The research work is
carried out for the algorithms that are used by the Facebook. These algorithm’s help’s in making sense
of the huge number of posts that people and business share every day in
Facebook. Facebook is a social networking website that allows users
to create profiles, post information, share information and put comment on any
activity. Facebook provides user news,
interesting contents, games, newsfeeds, chat, live video etc. This article represents
a discussion about Facebook, various algorithms used by it. All the
communication (Posts) on Facebook is performed by these algorithms. And the two
algorithm’s are Edge Rank algorithm and News Feed algorithm.
Keywords: Facebook, Newsfeed, Chatbox, Posts, EdgeRank.
Facebook is a
trending social networking website. It was created by ‘Mark Zuckerberg’ and ‘Edvard
Saverin’ in the year 2004. Facebook provides user news, interesting contents,
games, newsfeeds, chat, live video etc. Users can share the posts. Facebook is available in 37
different languages. A user can make his/her profile visible to everyone by
keeping the profile public. Facebook offers a privacy option to its users. By
using privacy they can set their profile exactly who can
see their posts. Nearly, Facebook has more than
two billion monthly active users till June-20179.
Figure1: Login and Sign up page of Facebook. (reference
There are few parts, which plays a major role in Facebook
News Feed was introduced in the year 2006. Newsfeed enabled spammers and other
users to take control over features of Facebook like upcoming events, birthday
events, etc., by creating fake events or which are not usual and fake birthdays
to attract attention towards their profile1.This feature was soon desirable by
Facebook users with the reason that it was containing unwanted information, and
it is too easy for users to track the information of other users 16. Then for this in convenience, the
company introduced the privacy settings2. In the year 2010, Facebook granted a patent on certain aspects of its
News Feed 4. One of the features is photo
album. In the Facebook, posting of photos takes a major role than any other
applications. In the early time, Facebook kept a limitation of 60 photos 10
per album, but now the limit has been increased to 200 photos per album. User
can also share the photos
to our friend’s, or can
hide the photos by using privacy settings 1214. By using these privacy settings, no one can see our
posts, or photos. And also user can tag our friends while posting
photo’s and the tagged ones will receive the message that they have been
tagged. And also users can find gaming in this. Facebook has seen 150M monthly
users with 2.4 times the installation of apps. Facebook uses Edge Rank
Algorithm for sorting and displaying the news in the users news feed.
This feature is used to communicate with the
friends and others who have been member in Facebook. It is used to text
messages, can do calls, and video calls. Facebook Messenger is an instant
messaging app. The main purpose of this app is chat heads. Chat heads are a
small part in Facebook, which is taking major role now. If user wants to chat
with someone the chat box can be pop up. Text messaging has created revolution
in the communication society. So there is no need of talking in calls. Facebook
chatting made everything easy, and interesting, by offering stickers, emojis, gifs.
And also user have options in chatbox. In that user have chat sounds, emoji’s,
block settings, advanced settings, hide Groups, turn off chat, turn off
video/voice calls, turn off post tabs, close all chat tabs.
In the home page an option called Groups. By using
this, user can create groups. Groups can
be created on favorite books, articles, paints, and also related to business.
Group’s helps the user to find friends with common interest, and in groups
everyone can share photos 10, and information. Users can do business (like
selling products, electronic goods, do shopping, etc.) on Facebook group. Groups also have privacy settings 6. One has to request before joining the group. After the
acceptance of group creator, user can join and
share and access that group.
Pages: A Facebook page is a public profile. It is
used to share and mostly used for business purpose. It is used to support
businesses, brands, shopping websites, and for social media 1. Page contains
fans or followers. Pages can be accessible to everyone even though they do not
follow the page. Pages promote events, updates, statuses, videos, photos and
User can post in pages, groups or in own profile. Posts can be a video, photo,
information, news, and links. Home page uses Featured Images. If a featured
image is posted then it takes first priority. When a post is shared in the Facebook,
it stores the image from that in its cache to use in the future. In posts, user
can like, share, and comment 1. And while posting a post user can also tag
some of your friends. For posts also, privacy settings can be applied. And if user
find any post unusual, user can also report that post. And posts also have some
options like save the link, turn off notification, and don’t allow remixes for
the post. Posts are used by pages and groups for many purposes like business,
fan pages, social media. Comments plays a major role in posts.
on Facebook: Facebook provides gaming applications for
its user. To access the games in Facebook or to play the games in Facebook user
have to login to the account using email id or phone number and password. And Facebook
started developing games 3,. A group of people can play single game, or an
individual can play the games.
