Support Vector Classifier, Explained: A Visual Guide with Mini 2D Dataset
CLASSIFICATION ALGORITHM
⛳️ More [Classification ALGORITHM](https://medium.com/@samybaladram/list/classification-algorithms-b3586f0a772c), explained: · [Dummy Classifier](https://towardsdatascience.com/dummy-classifier-explained-a-visual-guide-with-code-examples-for-beginners-009ff95fc86e) · [K Nearest Neighbor Classifier](https://towardsdatascience.com/k-nearest-neighbor-classifier-explained-a-visual-guide-with-code-examples-for-beginners-a3d85cad00e1) · [Bernoulli Naive Bayes](https://towardsdatascience.com/bernoulli-naive-bayes-explained-a-visual-guide-with-code-examples-for-beginners-aec39771ddd6) · [Gaussian Naive Bayes](https://towardsdatascience.com/gaussian-naive-bayes-explained-a-visual-guide-with-code-examples-for-beginners-04949cef383c) · [Decision Tree Classifier](https://towardsdatascience.com/decision-tree-classifier-explained-a-visual-guide-with-code-examples-for-beginners-7c863f06a71e) · [Logistic Regression](https://towardsdatascience.com/logistic-regression-explained-a-visual-guide-with-code-examples-for-beginners-81baf5871505) ▶ [Support Vector Classifier](https://towardsdatascience.com/support-vector-classifier-explained-a-visual-guide-with-mini-2d-dataset-62e831e7b9e9) · [Multilayer Perceptron](https://towardsdatascience.com/multilayer-perceptron-explained-a-visual-guide-with-mini-2d-dataset-0ae8100c5d1c)
"Support Vector Machine (SVM) for classification works on a very basic principle – it tries to find the best line that separates the two classes." But if I hear that oversimplified explanation one more time, I might just scream into a pillow.
While the premise sounds simple, SVM is one of those algorithms packed with mathematical gymnastics that took me an absurd amount of time to grasp. Why is it even called a ‘machine'? Why do we need support vectors? Why are some points suddenly not important? And why does it have to be a straight line – oh wait, a straight hyperplane??? Then there's the optimization formula, which is apparently so tricky that we need another version called the dual form. But hold on, now we need another algorithm called SMO to solve that? What's with all the dual coefficients that scikit-learn just spits out? And if that's not enough, we're suddenly pulling off this magic ‘kernel tricks' when a straight line doesn't cut it? Why do we even need these tricks? And why do none of the tutorials ever show the actual numbers?!
In this article, I'm trying to stop this Support Vector Madness. After hours and hours trying to really understand this algorithm, I will try to explain what's ACTUALLY going on with ACTUAL numbers (and of course, its visualization too) but without the complicated maths, perfect for beginners.
Definition
Support Vector Machines are supervised learning models used mainly for classification tasks, though they can be adapted for regression as well. SVMs aim to find the line that best divides a dataset into classes (sigh…), maximizing the margin between these classes.
"Support vectors" are the data points that lie closest to the line and can actually define that line as well. And, what's with the "Machine" then ? While other Machine Learning algorithms could include "Machine," SVM's naming may be partly due to historical context when it was developed. That's it.
