Hello folks, towards onwards I’m going to start new articles , tutorials , tips and tricks about Machine learning , Data science and related to this field. hope this articles will help you all. today going to add an article about What is Bias and Variance in Machine learning and what the role of these two words.

Lets get started : –

Or just remember this,

**Bias = under-fitting.**

**Variance = over-fitting.**

Most of us who already working or who just started data science – may obviously face the term

BiasandVariance.Here I will try to explain these in simple words, hope you’ll find it as useful.

error(X) = noise(X) + bias(X) + variance(X)In very simple word –

Bias, when we train the model on the data being used for training(Train data), error between predicted value and actual value is termed as Bias.And what is means if bias is high or low, So here it is :

👉 If the Bias is high that means model is not able to predict the output values of train data correctly

And

👉if the Bias is low which means model is able to predict output value of train data correctly or the model is copying all the data.

Variance is opposite of Bias [Variance = Accuracy]“

”Variance, the error between predicted value and actual value of test data or unseen data.and here also we have, low variance and high variance :

👉If the Variance is high which means model is not able to predict the output values of unseen or test data correctly and vice versa.

✅Bias and Variance have negative correlation, that means if the bias is high then variance is low and vice versa.

✅High Variance means model is over fitted which means model has remember all the train data and it is tightly bound on that, so it has low bias or able to predict the correct values on train data.