stud
bud
Sign up
Log in
Home
Library
Explore
SAT
More
Data Science Flashcards | StudBud
Explore
/
Data Science
Data Science
12 terms
·
StudBud Administration
·
25 plays
#FooBar
Flashcards
12 cards
F
Practice quiz
P
Host live
L
Questions
12
Answers
1
What is the most important factor behind the advancement of Machine Learning
Smarter algorithms
Faster computations
Programming languages such as python
Cheaper computations
2
Why is data science difficult?
Because you can't improve all manual tasks
Because it requires the understanding of complex math
It's not...
Because it requires the selection of the right algorithm.
3
What is most helpful in improving prediction accuracy?
Data consistency (lack of empty values etc.)
Domain knowledge
Choice of programming language
Computational speed
4
Which is a valid expression... (least nonsense)
Vertebrates and Invertebrates
Predictive Forecasting
Artificial Intelligence and Machine Learning
Mathematics and Statistics
5
Important types of Machine Learning does not include...
Reinforcement Learning
Supervised and Semi-supervised Learning
Unsupervised Learning
Improvised Learning
6
Machine Learning is.....
a super-set of Robotic Process Automation (RPA)
a sub-set of Artificial Intelligence (AI)
a sub-set of Deep Learning (DL)
a super-set of Natural Language Processing (NLP)
7
What should a data scientist ideally spend most time on?
Selecting the best algorithm
Cleaning data
The toilet
Feature Engineering
8
Which problem can't be solved using any type of Machine Learning
Live translation between two people
Separate vegetables based on ripeness
Identify possible computer network intrusion
Finding the optimal route between a number of cities
9
What is a ROC curve
a curve that plots true positive rate against the true negative rate
the flight path of a mythical bird from the Brothers Grimm stories
a curve visualizing the probability of false alarm
the sensitivity or recall as a function of fall-out
10
Subset sum is
a special case of knapsack problem
not an optimization problem
not solvable in time exponential in n
not NP-hard
11
How much higher was the price of calculating 1 teraFLOPS in Aug 2007 vs. Oct 2017 (2017 USD)
approx. 1900x
approx. 20x
approx. 170x
approx. 5x
12
What is true about KNN and k-means
KNN is an unsupervised classification algorithm
k-means is a supervised clustering algorithm
KNN needs labeled points
They are the same