Currently Empty: ₹0.00
Machine Learning
Supporting Vector Machine
Support Vector Machine (SVM) is a type of machine learning algorithm that helps computers classify or sort things into categories.
Imagine you have two types of fruits–apples and oranges–and you want a computer to tell them apart based on their size and weight. SVM looks at the data and draws a line (or curve) that best separates the two types.
The goal is to find the clearest boundary between the groups, so the computer can make accurate decisions even for new fruits it hasn’t seen before.
If the fruits are hard to separate with a straight line, SVM can even bend the space (using something called a “kernel”) to draw a better boundary.
How SVM is used ? (Step-by-Step)
- Collect Data
Get data labeled, such as emails labeled spam or not spam, then images labeled for example, a cat or a dog. - Feature Selection
Select the aspects considered important from the eye of an expert.
Example: The features could be the number of links, certain words, or attachments in detecting spam email. - Train the SVM Model
The willingness of the algorithm to draw a boundary line (or curve) with great separating power between the categories. - Support Vectors and Margin Discovery
SVM tries to identify the support vectors and maximize the margin. - Test the Model
From new unlabeled data, check whether the model gives the right category. - Predict
SVM automatically checks on which side of the boundary line the new data point lies and makes a prediction accordingly.
SVM is used when you need a machine to make smart decisions-yes or no, spam or not, cat or dog-based on learning patterns from past data.
SVM is often used in:
- Spam email detection
- Face and image recognition
- Medical diagnosis
- Handwriting analysis