September 24, 2018

Sreekanth B

Hewlett Packard Data Science Recently Asked Interview Questions Answers

Explain cross-validation.?

It is a model validation technique for evaluating how the outcomes of a statistical analysis will generalize to an independent data set. It is mainly used in backgrounds where the objective is forecast and one wants to estimate how accurately a model will accomplish in practice. The goal of cross-validation is to term a data set to test the model in the training phase (i.e. validation data set) in order to limit problems like overfitting and gain insight on how the model will generalize to an independent data set.

What is Collaborative Filtering?

The process of filtering used by most recommender systems to find patterns and information by collaborating perspectives, numerous data sources, and several agents.

Do gradient descent methods at all times converge to a similar point?

No, they do not because in some cases they reach a local minima or a local optima point. You would not reach the global optima point. This is governed by the data and the starting conditions.

What is the goal of A/B Testing?

This is a statistical hypothesis testing for randomized experiments with two variables, A and B. The objective of A/B testing is to detect any changes to a web page to maximize or increase the outcome of a strategy.
Hewlett Packard Data Science Recently Asked Interview Questions Answers
Hewlett Packard Data Science Recently Asked Interview Questions Answers

What are the drawbacks of the linear model?

Some drawbacks of the linear model are:

The assumption of linearity of the errors.
It can’t be used for count outcomes or binary outcomes
There are overfitting problems that it can’t solve

What is the Law of Large Numbers?

It is a theorem that describes the result of performing the same experiment a large number of times. This theorem forms the basis of frequency-style thinking. It says that the sample mean, the sample variance and the sample standard deviation converge to what they are trying to estimate.

What are confounding variables?

These are extraneous variables in a statistical model that correlate directly or inversely with both the dependent and the independent variable. The estimate fails to account for the confounding factor.

In a class of 30 students, approximately what is the probability that two of the students have their birthday on the same day (defined by same day and month) (assuming it’s not a leap year)?

For example – Students with birthday 3rd Jan 1993 and 3rd Jan 1994 would be a favorable event.

A) 49%

B) 52%

C) 70%

D) 35%

Ans: (C)

The total number of combinations possible for no two persons to have the same birthday in a class of 30 is 30 * (30-1)/2 = 435.

Now, there are 365 days in a year (assuming it’s not a leap year). Thus, the probability of people having a different birthday would be 364/365. Now there are 870 combinations possible. Thus, the probability that no two people have the same birthday is (364/365)^435 = 0.303.

Thus, the probability that two people would have their birthdays on the same date would be 1 – 0.303 = 0.696

Ahmed is playing a lottery game where he must pick 2 numbers from 0 to 9 followed by an English alphabet (from 26-letters). He may choose the same number both times.

If his ticket matches the 2 numbers and 1 letter drawn in order, he wins the grand prize and receives $10405. If just his letter matches but one or both of the numbers do not match, he wins $100. Under any other circumstance, he wins nothing. The game costs him $5 to play. Suppose he has chosen 04R to play.

What is the expected net profit from playing this ticket?

A) $-2.81

B) $2.81C) $-1.82

C) $-1.82

D) $1.82

Ans: (B)

Expected value in this case

E(X) = P(grand prize)*(10405-5)+P(small)(100-5)+P(losing)*(-5)

P(grand prize)=  (1/10)*(1/10)*(1/26)

P(small) = 1/26-1/2600, the reason we need to do this is we need to exclude the case where he gets the letter right and also the numbers rights. Hence, we need to remove the scenario of getting the letter right.

P(losing ) = 1-1/26-1/2600

Therefore we can fit in the values to get the expected value as $2.81

Assume you sell sandwiches. 70% people choose egg, and the rest choose chicken. What is the probability of selling 2 egg sandwiches to the next 3 customers?

A) 0.343
B) 0.063
C) 0.147
D) 0.027

Ans: (C)

The probability of selling Egg sandwich is 0.7 & that of a chicken sandwich is 0.3. Now, the probability that next 3 customers would order 2 egg sandwich is 0.7 * 0.7 *0.3 = 0.147. They can order them in any sequence, the probabilities would still be the same.

Explain star schema.?

It is a traditional database schema with a central table. Satellite tables map IDs to physical names or descriptions and can be connected to the central fact table using the ID fields; these tables are known as lookup tables and are principally useful in real-time applications, as they save a lot of memory. Sometimes star schemas involve several layers of summarization to recover information faster.

How regularly must an algorithm be updated?

You will want to update an algorithm when:

You want the model to evolve as data streams through infrastructure

The underlying data source is changing

There is a case of non-stationarity

What is Data Science? Also, list the differences between supervised and unsupervised learning.?

Data Science involves using automated methods to analyze massive amounts of data and to extract knowledge from them. By combining aspects of statistics, computer science, applied mathematics, and visualization, data science can turn the vast amounts of data the digital age generates into new insights and new knowledge.

Supervised Learning vs Unsupervised Learning
Supervised Learning

Unsupervised Learning

1. Input data is labeled. 1. Input data is unlabeled.
2. Uses training dataset. 2. Uses the input data set.
3. Used for prediction. 3. Used for analysis.
4. Enables classification and regression. 4. Enables Classification, Density Estimation, & Dimension Reduction

What are feature vectors?

A feature vector is an n-dimensional vector of numerical features that represent some object. In machine learning, feature vectors are used to represent numeric or symbolic characteristics, called features, of an object in a mathematical, easily analyzable way.

Explain the steps in making a decision tree.?

Take the entire data set as input.
Look for a split that maximizes the separation of the classes. A split is any test that divides the data into two sets.
Apply the split to the input data (divide step).
Re-apply steps 1 to 2 to the divided data.
Stop when you meet some stopping criteria.
This step is called pruning. Clean up the tree if you went too far doing splits.

Subscribe to get more Posts :