Copyright 2019© University of Southampton Page 1 of 8


UNIVERSITY OF SOUTHAMPTON COMP6217W1


SEMESTER 2 EXAMINATIONS 2018/19

THE SCIENCE OF ONLINE SOCIAL NETWORKS

Duration 120 mins (2 hours)


This paper contains 6 questions.

Answer ANY FOUR questions.

Each question carries 1/4 of the total marks for the exam paper and you
should aim to spend about 30 minutes on it.

An outline marking scheme is shown in brackets to the right of each question.

Only University approved calculators may be used.

A foreign language translation dictionary (paper version) is permitted provided
it contains no notes, additions or annotations.



















COMP6217W1

Copyright 2019© University of Southampton Page 2 of 8

COMP6217W1

TURN OVER

Copyright 2019© University of Southampton Page 3 of 8
Question 1.

Digital Natives and Digital Immigrants were classifications
put forward by Prensky in 2001 as a way of thinking about
how students engaged with technology in contrast to their
teachers and lecturers.

a. Give a brief definition of Digital Natives and Digital
Immigrants, explaining in each case some of their
assumed characteristics.
(8 marks)

b. In 2010 White and Cornu suggested an alternative
classification based on the idea of Digital Visitors who
view the web as a tool (as a means of completing
defined tasks, possibly anonymously), and Digital
Residents who view the web as a place (where they
connect to others, and build an identity). Explain the
key differences between this approach to
characterising people and the Natives/Immigrants
approach described by Prensky.
(8 marks)

c. Imagine that you were asked to design a real-time
social media chat application focused on University
students. Explain at least two ways in which
considering the students as Digital Natives or
Immigrants might impact the design of your application,
and at least two ways in which changing that view to
Digital Visitors/Residents might alter that design.

Some marks are reserved for how well the changes are
linked to the design impacts. You are free to speculate
about the exact functionality of the app in order to
illustrate your points.
(9 marks)


普伦斯基在2001年提出了数字原住民和数字移民的分类，作为一种思考学生与老师和讲师相
比如何参与技术的方式。
2010年，White和Cornu提出了另一种分类方法，即将网络视为一种工具（作为完成既定任务的一种手段
，可能是匿名的）的数字访问者和将网络视为一个地方（在那里他们与他人联系，并建立身份）的数
字居民。解释这种描述人的方法和普伦斯基描述的本地人/移民方法之间的关键区别。
假设你被要求设计一个针对大学生的实时社交媒体聊天应用程序。请至少说明两种方式，即将学生视为
数字本地人或移民可能会影响应用程序的设计，以及至少两种方式，即将学生视为数字访问者/居民可能
会改变应用程序的设计。保留了一些标记，说明这些更改与设计影响的关联程度。你可以自由推测应用
程序的确切功能，以说明你的观点。
给出数字原住民和数字移民的简要定义，解释每种情况下他们假定的一些特征。
COMP6217W1

Copyright 2019© University of Southampton Page 4 of 8
Question 2.

With so much of our personal data going online via social
media services many people have become worried about the
issue of Privacy.

a. List and briefly describe at least eight of the ten
technological and social Privacy Protection Strategies
identified by Asiri and Millard (2018) that social media
users deploy to protect and manage their online
privacy.
(8 marks)

b. Using a fictive social media example explain why the
principle of Information Asymmetry is a challenge to
the notion of Informed Consent. Make sure you define
both concepts in your answer.
(8 marks)

c. There are a number of approaches to solving the
privacy problems associated with personal data, three
of these are Anonymity, Data Amnesia, and the Right
to Be Forgotten. Describe each and how they are
related.
(9 marks)


