Simple Approach to (Without) SVM Algorithm (Create Hyperplane Base Regression Of Closest Pair)

Hi All Student, we apologize for the delay in posting “Sentiment Analysis Document Using Support Vector Machines (SVMs) With Linier & Polynomial Kernel Without Matlab ToolBox” . But it’s okey. 😀

In this week we will provide an overview of a technique which it’s think is a very simple approach to be implemented in making comparisons with the results hyperplane formed of Support Vector Machine (SVM) on linear data to separate the two classes (binary classification), based Linear Regression method on nearest points (Closest Pair) is formed of two points between classes to take its midpoint. Obviously this method is very vulnerable to errors in the formation of the hyperplane line, because they do not have a concept of Support Vector, in the sense that all points are considered all result of midpoint or it’s just take few of midpoint.

This is the linear regression formula (y = a + b*x) :

Tracking animation base Closest Pair :

Hyperplane_Data_Random_Base_Regression_Closest_Pair

View results part 1 :

Hyperplane Data Random Base Regression (Closest Pair)_1

View results part 2 :

Hyperplane Data Random Base Regression (Closest Pair)_2

View results part 3 :

Hyperplane Data Random Base Regression (Closest Pair)_3

View results part 4 :

Hyperplane Data Random Base Regression (Closest Pair)_4

You can download This Matlab Code All About “Simple_Approach_Hyperplane_Of_SVM_Base_Regression_Closest_Pair” at  (Simple_Approach_Hyperplane_Of_SVM_Base_Regression_Closest_Pair.zip). Note : “After Download it, To Extract File (Simple_Approach_Hyperplane_Of_SVM_Base_Regression_Closest_Pair.zip), You must Rename Extension *.doc to *.zip”

To Running the program, double click Simple_Approach_Hyperplane_Of_SVM_Base_Regression_Closest_Pair.m file.  Enjoy with matlab code, especially for your research.

Any Suggestions, Question and Other, Send to My Email : matlabfreecode@gmail.com
(CMIIW & PMIIW).