Fake News Detection 

Pre-Project, Report 

Authors 
● Rahul 

● Abhishek Rathore 

● Navneet kr. Chaurasia 

Supervisor 
● Dr. Shaima Qureshi 

Abstract 
These days a lot of information is being shared over social media and we are not able to 
differentiate between which information is fake and which is real. People immediately start 
expressing their concern or sharing their opinion as soon as they come across a post, without 
verifying its authenticity. This further results in spreading of it. Fake news and rumors are the 
most popular forms of false and unauthenticated information and should be detected as soon 
as possible for avoiding their dramatic consequences. 

Because the issue of fake news detection on social media is both challenging and relevant, we 
conducted this survey to further facilitate research on the problem. In this report, we present a 
comprehensive review of detecting fake news on social media, including fake news 
characterizations on psychology and social theories, existing algorithms from a data mining 
perspective, evaluation metrics and representative datasets. We also discuss related research 
areas, open problems, and future research directions for fake news detection on social media. 

Acknowledgements 

 
2  

Contents 

Introduction 
Nowadays, the Internet has become an integral part of our life. It is not exaggerating to say that 
it has become the main part of our lifestyle. The role of print media e.g. newspapers and 
electronic media e.g. Television, Radio, News channels in communicating has reduced with the 
onset of social media e.g. Facebook, Twitter, Instagram, YouTube, WhatsApp etc. The growth of 
social media platforms has played an important role in this transformation. The reach of Social 
media is far more than the other media e.g. each house would have a single Television set/ 
Radio/ newspaper, but every member of the house nowadays has access to electronic gadgets. 
With a far more reach the speed with which information is shared has increased multifold over 
social media. Lot of people use social media not only to keep in touch with their friends but also 
to gather news around us. Social media is more relevant for news consumption. Flip side of this 
is, without substantiation/ false information also spreads very fast over social media. The 
fine-grained fake news detection is still a big challenge from a research perspective. The key 
motivations of this survey are summarized as follows: 

● Fake news on social media has been occurring for several years; however, there 
is no agreed upon definition of the term “fake news”. To better guide the future 
directions of fake news detection research, appropriate clarifications are 
necessary. 
● Social media has proved to be a powerful source for fake news dissemination. 
There are some emerging patterns that can be utilized for fake news detection in 
social media. A review on existing fake news detection methods under various 
social media scenarios can provide a basic understanding of the state-of-the-art 
fake news detection methods. 
● Fake news detection on social media is still in the early age of development, and 
there are still many challenging issues that need further investigations. It is 
necessary to discuss potential research directions that can improve fake news 
detection and mitigation capabilities. 

 
3  

To facilitate research in fake news detection on social media, in this survey we will review 
two aspects of the fake news detection problem: c
​ haracterization​ and ​detection​. we will first 
describe the background of the fake news detection problem using theories and properties from 
psychology and social studies; then we present the detection approaches. Our major 
contributions of this survey are summarized as follows: 

● We discuss the narrow and broad definitions of fake news that cover most existing 
definitions in the literature and further present the unique characteristics fake news on 
social media and its implications compared with the traditional media; 
● We give an overview of existing fake news detection methods with a principled way to 
group representative methods into different categories; 
● We discuss several open issues and provide future directions of fake news detection in 
social media. 

Background/Literature Survey 

1. Fake News Characterisation 

In this section, we introduce the basic social and psychological theories related to fake 
news and discuss more advanced patterns introduced by social media. Specifically, we 
first discuss various definitions of fake news and differentiate related concepts that are 
usually misunderstood as fake news.We then describe different aspects of fake news on 
traditional media and the new patterns found on social media 

1.1 Definitions of Fake News 

Fake news has existed for a very long time, nearly the same amount of time as news 
began to circulate widely after the printing press was invented in 14397 . However, there is no 
agreed definition of the term “fake news”. Therefore, we first discuss and compare some widely 
used definitions of fake news in the existing literature, and provide our definition of fake news 
that will be used for the remainder of this survey. A narrow definition of fake news is ​news 
articles that are intentionally and verifiably false and could mislead readers.​There are two key 
features of this definition: a
​ uthenticity​ and i​ ntent​. First, fake news includes false information 
that can be verified as such. Second, fake news is created with dishonest intentions to mislead 

 
4  
consumers. This definition has been widely adopted in recent studies. Broader definitions of 
fake news focus on the either authenticity or intent of the news content. Some papers regard 
satire news as fake news since the contents are false even though satire is often 
entertainment-oriented and reveals its own deceptiveness to the consumers. Other literature 
directly treats deceptive news as fake news which includes serious fabrications, hoaxes, and 
satires.In this report, we use the narrow definition of fake news.Formally, we state this definition 
as follows, 

1.2 Fake News on Traditional News Media 

Fake news itself is not a new problem. The media ecology of fake news has been changing over 
time from newsprint to radio/television and, recently, online news and social media.We denote 
“traditional fake news” as the fake news problem before social media had important effects on 
its production and dissemination. Next, we will describe several psychological and social 
science foundations that describe the impact of fake news at both the individual and social 
information ecosystem levels. 

