Support Vector Machine: Principles, Parameters, and Applications

Gholami, Raoof; Fakhari, Nikoo

doi:10.1016/b978-0-12-811318-9.00027-2

Cited by 199 publications

(140 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It discriminates the data by a hyperplane that fairly separates two classes. Along with the linear classification, SVM can also perform non-linear classification by mapping the input to n-dimensional space where coordinates represent the value of n (number of features) [10] [24].…”

Section: Support Vector Machine(svm)mentioning

confidence: 99%

Perinatal Hypoxia Diagnostic System by using Scalable Machine Learning Algorithms

Kaur*¹,

Khullar²,

Singh³

et al. 2019

IJITEE

View full text Add to dashboard Cite

Collaborating big data and machine learning approaches in healthcare can help in improving clinical decision making and treatment by identifying and accumulating accurate features. Prenatal hypoxia can also be identified by cardiotocography (CTG) monitoring that helps in identifying the condition of the fetus. Imposing the data over distributed approaches can help in fast computation to rate the fetal and mother wellbeing before delivery. Our research aims to propose and implement a scalable Machine learning Algorithm based perinatal Hypoxia diagnostic system for larger datasets. This system was implemented on the CTG dataset using python and pyspark models like SVM, Random Forest, and Logistic regression. In the proposed method experiment results contributing to spark RF are more accurate than other techniques and achieved the precision of 0.97, recall of 0.99, f-1 score of 0. 98, AUC of 0.97 and gained 97% accuracy

show abstract

Section: Support Vector Machine(svm)mentioning

confidence: 99%

Perinatal Hypoxia Diagnostic System by using Scalable Machine Learning Algorithms

Kaur*¹,

Khullar²,

Singh³

et al. 2019

IJITEE

View full text Add to dashboard Cite

show abstract

“…Support Vector Machine was introduced after in the 1990s and used for many engineering application [23]. Support Vector Machine is an algorithm developed for binary classification by Cortes & Vapnik.…”

Section: Support Vector Machine Based Classificationmentioning

confidence: 99%

“…In non-linear form classification kernel trick, it mapped the input from low dimensional feature space to high dimension feature space. Support Vector Machine algorithm provides a solution for a limited number of training data in more time and they consume more time for large databases [23]. It is used for text and hypertext categorization, classification of images, image segmentation, and hand-written recognition of character.…”

Section: Support Vector Machine Based Classificationmentioning

confidence: 99%

Empirical Performance Analysis of Decision Tree and Support Vector Machine based Classifiers on Biological Databases

Amjad¹,

Ali²,

Rafiq³

et al. 2019

IJACSA

View full text Add to dashboard Cite

The classification and prediction of medical diseases is a cutting edge research problem in the medical field. The experts of machine learning are continuously proposing new classification methods for the prediction of diseases. The discovery of classification rules from medical databases for classification and prediction of diseases is a challenging and nontrivial task. It is very significant to investigate the more promising and efficient classification approaches for the discovery of classification rules from the medical databases. This paper focuses on the problem of selection of more efficient, promising and suitable classifier for the prediction of specific diseases by performing empirical studies on bunch mark medical databases. The research work under the focus concentrates on the benchmark medical data sets i.e. arrhythmia, breast-cancer, diabetes, hepatitis, mammography, lymph, liver-disorders, sick, cardiotocography, heart-statlog, breast-w, and lung-cancer. The medical data sets are obtained from the open-source UCI machine learning repository. The research work will be investigating the performance of Decision Tree (i.e. AdaBoost.NC, C45-C, CART, and ID3-C) and Support Vector Machines. For experimentation, Knowledge Extraction based on Evolutionary Learning (KEEL), a data mining tool will be used. This research work provides the empirical performance analysis of decision tree-based classifiers and SVM on a specific dataset. Moreover, this article provides a comparative performance analysis of classification approaches in terms of statistics.

show abstract

“…The exact relationships between predictions and outcomes are unknown, hence selecting an optimal kernel function presents a challenge. There are no clear rules for selecting a single optimal kernel, but cross-validation is usually implemented[13, 14,15]. An interesting property of kernels is that a linear com-100 bination of two kernel functions results in another kernel function [11].…”

mentioning

confidence: 99%

Fenchel duality of Cox partial likelihood and its application in survival kernel learning

Wilson

Sun

et al. 2020

Preprint

View full text Add to dashboard Cite

The Cox proportional hazard model is the most widely used method in modeling time-to-event data in the health sciences. A common form of the loss function in machine learning for survival data is also mainly based on Cox partial likelihood function, due to its simplicity. However, the optimization problem becomes intractable when more complicated regularization is employed with the Cox loss function. In this paper, we show that a convex conjugate function of Cox loss function based on Fenchel Duality exists, and this provides an alternative framework to optimization based on the primal form. Furthermore, the dual form suggests an efficient algorithm for solving the kernel learning problem with censored survival outcomes. We illustrate the application of the derived duality form of Cox partial likelihood loss in the multiple kernel learning setting to their flexibility and efficiency [2]. Cox PH models are the most widely used model in health and clinical sciences, while the AFT model is more popular in 5 areas like engineering. The Cox model is a semi-parametric regression method where no assumptions imposed on the baseline hazard function. The parametric regression coefficients quantify the effect size of each covariate and the exponential of the coefficient is interpreted as the unit increase of the hazard ratio. The Cox model works well in practice because it can tolerate a modest 10 deviation from the PH assumption. The partial likelihood in the Cox model is defined as the probability that one individual will experience the event at a time t among those who have survived longer than t. It was shown that maximizing partial likelihood provides an asymptotically efficient estimation of regression coefficients [3]. The Cox model has been successfully extended to various high-15 dimensional settings, where the number of features is more than the number of samples. Additionally, the log partial likelihood (LPL) is differentiable and convex. Thus the combination of Cox LPL, and l1 (lasso), l2 (ridge), or elastic net penalties can be directly solved by standard Newton-Rapshon method. Li and Luan [4] pioneered methods for kernel Cox regression in the framework the pe-20 nalization framework from the view of function estimation in reproducing kernelHilbert spaces. Furthermore, the convex combined loss function often guarantees the convergence of efficient optimization algorithms including coordinate descent [5], which is the core algorithm implemented in a popular R package for penalized regression glmnet. The Cox LPL has also been adopted in many ma-25 chine learning approaches as a loss function to the survival setting. For example, Ridgeway [6] adapted the gradient boosting method for the Cox model, which is implemented in the R package gbm. Li and Luan [7] also considered a boosting procedure using smoothing splines to estimate the proportional hazards models. More recently, the application of neural network-based deep learning techniques 30 to the Cox PH model has begun to receive attention [8]. These machine lear...

show abstract

Support Vector Machine: Principles, Parameters, and Applications

Cited by 199 publications

References 24 publications

Perinatal Hypoxia Diagnostic System by using Scalable Machine Learning Algorithms

Perinatal Hypoxia Diagnostic System by using Scalable Machine Learning Algorithms

Empirical Performance Analysis of Decision Tree and Support Vector Machine based Classifiers on Biological Databases

Fenchel duality of Cox partial likelihood and its application in survival kernel learning

Contact Info

Product

Resources

About