

English

Recently, Gene expression profiling by microarray technique has been effectively utilized for classification and diagnostic guess- ing of cancer nodules in the field of medical sciences. But the techniques used for cancer classification is still in its lower level. There are various drawbacks in the existing classification techniques such as low testing accuracy, high training time, unreliability, etc. Moreover, mi- croarray data consists of a high degree of noise. Gene ranking techniques such as T-Score, ANOVA, etc are later proposed to overcome those problems. But those approaches will sometimes wrongly predict the rank when large database is used. To overcome these issues, this paper mainly focuses on the development of an effective feature selection and classification technique for microarray gene expression cancer diagnosis for provide significant accuracy, reliability and less error rate. In this paper, Wrapper feature selection approach called the GA-FSVML approach is used for the effective feature selection of genes. In FSVML, the RBF kernel function in SVM is trained using modi- fied Levenberg Marquadt algorithm. This approach proposes a Fast SVM Learning (FSVML) technique for the classification tasks. The ex- periment is performed on lymphoma data set and the result shows the better accuracy of the proposed FSVML with GA-FSVML classifica- tion approach when compared to the standard existing approaches.

Cycle-breaking acceleration of SVM training

Fast SVM training is an important goal for which many proposals have been given in the literature. In this work we will study from a geometrical point of view the presence, in both the Mitchell–Demyanov–Malozemov (MDM) algorithm and Platt's Sequential Minimal Optimization, of training cycles, that is, the repeated selection of some concrete updating patterns. We shall see how to take advantage of these cycles by partially collapsing them in a single updating vector that gives better minimizing directions. We shall numerically illustrate the resulting procedure, showing that it can lead to substantial savings in the number of iterations and kernel operations for both algorithms.

Comments on the "Core Vector Machines: Fast SVM Training on Very Large Data Sets"

In a recently published paper in JMLR, Tsang et al. (2005) present an algorithm for SVM called Core Vector Machines (CVM) and illustrate its performances through comparisons with other SVM solvers....

Fast SVM training algorithm with decomposition on very large data sets

Training a support vector machine on a data set of huge size with thousands of classes is a challenging problem. This paper proposes an efficient algorithm to solve this problem. The key idea is to introduce a parallel optimization step to quickly remove most of the nonsupport vectors, where block diagonal matrices are used to approximate the original kernel matrix so that the original problem can be split into hundreds of subproblems which can be solved more efficiently. In addition, some effective strategies such as kernel caching and efficient computation of kernel matrix are integrated to speed up the training process. Our analysis of the proposed algorithm shows that its time complexity grows linearly with the number of classes and size of the data set. In the experiments, many appealing properties of the proposed algorithm have been investigated and the results show that the proposed algorithm has a much better scaling capability than Libsvm, SVMlight, and SVMTorch. Moreover, the good generalization performances on several large databases have also been achieved.