Chapter 7 application (7.3&7.4)

Today, I finished Chapter 7 (7.3&7.4) and found something useful for my research.

Summary

SVM: Support Vector Machines

SVMs are a class of powerful, highly flexible modeling techniques.

The SVM regression coefficients minimize

There are several aspects of this equation worth pointing out. First, the use of the cost value effectively regularizes the model to help alleviate the over-parameterized problem. Second, the individual training set data points are required for new predictions. Only a subset of training set data points where , are needed for prediction. Since the regression line is determined using these samples, they are called the support vectors as they support the regression line. Which kernel function should be used? This depends on the problem. Note that some of the kernel functions have extra parameters. These parameters, along with the cost value, constitute the tuning parameters for the model.

K-Nearest Neighbors

Two commonly noted problems are computational time and the disconnect between local structure and the predictive ability of KNN.

Next week, I will continue to read the book and try to finish it before the holiday.