EDGE RANK ALGORITHM: -Edge rank is introduced by
Introduction to EdgeRank:
Edge rank is an algorithm which use to determine
what posts are to be displayed in the news feed. Edge rank is the face book
algorithm that decides which stories appear in each user’s news feed. This
algorithm hides moving stories, so if your story doesn’t score well, no one can see it.
When user login to your Facebook the first thing you
see is newsfeed. News feed is the posts
which have been posted by our friends, group and pages. Every action their
friends take is a potential newsfeed story. Facebook calls these actions as “Edges”.
Whenever user post a status updates, likes, comments, shares, then it generates
an “Edge”. And the story about that edge might show in the user’s personal
So face book created an algorithm to predict how
interesting each story will be to each other user and the face book calls this
algorithm as edge rank as it ranks the edges. Then they sort user’s newsfeed to
show the top ranked stories for that particular news.
Facebook has two algorithms one is proven and other
has not been conclusively proven. The edge rank algorithm ranks stories and the
second algorithm sort the news feed. This newsfeed algorithm includes a
randomization element and a keyword aggregator 9.
Working process of EdgeRank:
Edge rank is a credit rating. It is invisible, it is
important, it is unique to each user’s and no one other than Facebook exactly
knows how it works.
Edge rank algorithm contains three parts: –
1. Affinity Score
2. Edge Weight
3. Time Decay
Score: Affinity score means connecting a particular userto
example:- Satya and Niharika are two friends on Facebook
both of them are having fifty mutual friends and frequently posts stories on
walls of each other they will have the high affinity score with both of them.
So, Facebook knows they are probably wanted to see his status updates.
Facebook calculates the affinity score by looking at ecliptic actions that
users take and also considering the factors
1. The strength of the factors
2. How close the person who looks the action was to user.
3. How long ago they took the action.
Actions include clicking, liking, commenting, tagging, sharing and friending.
Each of these actions has different weights and reflects the effort required
for the action. Commenting on something gives more score than merely liking it,
which is worth more than clicking on it. Passively viewing status updates in
your newsfeed doesn’t count affinity score unless user interact with it.
Affinity score measures not only user action but also user friend’s action and
their friend’s action.
example: If user commented on a page, it scores more than if
user’s friend commented, which is worth more than if a friend of friend
commented. Not all friends’ actions are treated equally.
if user used to interact with someone a lot, but
less so now, then their influence will start to have a progressively smaller
part of its visible surface illuminated. So that, the score appears less.
Technically a Facebook is just multiplying each action by 1/x; where x is the
time, since the action happened. Affinity score is one way, user friend has a
different affinity score to user.
Each category of edges has a default weight. In this
edge weight comments 10 are worth more than likes. In this user are more
likely to see a story in your newsfeed about user commenting on a post than a
story about your liking a post.
If the user reacts to a same related topic Facebook
change the edge weights according to user interactions. In order to publicize
the features to user interactions, Facebook is giving more edge weight to the
For example: – when Facebook places roledout, space
check-ins had a very high default weight for a few months and your news feed
was probably inundated with stories like “John checked into old navy”.
Generally, after a few weeks or months Facebook dials the new feature back to a
more reasonable weight.
This part talks about what Facebook refers to as
Time Decay. This is mainly used to understand why Facebook shows some updates
more than others. And why some people seem to stay in your newsfeed. Edge rank
is a running score but not a single score. When a user logins to Facebook,
their newsfeed is completely fills with the edges which have a height score at
a very particular interval of time. As the story gets older and older edge rank
will be decreased because it is old story.
Generally, Facebook seems to be adjusting its time
decay based on following factors:-
long does the user logged off from Facebook.
frequently the user logs into Facebook.
Checking of user edge rank:-
User can measure the edge rank by seeing how many
people user reached. User can also measure by how many comments or check-ins user
The process of checking user edge rank score by
score between user and edge creator
for this edge type
decay factor based on how long ago the edge
NEWS FEED ALGORITHM:-
News feed algorithm helps in make sense of the huge
number of posts that people share every day.
The elements of the newsfeed algorithm are:-
Newsfeed visibility is based on a
score of how relevant Facebook considers the story to be and takes into account
an immense number of factors. A story can be a status update, photo, and video,
comment, like, tag, event, relationship status change or any of the other
pieces of content that user see in news feed.
