Dhiraj, a data scientist and machine learning evangelist, continues his teaching of machine learning algorithms by explaining both through lecture and practice the Support Vector Machine (SVM) algorithm in Python in this video series. Click here to watch all of Dhiraj Kumar’s machine learning videos. Learn all about SVM in this video series covering these seven topics:

**Introducing Support Vector Machines (SVMs)**. This first topic in the Support Vector Machine (SVM) series introduces this machine learning classification algorithm. SVM performs well even with a limited amount of data. Data points are inputed and the output is the hyper plane. The hyper plane is a line that separates the data, and this line is called the decision boundary. We explain how to use SVM with non linear data. Kernel Tricks are also covered.**Support Vector Machine (SVM) Advantages and Disadvantages**. This second topic in the Support Vector Machine (SVM) series covers where SVM works well and where it doesn’t work well. SVM works well with data that has a clear margin, in high dimensional spaces, is very memory efficient, and when the number of dimensions is greater than the number of samples. SVM does not work well with large data sets, with overlapping classes, when the data is non-probabilistic, and when the number of features for each data point exceeds the number of training data samples.**Support Vector Machine (SVM) Regression**. This third topic in the Support Vector Machine (SVM) series explains how to perform regression analysis with the Support Vector Machine (SVM). When the Support Vector Machine (SVM) is used for regression, it is called Support Vector Regression (SVR). SVR does not depend on the dimensionality of the input space. Penalty Factors and epsilons are discussed as well. Python is used to show how to perform regression analysis.**Support Vector Machine (SVM) Classification**. This fourth topic in the Support Vector Machine (SVM) series focuses on the Support Vector Machine (SVM) classifier. The classification concepts of Hyper Plane, Boundary Line, Support Vector, and Kernel are discussed as well. Maximum margin and hard margin are compared, and as with all prior topics, all concepts are demonstrated with Python in the Jupyter notebook.**Support Vector Machine (SVM) Parameter Tuning**. This fifth topic in the Support Vector Machine (SVM) series explains how to tune different parameters of SVMs. The three different parameters are Kernel, Epsilon, and C-Penalty Co-efficient. Python is used to show how to perform parameter tuning.**Support Vector Machine (SVM) Prediction**. This sixth topic in the Support Vector Machine (SVM) series explains how to do prediction after our SVM model is built. We will discuss how to store and share predictions. Python is used to show how to perform prediction.**Support Vector Machine (SVM) Evaluation**. This seventh topic in the Support Vector Machine (SVM) series explains how to evaluate a Support Vector Machine (SVM) model. Once the machine learning model has been evaluated, we can use the feedback to improve the model until our model produces the desired accuracy. We will use a Confusion Matrix in Python to evaluate our Support Vector Machine (SVM) model.

- Introducing Support Vector Machines (SVMs) 00:11:32
- Support Vector Machine (SVM) Advantages and Disadvantages 00:07:53
- Support Vector Machine (SVM) Regression 00:09:01
- Support Vector Machine (SVM) Classification 00:10:05
- Support Vector Machine (SVM) Parameter Tuning 00:10:10
- Support Vector Machine (SVM) Prediction 00:05:17
- Support Vector Machine (SVM) Evaluation 00:05:29