Polynomial Regression: Delivered By: Garima Koushik

You might also like

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 11

Polynomial Regression

Delivered by : Garima Koushik


Why Polynomial Regression ?
 What if data points are scattered like this?
 After applying the Linear Regression model, it will look
something like this –
Polynomial Regression
 It is same as Linear Regression but it is used for non linear
distribution of the data.
 After applying the Polynomial Regression technique on the
same data , the best fit line will be like as below-
Under fitting
 It occurs when a straight line is unable to capture the pattern
in data.
 To overcome under fitting, we increase the complexity of
model.
 The linear model,

 Can be transferred to
 After fitting cubic curve i.e. Degree = 3
Over fitting
 When degree of the model is set to a very high number than
model starts capturing the noise in the data, it is called as
over fitting.
Bias and Variance
 Bias refers to the error due to the model’s simplistic
assumptions in fitting the data. A high bias means that the
model is unable to capture the patterns in the data and this
results in under-fitting.
 Variance refers to the error due to the complex model
trying to fit the data. High variance means the model passes
through most of the data points and it results in over-
fitting the data.
Which one is correct model?
Outcome
 Machine learning model with low variance and low bias
 As complexity increases, bias will also decreases but variance
increases
 It becomes difficult to have both low bias and variance
 So, we try to achieve good model that will perform well on
train and unseen data.

You might also like