Psychological Foundations of Fake News.​ Humans are naturally not very good at differentiating 
between real and fake news. There are several psychological and cognitive theories that can 
explain this phenomenon and the influential power of fake news. Traditional fake news mainly 
targets consumers by exploiting their individual vulnerabilities.There are two major factors 
which make consumers naturally vulnerable to fake news: (i) Naı̈ve Realism: consumers tend 
to believe that their perceptions of reality are the only accurate views, while others who disagree 
are regarded as uninformed, irrational, or biased; and (ii) Confirmation Bias: consumers prefer to 
receive information that confirms their existing views. Due to these cognitive biases inherent in 
human nature, fake news can often be perceived as real by consumers. Moreover, once the 
misperception is formed, it is very hard to correct it. Psychology studies shows that correction 
of false information (e.g., fake news) by the presentation of true, factual information is not only 
unhelpful to reduce misperceptions, but sometimes may even increase the misperceptions, 
especially among ideological groups. 

Social Foundations of the Fake News Ecosystem. C

​ onsidering the entire news consumption 
ecosystem, we can also describe some of the social dynamics that contribute to the 
proliferation of fake news. Prospect theory describes decision making as a process by which 
people make choices based on the relative gains and losses as compared to their current state. 

 
5  
This desire for maximizing the reward of a decision applies to social gains as well, for 
instance, continued acceptance by others in a user’s immediate social network. As 
described by social identity theory and normative influence theory, this preference for social 
acceptance and affirmation is essential to a person’s identity and self-esteem, making users 
likely to choose “socially safe” options when consuming and disseminating news information, 
following the norms established in the community even if the news being shared is fake news. 

This rational theory of fake news interactions can be modeled from an economic game 
theoretical perspective by formulating the news generation and consumption cycle as a 
two-player strategy game. For explaining fake news, we assume there are two kinds of key 
players in the information ecosystem: publisher and consumer. The process of news publishing 
is modeled as a mapping from original signal s to resultant news report a with an effect of 
distortion bias b, i.e., s → a, where b = [−1, 0, 1] indicates [lef t, no, right] biases take effect on the 
news publishing process. Intuitively, this is capturing the degree to which a news article may be 
biased or distorted to produce fake news. The utility for the publisher stems from two 
perspectives: (i) short-term utility: the incentive to maximize profit, which is positively correlated 
with the number of consumers reached; (ii) long-term utility: their reputation in terms of news 
authenticity. Utility of consumers consists of two parts: (i) information utility: obtaining true and 
unbiased information (usually extra investment cost needed); (ii) psychology utility: receiving 
news that satisfies their prior opinions and social needs, e.g.,confirmation bias and prospect 
theory. Both publisher and consumer try to maximize their overall utilities in this strategy game 
of the news consumption process. We can capture the fact that fake news happens when the 
short-term utility dominates a publisher’s overall utility and psychology utility dominates the 
consumer’s overall utility, and an equilibrium is maintained. This explains the social dynamics 
that lead to an information ecosystem where fake news can thrive. 

1.3 Fake News on Social Media 

In this subsection, we will discuss some unique characteristics of fake news on social media. 
Specifically, we will highlight the key features of fake news that are enabled by social media. 
Note that the aforementioned characteristics of traditional fake news are also applicable to 
social media. 

 
6  

Malicious Accounts on Social Media for Propaganda.While many users on social media are 
legitimate, social media users may also be malicious, and in some cases are not even real 
humans. The low cost of creating social media accounts also encourages malicious user 
accounts, such as social bots, cyborg users, and trolls. A social bot refers to a social media 
account that is controlled by a computer algorithm to automatically produce content and 
interact with humans (or other bot users) on social media. Social bots can become malicious 
entities designed specifically with the purpose to do harm, such as manipulating and spreading 
fake news on social media. Studies show that social bots distorted the 2016 U.S. presidential 
election online discussions on a large scale, and that around 19 million bot accounts tweeted in 
support of either Trump or Clinton in the week leading up to election day8. Trolls, real human 
users who aim to disrupt online communities and provoke consumers into an emotional 
response, are also playing an important role in spreading fake news on social media. For 
example, evidence suggests that there were 1,000 paid Russian trolls spreading fake news on 
Hillary Clinton9 . Trolling behaviors are highly affected by people’s mood and the context of 
online discussions, which enables the easy dissemination of fake news among otherwise 
“normal” online communities. The effect of trolling is to trigger people’s inner negative 
emotions, such as anger and fear, resulting in doubt, distrust, and irrational behavior. Finally, 
cyborg users can spread fake news in a way that blends automated activities with human input. 
Usually cyborg accounts are registered by humans as camouflage and set automated programs 
to perform activities in social media. The easy switch of functionalities between human and bot 
offers cyborg users unique opportunities to spread fake news. In a nutshell,these highly active 
and partisan malicious accounts on social media become the powerful sources and 
proliferation of fake news. 