有许多方法可以解决与个人数据相关的隐私问题，其中三种是匿名、数据失忆和被遗忘的权利
。描述每一项以及它们之间的关系。
列出并简要描述Asiri和Millard（2018）确定的社交媒体用户用于保护和管理其在线隐私的十种技术和
社交隐私保护策略中的至少八种。
随着我们的个人数据通过社交媒体服务上网，许多人开始担心隐私问题。
用一个虚拟的社会媒体例子来解释为什么信息不对称原则是对知情同意概念的挑战。确保你
在回答中定义了这两个概念。
COMP6217W1
TURN OVER
Copyright 2019© University of Southampton Page 5 of 8
Question 3.
Consider a social network structure that includes a number of
clusters.
a) Discuss the role of clusters in the success or failure of
information cascades in this network. Explain how clusters
either enhance information cascades or are obstacles to
them. In your explanation include density of clusters and
thresholds for the spread of cascades. You answer should
be around one page.
(15 marks)
b) What is the cascade capacity of a network? Is there a
network in which cascade capacity exceeds ½ and what
does the answer to this question imply for new technology
adoption by people on a social network? You answer should
be around half a page.
(10 marks)
讨论集群在这个网络中信息级联的成败中的作用。解释集群是如何增强信息级联的，还是它们的障碍。在
你的解释中包括了簇的密度和级联扩散的阈值。你的答案应该在一页左右。
网络的级联容量是多少？是否存在级联容量超过?这个问题的答案对社交网络上的人们
采用新技术意味着什么？你的答案应该在半页左右。
COMP6217W1
Copyright 2019© University of Southampton Page 6 of 8
Question 4.
Consider the use of graph theory for online social network
analysis.
a) Provide an adjacency matrix for the graph below.
(2 marks)
b) Explain what is outdegree and indegree centrality and what
would a high outdegree and indegree centrality imply for the
vertices in the example that you have provided.
(3 marks)
c) Explain what is eigenvector centrality and its difference to
degree centrality. Provide an outline of the steps involved in
calculating eigenvector centrality.
(9 marks)
d) Provide all examples of balanced and unbalanced graphs
with three nodes, using diagrams as needed. Briefly explain
why each graph is balanced or unbalanced.
(11 marks)
解释什么是outdegree和indegree中心性，以及对于您提供的示例中的顶点，高outdegree和indegree中心性
意味着什么。
提供三个节点的平衡图和非平衡图的所有示例，根据需要使用图表。简要解释为什么每个图形是平衡的
或不平衡的。
为下图提供一个邻接矩阵。
解释什么是特征向量中心性及其与度中心性的区别。提供计算特征向量中心性的步骤概要。
COMP6217W1
Copyright 2019 © University of Southampton Page 7 of 8
Question 5
Consider the case of Facebook as an online social network
provider whose business model includes users, advertisers and
application developers.
a) Explain what are direct networks effects and cross-network
effects.
(5 marks)
b) Discuss whether cross-network effects are observed in
Facebook and whether they are reflected in its revenue
model. Your answer should be around half a page in length.
(10 marks)
c) Everett M. Rogers proposes a number of criteria that
determine the adoption of new products and technologies:
relative advantage, compatibility, complexity, trialability and
observability. Discuss whether each of the above criteria
relate to the adoption of Facebook by a new user.
(10 marks)
TURN OVER
以Facebook为例，它是一家在线社交网络提供商，其商业模式包括用户、广告商和应用程序开发人员。
解释什么是直接网络效应和跨网络效应。
埃弗雷特M。罗杰斯提出了一系列决定新产品和新技术采用的标准：相对优势、兼容性、复
杂性、可测试性和可观察性。讨论上述标准是否与新用户采用Facebook有关。
)讨论Facebook是否存在跨网络效应，以及这些效应是否反映在其收入模型中。你的答案应该在半页左右。
COMP6217W1


Copyright 2019© University of Southampton Page 8 of 8
Question 6.

Consider a game with the following payoff matrix:

B
S P
A S (8, 8) (3, 9)
P (9, 3) (5, 5)


a) Explain what is a dominant strategy and whether there are
dominant strategies for A and B. Also explain whether there
are strictly dominant strategies for A and B in the above
matrix.
(8 marks)

b) Explain which choice of strategies for A and B are Pareto
optimal and which ones are socially optimal.

(7 marks)

c) Consider the Ultimatum Game, where a person A is given a
dollar to split with a person B. Argue about what is the
rational approach for A to determine the split and what
experiments have shown. Discuss whether game-theoretic
approaches could negotiate any differences in behaviour
observed during experimentation. Your answer should be no
longer than ½ page.

(10 marks)





END OF PAPER
考虑最后通牒游戏，其中一个人a得到一美元与一个人B分开。争论什么是确定分裂的合理方
法，以及实验表明了什么。讨论博弈论方法是否可以解决实验中观察到的行为差异。你的答
案不应该超过?第页。
)解释A和B的哪些策略选择是帕累托最优的，哪些是社会最优的。
解释什么是主导策略，a和B是否有主导策略。同时解释上述矩阵中A和B是否存在严格的
支配策略。

欢迎咨询51作业君