$30
CS 498: Computational Advertising
Homework 5
1 General Instructions
• This assignment is due at 11:59 PM on the due date. We will be using Gradescope for collecting this
assignment. The homework MUST be submitted in pdf format on gradescope Contact TAs if you
face technical difficulties in submitting the assignment. We shall NOT accept any late submission!
• Please make sure to appropriately map/assign the pages of your submitted pdf to each sub-question
listed in the homework outline. Handwritten answers are not acceptable. Name your pdf file as
YourNetid-HW5.pdf
• For all questions, you need to explain the logic of your answer/result for every subpart. A result/answer without any explanation will not receive any points.
• It is OK to discuss with your classmates and your TAs regarding the methods, but it is NOT OK to
work together or share code. Plagiarism is an academic violation to copy, to include text from other
sources, including online sources, without proper citation. To get a better idea of what constitutes
plagiarism, consult the CS Honor code (http://cs.illinois.edu/academics/honor-code) on academic
integrity violations, including examples, and recommended penalties.There is a zero tolerance policy
on academic integrity violations; Any student found to be violating this code will be subject to
disciplinary action.
• Please use Piazza if you have questions about the homework. Also feel free to send TAs emails and
come to office hours.
• Please find the link to the Liu and Tang paper pdf below 1
.
2 Question 1 (2 points)
What is homophily? Why is this important in behavioral targeting that uses social information?
3 Question 2 (6 points)
Please answer the following questions based on the Liu and Tang paper [1] which was discussed in class. Each part carries 2 points.
1http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.309.5609&rep=rep1&type=pdf
1
1. The paper talks about a trade-off between an individual’s profile and social information
in behavioral targeting. Identify a case when using social information is likely to yield
poorer results. Explain why.
2. Identify a case when using social information is going to be very valuable. Explain
why.
3. Refer to Figure 2 from the paper(while is also given in class slide 30). Lets look at the
ad clicks for the category Hispanic targeting in this figure. Explain why the likelihood
of clicking on a display ad is so high if at least one friend has clicked on an ad in this
category compared to the case when no friends have clicked on any ad.
4 Question 3 (2 points)
If you are an advertiser and you develop a behavioral profile for an user, should you focus
on their long term or short term interests? How will you make decisions on what you do?
Explain with reasons.
5 Question 4 - Advertising models (10 points)
This is a 3 part question. The questions are based on the discussions that was carried out
in-class about advertising model on 10/25 and 11/01. The first question carries 4 points,
while the others carry 3 points each.
1. This question is based on the advertising model of Stack Exchange. Their advertising
model is as follows—It has high bar for entry, you need to have at least 10000 USD
to spend. There are no auctions. The advertising is based on contract. There is no
real-time bidding. Discuss and identify two positive and two negative aspects of this
advertsing model compared to real time bidding.
2. The second question is based on the precise micro-targeting practices of consumers
that happens during behavioral targeting and computational advertising in general.
Advertisers micro-target consumers at fine grained level. However, should we consider
such micro-targeting as predatory or unfair for more vulnerable consumers like the
elderly or children? In other words, is micro-targeting in the context of vulnerable
population a positive or a negative development? Discuss.
3. Based on your discussion of micro-targeting scenario in the previous part, now consider
this case— the display of political ads in a platform like Facebook is based on microtargeting. Similar to Twitter, should Facebook ban all political ads on its platform?
Discuss.
2
References
[1] Kun Liu and Lei Tang. “Large-scale behavioral targeting with a social twist”. In: Proceedings of the 20th ACM international conference on Information and knowledge management. ACM. 2011, pp. 1815–1824.
3