How can we tell if two users on Instagram are best friends? (answer question) Ok, how can we use this algorithm in the product?
Analytical Scientist Interview Questions
363 analytical scientist interview questions shared by candidates
We have two types of reviewers: careful reviewer (80% of reviewers) and lazy reviewers (20% of reviewers). Careful reviewers rate a post positive 60% of time and negative 40% of time). Lazy reviewers however rate a post positive 100% of time. A) what is the probability that a random ad is reviewed positively? B) If an ad gets a negative review, what is the probability that it's reviewed by a lazy reviewer? C) If 3 ads are reviewed positively in a row, what is the probability that they are reviewed by a lazy reviewer? D) Some as above with n positively reviewed ads in a row. What happens when n goes to infinity? E) If we have very few labeled data, how can we build a model to distinguish between careful and lazy reviewers?
You have 24 hours, how would you measure what % of FB stories are fake news? How would you refine your process if you have more time?
If the number of Facebook friend requests dropped by 10%, how would you investigate the problem as a data scientist?
You're at a casino with two dice, if you roll a 5 you win, and get paid $10. What is your expected payout? If you play until you win (however long that takes) then stop, what is your expected payout?
How would you analyze a dataset with 1,000,000 observations and 10,000 variables.
You play a $5 gamble that pays off 2:1, you will continue to play the gamble until you win with an unlimited budget. What is the expected value of the game for your strategy?
How would you design a Flow Assay to determine blood sample quality. What are the main components in designing a quality assay.
What was the biggest challenge
Q: where you see yourself after 5 years?
Viewing 11 - 20 interview questions