# Midterm Project

Chapter 1
1. Why is observing and describing a social network at the population level, as opposed to the
individual level, necessary? What benefits does the population-based analysis of a social
network provide to us? What are *at least* four important network dynamics that can be
studied at the population level? Describe each network dynamic and give at least one
example of EACH of the four network dynamics you can find in the real world. (a total of 4
network dynamics, and a total of 4 examples [1 example per network dynamic]). (10 points)

3

Chapter 2
2. Using Nodes and Edges, draw a graph that has a node that is a gatekeeper for other nodes.
You can either draw a graph on another sheet of paper, take a picture, and include in your
answer, OR you can use the shapes in the word processing software. Either way, make sure
that your handwriting or text in the figure is clearly legible to the grader. If we can’t read
your text, we cannot give you full credits. This figure will not count towards the page limit.
(5 points)

2.1 Based on the figure to the left, explain
what breadth-first search is and how you
would measure the distance between “you”
and “Tom”. Then, explain what implications,
benefits, and lessons the breadth-first search
bring to you. A well-known example of a
searches is the discovery of a “small world
phenomenon” where everyone in the world
is connected with <7 distance. Do NOT use
this example of a small world phenomenon.
List another, different example of benefits of
calculating distance. 5 BONUS POINTS: if you
find an example usage of breadth-first search

in an academic article published in a peer-
reviewed journal. (10 points)

Chapter 3.
3. Consider the graph on the right, in which each edge —
except the edge connecting b and c — is labeled as a
strong tie (S) or a weak tie (W). According to the theory
of strong and weak ties, with the strong triadic closure
assumption, how would you expect the edge connecting
b and c to be labeled? Give a brief (1-3 sentence)
Tom

4
3.1 Problem 2 – In the social network depicted below in Figure 3.23 with each edge labeled as
either a strong or weak tie, which two nodes violate the Strong Triadic Closure Property?

3.2 Explain why most people get information about their jobs from acquaintances not from
close friends/family members based on the notion of “strength” of weak ties. Make sure
you explain what a weak tie is, why/how a weak tie is formed (using strong triadic closure
property), and what benefits accrues from the triadic closure property. You can find a
partial answer to this question in Figure 2.8. Reading this entire chapter and the lecture
notes (“Eureka!” slides) will give you two sets of reasons. Mention these two sets of reasons
explaining from strength of weak ties. (15 points)
Chapter 4.
4 Explain how Christaki and Fowler found that obesity is contagious. Identify the three sets of

factors they tested (listed in your textbook page 83, Chapter 4 lecture notes as well as Ted-
Talk shown during the lecture), and illustrate how they used the homophily test to conclude

that obesity is contagious and what mechanism underlies the contagion. Be sure to refer to
the homophily test that was used to assess if the plot (Figure 4.2) exhibits homophily in
your textbook. Be sure to modify the variables (or parameters) in this homophily test to be
applicable to the assessment of homophily in the contagion of obesity. Then, explain what
underlying mechanism Christaki and Fowler have suggested to explain the contagion of
obesity. Which mechanism—either social influence or selection—explained the contagion of
obesity? (15 points)
4.1 Given a network showing pairs of people who share activities, we can try to reconstruct an
affiliation network consistent with this data. For example, suppose that you are trying to
infer the structure of a bipartite affiliation network, and by indirect observation you’ve
obtained the projected network on just the set of people (refer to Exercise 2 page 105, and
discussed in the recitation October 1st). There is an edge joining each pair of people who
share a focus. This projected network is shown below in Figure 4.22.

5
– Draw an affiliation network involving these six people, together with four foci that you
should define, whose projected network is the graph shown above in Figure 4.22. (10
points)
Chapter 5.
5 For the diagram below in Figure 5.4, Assuming that the network is balanced, please answer
why the question marks are positive or negative. (5 points)

6
5.1 In the Labeled Graph below, please explain if the graph is balanced or unbalanced in 1-2
sentences. (5 points)

5.2 Describe how you would use structural balance theory if you were to develop an automatic
recommender system for products for your customers. Automatic and intelligent
recommender systems generate and suggest products that customers are likely to
purchase, based on the structural balance of a social network as presented in Qi et al.’s
paper (to be shared during the lecture on October 6th). (15 points)
L. Qi et al., ‘‘Structural balance theory-based e-commerce recommendation over big
rating data,’’ IEEE Trans. Big Data., Sep. 2016, doi: 10.1109/TBDATA.2016.2602849.
You do not need to understand the programming code or equations in this article. Focus on
their ideas about how to use structural balance theory in order to increase the accuracy of
their recommendations (i.e., the likelihood that their recommended products are purchased
by the customers).

*** End of the Questions ***