Interview Questions Set 1(21 to 30)

#datascience #statistics

Akash Borgalli Dec 11 2021 · 2 min read
Share this

21. let's suppose I have appeared in 3 interviews, what is the probability that I am able to crack at least 1 interview?

Figure 1

22. Explain Gaussian Distribution in your own way.

Ans:  It is also called Normal Distribution. It's like a bell shape curve when you plot a graph where most of the data lie near the mean meaning center of the data points. It would be symmetrical on right and left sides of the mean.

23. What do you understand by 1st,2nd, and 3rd Standard Deviation from Mean?

Ans: Standard deviation talks about the spread of the data in a distribution with respect to different data. In Normal Gaussian Distribution. It says that within -1 to 1st standard deviation you would be having 68% of data and from -2 to +2 standard deviation you hold 95% of data and last but not least from -3 to +3 you hold 99.7% of data. It is also called as 68-95-99 rule which is also known as the Empirical rule.

Figure 2

24. What do you understand by variance in data in simple words?

Ans: It talks about how far your data point is away from the mean.

25. If the variance of the dataset is too high, in that case How you will be able to handle it or decrease it?

Ans: Bias and Variance both come under reducible errors. This is a type of error that can be reduced and controlled to get higher accuracy. Bias means an error that you see after training the model. Variance is an error after testing the model. To handle high variance you can go for multiple final models instead of one or do early stopping at the time of training.

26. Explain the relationship between Variance and Bias.

Ans: The relationship with respect to datapoints in case of variance where best-fit line holds a good amount of data points its considered as Overfitting model and datapoints that are not close to best-fit line shows bias. In short, it's not trying to understand relations. It's called underfitting of the model.

Figure 3

27. Tell me what kind of graph-based approach I will be able to apply to find out the standardization of Dataset?

Ans: Histogram

28. What do you understand by Z Value given in Z Table?

Ans:  Its tells no. of standard deviations a value is away from the mean. Let's suppose if Z value is positive that means the data point is above the mean whereas if Z value is negative which shows that it is below the mean.

29. Do you know a Standard Normal Distribution Formula?

Ans:  Yes its z = (x-mean) / standard deviation.

30. Can you please explain the critical region in your way?

Ans: This critical region comes under hypothesis testing. Critical regions are regions that are rejection regions that are separated by positive and negative z values which are also known as critical values which are on both sides.

Figure 4
Read next