Page 281 - Artificial Intellegence_v2.0_Class_11
P. 281

120


                               100


                                80
                              Marks (Y)   60





                                40


                                20


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

                 Observations about the line:

                                  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 vertical distance between the observed responses in the dataset and the line of best fit is called the residual
                    error (e). 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.




                                                                                                 Regression     279
   276   277   278   279   280   281   282   283   284   285   286