This is concerned with how the user
is interested in creation of the story. Interest is also gauged by how long user
spent in the past on stories that user has posted, how the user reacted to
those stories like, comment, share etc.
In this it considers how well the
story is performing with other users and the kind of reactions that the post is
getting. If your friends have liked the post user are more likely to see it pop
into your own news feed.
It checks the user performances
like is the page is verified? What type of page is it? Does it have complete
profile information? How old is that page? Have your fans elected to see your
content first? On the other hand it may not even be a page; posts from Facebook
groups enjoy a lot more visibility due to a different algorithm at work.
Facebook itself weights some types
of post more highly than others. So some content types are given a higher score
and are promoted, and less popular contents will get a lower score.
This is to do with the age of a
post. Generally the longer story has been live the less likely it is to be
seen. There will be exceptions: with personal posts. Recency also raises the
question of when the best time to post. When are your fans on Facebook?
visibility = (interest in page) * (post performance) * (past page performance)
(Type of content)*(recency)
Facebook is having
algorithms which are used for sorting posts in user’s news feed which is not
practically proved and one more algorithm which gives a rank to user its called
edge rank. By following the above steps user’s score is decided and it pays the
users to understand how user can affect it so that user don’t end up skipping
your news feed.Facebook select the most relevant and engaging stories to show
in the newsfeed, it’s called news feed algorithm. So here we conclude about the
Facebook, and its basic algorithm which is used for users and they plays a
major role for giving ranks to users and manage posts.
T.(2010)Branching out: Facebook Pages and Facebook Advertising. In: Social
Networking Spaces. Apress.Springer, Berlin, Heidelberg.
a., Coudert F.(2011) Privacy Settings in Social networking Sites: Is It Fair?. In: Fischer-Hubner S., Duquenoy
p., Hansen M., Leenes R., Zhang G.(eds) Privacy and Identify management for
Life.Privacy and Identify 2010. IFIP Advances in Information and Communication
Technology, vol 352. Springer, Berlin, Heidelberg.
3. Wayne Graham, “Beginning
Facebook Game Apps Development” pp 1-14, published in Springer.
4. Roosendaal A. (2012) We Are All Connected to Facebook … by Facebook! In:
Gutwirth S., Leenes R., De Hertt P., PoulletY . (eds) European Data Protection:
In Good health?.Springer, Dordrecht
5.Agre, Philip E., and Marc Rotenberg.
1997. Technology and privacy: The new landscape. Cambridge: MIT press, Google
6. SApargaliyey D. (2012) Using a Facebook
Closed-Group as Part of an Online Course. In: Li K.C., Wang F.L.Yuen K.S.,
Cheung S.K.S., Kwan R. (Eds) Engaging Learners Through Emerging Technologies.ICT 2012. Communications in Computer and
Information Science.Vol 302.Springer, Berlin, Heidelberg.
7. Hung YH.,Lai HY.(2015)Effects of Facebook Like and conflicting Aggregate
Rating and Customer Comment on Purchase Intentions In:Antona M.,Stephanidis
c.(eds) Universal Access in Human-ComputerInteraction.Access to Today’s
Technologies, UAHCI 2015,Lecture Notes in Computer Science, Vol 9175.Springer, Cham.
8. Shankar Setty, RajendraJadi, Sabya Shaik,Classification of Facebook news feeds
and sentiment analysis., 01 December 2014, in IEEE.
9. KlimisNtalianis, AnastasiosDoulamis, Nikolaos
Doulamis, “Video abstraction in social media: Augmenting facebook’s EdgeRank
algorithm in video content presentation”, 13 February 2014, in IEEE.
10.Thao Thanh Nguyen, AllaG.Kravets, “Analysis
of the social network Facebook comments”, 19 December , in IEEE.
11. Mark D. Wood, Minwoo Park,”Exploring Photos
in Facebook”, 01 February, in IEEE.
12.Siti Zainab Ibrahim, Ann Blandford, Nadia
Bianchi-Berthouze, “Privacy Settings on Facebook: Their Roles and
Importance”, 25 February 2013, In
13. Springer, Cham,Damen S., Zannone N.(2014)
Privacy Implications of Privacy Settings and Tagging in Facebook. In: Jonker
W., petkovic M. (eds) Secure Data Management. SDM 2013.LectureNotesin Computer Science, vol
8425. Springer, Cham.
14.Springer, Berlin, Heidelberg, Awareness,
Information Sharing, and Privacy on the Facebook, 2006, in Springer.
15. facebook: https://www.facebook.com.