A comparative study of support vector machine and logistic regression article · January 021 citations reads 11 authors

1.3 Classification Tools 
This study focuses on only two classification tools, namely, the support vector machine and 
logistic regression. Both tools are popular machine learning tools and in the subsequent sections, 
we shall explain what they are. 
1.3.1 Support Vector Machine 
Support vector machine (SVM) is a supervised machine learning algorithm for both classification 
and regression problems. When SVM is used for classification, it separates classes through a 
hyperplane that has the maximum distance between them. The distance between the hyperplane 
and the first point of each class is called the margin while the points that are closest to the margin 
are termed the support vectors. It is a concept based on the hyperplane with a strong geometric 
connection that looks at data and sorts it into two categories (Cortes and Vapnik, 1995) explained 
support vector machine as a binary classifier from field of machine learning. SVM has 
implemented satisfactorily to a variety of realistic problems like face detection, bioinformatics, 
text and hypertext categorization, classification of images etc. 
 
 

---

O
RIENTAL JOURNAL OF SCIENCE & ENGINEERING VOL -2, ISS-1, FEB - 2021 
www.ojse.org
ojse©2019 
Page 87 
1.3.1.1 Support vectors, Margin lines and the Margin 
Support vectors are data points that are closer to the hyperplane and influences the position and 
orientation of the hyperplane. Support vectors maximizes the margin of the classifier and removing 
the support vectors will change the position of the hyperplane. In a given separating hyperplane, 
they are unique. A support vector is called a –ve support vector if the vector is from negative class 
and +ve support vector if the vector is from positive class. 
Figure 1: Graphical illustration of support vectors, margin lines and margin. 
Margin lines are hyperplanes that passes through the support vectors. A +ve margin line passes 
through the positive support vectors and the -ve margin line passes through the negative support 
vectors. The distance between the lines and support vectors are called the margin.