Support Vector Assertion - Practical Machine Learning Tutorial with Python p.22

TL;DR

Support Vector Machines use a decision boundary to classify new points based on their position in relation to the separating hyperplane.

Transcript

what is going on everybody and welcome to the 22nd machine learning with Python tutorial video specifically we're talking about support vector machines and classification in the previous tutorial we talked about vectors magnitude direction and dot product and before that we act we showed the high level intuition of a support vector machine now we'r... Read More

Key Insights

• 😥 Support Vector Machines classify new points by projecting them onto a vector perpendicular to the separating hyperplane.
• 🫥 The classification is determined by the dot product of the projected vector and the weight vector, plus the bias.
• 😚 Support vectors, the data points closest to the decision boundary, are used to derive the equations for classification.
• 🏋️ The equations for positive and negative support vectors are multiplied by the class value to find the weight vector and bias.
• 🏋️ The weight vector and bias are necessary to solve the equation used for classification.
• ❓ Support vectors are crucial in determining the decision boundary and separating hyperplane.
• 😥 The goal is to find the weight vector and bias that correctly classify new data points.

Questions & Answers

Q: How do Support Vector Machines classify new points?

Support Vector Machines classify new points by projecting them onto a vector perpendicular to the separating hyperplane. The resulting dot product, plus the bias, determines the classification.

Q: What happens if the dot product is greater than or equal to zero?

If the dot product of the projected vector and the weight vector, plus the bias, is greater than or equal to zero, the point is classified as positive.

Q: How are support vectors used in the classification process?

Support vectors, the data points closest to the decision boundary, are used to derive the equations for classification. These equations involve the dot product and bias.

Q: What are the constraints in finding the weight vector and bias?

The equations used to identify positive and negative support vectors are multiplied by the class value (1 or -1) to derive the equations for finding the weight vector and bias.

Summary & Key Takeaways

• Support Vector Machines (SVM) classify new data points by projecting them onto a vector perpendicular to the separating hyperplane.

• The classification is determined by whether the dot product of the projected vector and the weight vector, plus the bias, is greater than or equal to zero.

• Support vectors, which are the data points closest to the decision boundary, are used to derive the equations for classification.