02 End-End ML Project

Question 1

what is correlation

Answer 1

1. it measures the degree of association or relation between 2 or more variables.
2. it helps to determine the pattern between 2 variables.
   3.value ranges from -1 to +1.

Question 2

what does correlation of -1 mean

Answer 2

it indicates that if 1 variable increases another variable decreases and the relation is strong.

Question 3

what does correlation of +1 mean

Answer 3

it indicates that if 1 variables increases other variable also increases and the relation is strong.

Question 4

what does correlation 0 mean

Answer 4

the relation between both the variables is low.

Question 5

data processing required before calculating correlation

Answer 5

removing of outliers.

Question 6

formula of correlation

Answer 6

cov(x, y)/sdx*sdy

Question 7

what is covariance

Answer 7

it tells us the direction in which both the variables chenge.

Question 8

+ve covariance

Answer 8

both the variables are moving in the same direction

Question 9

-ve covariance

Answer 9

if 1 variable is increasing other variable is decreasing

Question 10

Formula of covariance

Answer 10

sum(Xi-Xmean)-(Yi-Ymean)/(n-1)

Question 11

types of data transformation

Answer 11

1. min-max scaler AKA Standardisation
2. normalization.

Question 12

what is min-max scaler

Answer 12

1. the transformed range between 0 - 1
2. it is highly affected by outliers

Question 13

formula of min-max scaler

Answer 13

(x-xmin)/(xmax-xmin)

Question 14

what is standardization

Answer 14

1. the mean of the transformed value is 0 and the variance will be 1.

Question 15

What are all Regression Model Metrics

Answer 15

1. Mean Sq Error (MSE)
2. Root Mean Sq Error (RMSE)
3. R-Square (R^2)
4. Adjusted R-Square

Question 16

What is MSE?

Answer 16

Mean of the difference between actual and predicted value.

Question 17

what is RMSE?

Answer 17

it the square root of the mean of square of difference between actual and predicted.
if RMSE is 90 than the gap between actual and predicted is 90.

Question 18

What is R^2?

Answer 18

with RMSE it is difficult to understand which model is better for different problems hence we use R^2.
formula = 1-(RSS/TSS) where RSS is Residual Sum of Sq and TSS is Total Sum of Sq.

Question 19

what is adjusted R^2?

Answer 19

R^2 does not change when new variables are added and hence we use adjusted R^2 to understand the performance of model when new features are added.