Page 332 - AI_Ver_3.0_class_11
P. 332

The vertical distance between the observed responses in the dataset and the line of best fit is called the residual
              error (e) as shown in the graph below:

                                     120

                                     100

                                      80

                                     Marks (Y)   60     residual (e)=Observed value -- Predicted Value

                                      40

                                      20

                                       0
                                        0     1     2    3     4     5    6     7    8     9    10
                                                           No. of Hours Studied (X)


              Regression—How good is the Line?
              1.  Linear regression aims to find the best-fitting straight line through the points.
              2.   If data points are closer to the line of best fit (less residual error), it means the correlation between the two
                  variables is higher. That means, the relationship between the two variables is strong.

              3.   The regression line is also called ‘Line of Least Squares of Errors’ because the lower the residual errors, the
                  better.

              4.  Each data point has one residual.


                                                                                     #Digital Literacy
                           Video Session

                     Scan  the  QR  code  or  visit  the  following  link  to  watch  the  video:  Linear  Regression
                     Algorithm

                     https://www.youtube.com/watch?v=E5RjzSK0fvY&t=527s








                             Brainy Fact

                     The least squares regression method was first published by mathematicians Legendre in 1805
                     and Carl Friedrich Gauss in 1809. Both used linear regression to predict the movement of planets
                     around the sun. Gauss later published an improved method in 1821.



              When Regression Analysis is Not Suitable

              It’s important to understand that regression analysis may not always be suitable in certain scenarios:
              ●   No Correlation: If there is no correlation between the variables, meaning they change independently of each
                  other, regression analysis will not yield meaningful insights or predictions.

                    330     Touchpad Artificial Intelligence (Ver. 3.0)-XI
   327   328   329   330   331   332   333   334   335   336   337