Title Regression-1 (1).pptx ✅

MACHINE LEARNING Dr.S.SAUDIA Assistant Professor CITE, M.S.University # REGRESSION CONCEPTS
# Contents REGRESSION CONCEPTS Regression concepts get applied in supervised kind of Machine Learning Algorithms. The concepts are helpful in situations: When prediction of a dependent variable value is to be made based on the linear relationship between the dependent variable values and the independent variable values. When a labelled dataset available for training the algorithm. When the output variable values are expected to be in a continuous range. Course Teacher: Dr.S.Saudia, AP, CITE, MSU
# REGRESSION CONCEPTS The linear regression relationship model/ regression line is arrived by minimizing the error between actual value of the dependent variable and the value predicted by the proposed regression model. These errors are called Residuals. The minimization technique followed is the Least Squares Technique. The best fit line obtained from the Least Squares Technique is called Regression Line. Residual analysis is already explained in lecture 14. Course Teacher: Dr.S.Saudia, AP, CITE, MSU
Least Squares Method. Steps: Residue is the distance between a data point and a particular line function drawn through the scatterplot. The datapoint should have been on the line. But it is a little distance away from the line. This contributes to the Error of the line function with respect to the data point or the Error between the actual value and the predicted value. Error/ Residue =y- f(x) So Residues are Errors in such situations. The Errors due to all the data points on the line are determined, squared and added up. This Sum of Squares is found between all the data points and all possible line functions. The line function corresponding to the least sum of Squares is the Best Fit Line for the data points in the scatter plot or it is the predictive model to predict the value of the dependent variable for any value f the independent variable value. So this method of finding the Best Fit line is called the Least Squares Method. Course Teacher:Dr.S.Saudia, AP, CITE, MSU. #