Description
Question 1 (True or False)
Provide brief explanations for your answers.
(a) Correlation implies causation.
(b) Data scientists do most of their work in Python and are unlikely to use other tools.
(c) Most data scientists spend the majority of their time developing new models.
(d) The use of historical data to make decisions about the future can reinforce historical biases.
(e) All data science investigations start with an existing dataset.
(f) Linear regression is an example of an unsupervised model.
(g) Linear regression can be used to confidently make predictions inside the range of data on
which it was trained on (assume that linear regression works well as a model for this data).
Question 2
You recently learned about Google’s new system for detecting breast cancer in mammograms
(https://www.nature.com/articles/s41586-019-1799-6). The system was trained on a large dataset
of annotated mammogram images from the UK and the USA, most of which were acquired on
devices made by Hologic. The paper shows that the model can be trained on the UK dataset and
still perform well on the USA dataset. Your friend finds the work exciting, and would like to use
Google’s pre-trained model to detect breast cancer in Brazil. Is this a good idea? Why or why
not?
Question 3
Your friend working at UCLA housing has been given the task of determining how students feel
about the new dorms. Your friend wants to accomplish this by scraping Twitter and news article
comment sections for tweets containing keywords and hashtags related to the new dorm and then
running them through a model that does sentiment analysis. This model contains an algorithm
that says whether the text exhibits positive, neutral, or negative sentiment. What are some of the
issues, if any, with what your friend proposes?
CS M148 Fall 2021 Homework 1
Question 4
You would like to see if you can predict the probability that a given student will take a particular
class in the upcoming quarter.
(a) What are some features you would try to gather to investigate this problem (e.g. student’s
year in school, professor teaching the course)?
(b) How would you formulate your labels? (Think about ordinal/categorical/numerical encoding,
min/max/possible values, etc.)
(c) How could you source/obtain/gather the above data?
Problem 5
In Project 1 you learned about imputing data, the step a data scientist must take to deal with
missing or null values in a dataset.
(a) List four different strategies you could reasonably use to address null values. For each, clarify
what the advantages and disadvantages to it are. Additionally, for each strategy, speculate on
what sorts of datasets it would likely be the most effective approach, and what sorts of data
would it be inadvisable for.
(b) Outliers can also be problematic points in a dataset. Which of the strategies you mentioned in
part (a) would also work for outliers, and why?
Problem 6
A jar contains 3 red, 2 yellow, and 1 blue marble. Aside from color, the marbles are
indistinguishable. Two marbles are drawn at random without replacement from the jar. Let X
represent the number of red marbles drawn and Y represent the number of blue marbles drawn.
(a) What is P(X = 1)?
(b) What is P(X = 0, Y = 1)?
(c) What is P(X = 1 | Y = 1)?
Problem 7
Evaluate the following features and determine if you would one hot encode them. Justify your
response:
(a) Walking pace, measured in miles per hour.
(b) Class name (freshman, sophomore, junior, senior).
(c) City in the US (part of an address).
(d) If someone has diabetes or not.
(e) Laptop brand.
(f) Interaction term of how much sleep one got the previous night (numerical value, in hours) and
test difficulty (3 possible values: easy, medium, hard) when trying to predict test score.