Description
2 Question 1 (5 points)
Scroll to Section 14.7, Exercises in the reference chapter linked on the first page. In this
question, you are required to solve Problem (2). In parts (a) and (b), you must show the
precise steps to arrive at your answers. They carry 2 points each. Part (c) is worth a single
point. You need to both, provide the right answer and an appropriate explanation for why
you think you obtained this result to receive this point.
Note: Wordy explanations will be penalized, be as concise as you can.
3 Question 2 (10 points)
In this question, you are required to solve Problem (3), Section 14.7 in the reference chapter.
Part (a) carries 2 points, while Parts (b) and (c) are worth 4 points each, split across the
computation of authority scores and your explanations for the results.
4 Programming Question 3 (10 points)
This question aims to provide you a better understanding of the HITS algorithm in action.
We will host the question via Hackerrank, the contest link will be posted on piazza soon.
• Please read the problem description carefully.
• The input will always be valid. We are mainly testing your understanding of the HITS
algorithm, not your coding skills.
• Please pay special attention to the output format. We will be using the hackerrank
based autograder and it is extremely important that your generated output satisfies
the requirement.
• We don’t have specific constrains for this programming question. The
only constrains are the standard environment constraints in hackerrank:
https://www.hackerrank.com/environment.
• The grading will be based on the test cases your code passes. You are provided with
two sample test cases to debug your code. For the final grading, we will use real-world
datasets. Make sure your code can scale to large graphs.
2
5 Question 4 (2 points)
Assume you know the K most profilic movie reviewers (highest hub-scores) in the above
programming question, and the set of movies each of them have reviewed. A new user
decides to review movies in a manner to maximize his potential hub-score, however he has a
limited budget (he can review a small fixed number of movies). What is a good strategy for
him to adopt given the above information?
(Limit your answer to a single line, wordy answers will be penalized)
Does the same strategy work for both large and small graphs? Hint: The effects of adding
a new user are much smaller in a large graph
3
