Support Vector Machines or SVMs may sound intimidating, but they really aren’t. When I first heard about them, I became a little curious and wanted to know more. After few working projects I began to understand their structure and uses and want to share a little of that with you.
Let’s take a simple classification problem, we are trying to classify two different types of cuisine, for simplicity, let’s assume that the two differentiating factors identified are: The ingredients of the food and the method of creating the food e.g. the recipe. Using a simple scatter plot the human eye can easy notice a few distinguishing factors.
Support Vectors are just the co-ordinates of individual observations. A Support Vector Machine is a supervised binary classification algorithm that creates a hyperplane by separating two classes of data by the largest margin and creating groups of similar data distinctive from other groups. It follows a technique called the kernel trick to transform the data and based on these transformations, it finds an optimal boundary between the possible outputs.
Example of Support vector machine:
- Handwritten digit recognition (data)
- Text Classification
- Face Detection
- DNA Sequencing