Echo Chamber Effect. Social media provides a new paradigm of information creation and 
consumption for users. The information seeking and consumption process are changing from a 
mediated form (e.g., by journalists) to a more disinter-mediated way [19]. Consumers are 
selectively exposed to certain kinds of news because of the way news feed appear on their 
homepage in social media, amplifying the psychological challenges to dispelling fake news 
identified above. For example, users on Facebook always follow like-minded people and thus 
receive news that promote their favored existing narratives [65]. Therefore, users on social 
media tend to form groups containing like-minded people where they then polarize their 
opinions, resulting in an echo chamber effect. The echo chamber effect facilitates the process 

 
7  
by which people consume and believe fake news due the following psychological factors: (1) 
social credibility, which means people are more likely to perceive a source credible if others 
perceive the source is credible, especially when there is not enough information available to 
access the truthfulness of the source; and (2) frequency heuristic, which means that consumers 
may naturally favor information they hear frequently, even if it is fake news. Studies have shown 
that increased exposure to an idea is enough to generate positive opinion of it [100; 101], and in 
echo chambers, users continue to share and consume the same information. Asresult, this echo 
chamber effect creates segmented, homogeneous communities with a very limited information 
ecosystem. Research shows that the homogeneous communities become the primary driver of 
information diffusion that further strengthens polarization. 

2. Fake News Detection 

In the previous section, we introduced the conceptual characterization of traditional fake 
news and fake news in social media. Based on this characterization, we further explore the 
problem definition and proposed approaches for fake news detection. 

2.1 Problem Definition 

In this subsection, we present the details of mathematical formulation of fake news 
detection on social media. Specifically, we will introduce the definition of key components of 
fake news and then present the formal definition of fake news detection. The basic notations 
are defined below, 

● Let a refer to a News Article. It consists of two major components: Publisher and 
Content. Publisher ​pa​​ includes a set of profile features to describe the original author, 
such as name, domain, age, among other attributes. Content c​a​ consists of a set of 
attributes that represent the news article and includes headline, text, image, etc. 
● We also define Social News Engagements as a set of tuples E = {e​it ​} to represent the 
process of how news spread over time among n users U = {u​1​ , u​2​ , ...u​n​,} and their 
corresponding posts P = {p​1​ , p​2​ , ..., p​n​} on social media regarding news article a. Each 
engagement e​it​ = {u​i ,​ p​i​ , t} represents that a user u​i​ spreads news articles using p​i​ at time 
t. Note that we set t = Null if the article a does not have any engagement yet and thus u​i 
represents the publisher. 

 
8  

 
9  

 
 
10  

Overview 
Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh euismod 
tincidunt ut laoreet dolore magna aliquam erat volutpat. Ut wisi enim ad minim veniam, quis 
nostrud exerci tation ullamcorper suscipit lobortis nisl ut aliquip ex ea commodo consequat. 

Duis autem vel eum iriure dolor in hendrerit in vulputate velit esse molestie consequat, vel illum 
dolore eu feugiat nulla facilisis at vero eros et accumsan. 

Goals 
1. Lorem ipsum dolor sit amet:​ Duis autem vel eum iriure dolor in hendrerit in vulputate 
velit esse molestie consequat, vel illum dolore eu feugiat nulla facilisis at vero eros et 
accumsan. 

2. Sed diam nonummy nibh euismod:​ Nam liber tempor cum soluta nobis eleifend option 
congue nihil imperdiet doming id quod mazim placerat facer possim assum.  

Specifications 
 
11  

Nam liber tempor cum soluta nobis eleifend option congue nihil imperdiet doming id quod 
mazim placerat facer possim assum. Typi non habent claritatem insitam; est usus legentis in iis 
qui facit eorum claritatem. Investigationes demonstraverunt lectores legere me lius quod ii 
legunt saepius. 

Milestones 

1. Lorem ipsum 
Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh 
euismod tincidunt ut laoreet dolore magna aliquam erat volutpat. 

2. Dolor sit amet 

Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh 
euismod tincidunt ut laoreet dolore magna aliquam erat volutpat. 

3. Consectetur adipiscing elit 

Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh 
euismod tincidunt ut laoreet dolore magna aliquam erat volutpat.