This document discusses support vector machines (SVM) for classification. It explains that SVM finds a hyperplane that separates classes by maximizing the margin between them. It provides examples of linear, polynomial, Gaussian, and sigmoid SVM kernels. It also discusses tuning SVM parameters like gamma and C, and the pros and cons of SVM, including that it works well with clear margins but does not perform as well on large or noisy datasets. The document is presented by Manish and provides an introduction to SVM for classification.

1. Data Science & Machine Learning Prepared by –
Manish
IT.mnchoudhary@gmail.com
Data Science & Machine Learning
Step by Step : Support Vector Machine
Date 28th
Nov

2. Data Science
Prepared by Manish IT.mnchoudhary@gmail.com
Classification Model : SVM - Linear
In Python, what is the class used to create a logistic regression classifier ?
•
SVM Linear
Linearly separate the data points
Support Vector Machine” (SVM) is a supervised machine learning algorithm which can be used
for both classification or regression challenges. However, it is mostly used in classification
problems. In this algorithm, we plot each data item as a point in n-dimensional space (where n is
number of features you have) with the value of each feature being the value of a particular coordina
Then, we perform classification by finding the hyper-plane that differentiate the two classes very
well (look at the below snapshot).

3. Data Science
Prepared by Manish IT.mnchoudhary@gmail.com
Classification Model : SVM - Linear
In Python, what is the class used to create a logistic regression classifier ?
•

4. Data Science
Prepared by Manish Email :
Classification Model : SVM : Kernel
In Python, what is the class used to create a logistic regression classifier ?
•
Linear Polynomial
Gaussian (RBF) Sigmoid

5. Data Science
Prepared by Manish Email :
Classification Model : SVC - Kernel
In Python, what is the class used to create a logistic regression classifier ?
•

6. Data Science
Prepared by Manish IT.mnchoudhary@gmail.com
Classification Model : SVM - RBF
In Python, what is the class used to create a logistic regression classifier ?
•

7. Data Science
Prepared by Manish IT.mnchoudhary@gmail.com
Classification Model : SVM
In Python, what is the class used to create a logistic regression classifier ?
•
How to tune Parameters of SVM?
sklearn.svm.SVC(C=1.0, kernel='rbf', degree=3, gamma=0.0, coef0=0.0, shrinking=True,
probability=False,tol=0.001, cache_size=200, class_weight=None, verbose=False, max_iter=-1,
random_state=None)
gamma: Kernel coefficient for ‘rbf’, ‘poly’ and ‘sigmoid’. Higher the value of gamma, will try to
exact fit the as per training data set i.e. generalization error and cause over-fitting problem.
svc = svm.SVC(kernel='rbf', C=1,gamma=0).fit(X, y)

8. Data Science
Prepared by Manish IT.mnchoudhary@gmail.com
Pros & Cons: SVM
In Python, what is the class used to create a logistic regression classifier ?
•
Advantage and Disadvantage associated with SVM
Advantage:
oIt works really well with clear margin of separation
oIt is effective in high dimensional spaces.
oIt is effective in cases where number of dimensions is greater than the number of samples.
oIt uses a subset of training points in the decision function (called support vectors), so it is
oalso memory efficient.
Disadvantage:
oIt doesn’t perform well, when we have large data set because the required training time is higher
oIt also doesn’t perform very well, when the data set has more noise i.e. target classes are overla
oSVM doesn’t directly provide probability estimates, these are calculated using an expensive
ofive-fold cross-validation. It is related SVC method of Python scikit-learn library.