ISSN 2278-3091

Riaz Ahmed Shaikh et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 2435 – 2440
Volume 10, No.3, May - June 2021
International Journal of Advanced Trends in Computer Science and Engineering
Available Online at http://www.warse.org/IJATCSE/static/pdf/file/ijatcse1311032021.pdf
https://doi.org/10.30534/ijatcse/2021/1321032021

An Analysis on Measuring Graph patterns in Social Networks

Riaz Ahmed Shaikh1 , Noor Ahmed Shaikh2, , Muhammad Hanif Tunio3, Yumna Magsi4, Sajid Ahmed
Ghangro5, Saima Shoro6
Department Of Computer Science
1
Shah Abdul Latif University Khairpur Mir;s, Sindh, Pakistan, [email protected]
2
Shah Abdul Latif University Khairpur Mir;s, Sindh, Pakistan, [email protected]
3
Shah Abdul Latif University Khairpur Mir;s, Sindh, Pakistan, [email protected]
4
Shah Abdul Latif University Khairpur Mir;s, Sindh, Pakistan, [email protected]
5
Shah Abdul Latif University Khairpur Mir;s, Sindh, Pakistan, [email protected]
6
Shah Abdul Latif University Khairpur Mir;s, Sindh, Pakistan, [email protected]
\

ABSTRACT
Concepts from graph theory might be useful in social
networking for a variety of reasons. Graphs have
recently been widely utilized in social networks for a
variety of reasons, including modelling and analysis
of SN architecture, SN operation modelling, SN user
analysis, and a variety of other topics.A graph
consists of nodes; just like that, a social media is a
kind of social network where a node is defined by
each person or organization. Via common interests,
relationships, mutual friends, expertise, common Figure 1 Graph
dislikes, views etc., these nodes in a social network
are interdependent on each other. The application and The vertices are basically the nodes of the networks
implications of graph theory on social networking, social and edges are the connection between those nodes.
network analysis, Interpersonal relations, and social The edges are of two types
network properties are focused in this paper. The
1. Directed Graph
main goal of his study was to relate the social
networks with graph theory.An Analysis of a few 2. Undirected Graph (Shown in Figure 2)
Facebook users, needed for social media structural
analysis, is also included.
Key words: Graph Theory, Social Network,
Facebook, Twitter.
1. GRAPH
Graph is a pictorial representation of statistical data
of a functional representation of the variable. [1] It
shows the general tendencies in quantitative behavior
of data therefore serve as a predictive function. A
graph is used to represent the network of edges and
vertices which are interconnected with each other.
[2][3][4] Shown in figure no: 1
Figure 2 Directed and Undirected Graph
Directed graph shows the direction of path from one
node to another however undirected graph does not

2435
Riaz Ahmed Shaikh et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 2435 – 2440

describe the particular path towards the nodes. On the On the basis of nodes contents network can be
basis of presence and absence of cycles the graphs divided into two types: (Illustrated in Figure 4)
can be classified as:
2.1 Social and Economic Networks
1. Cyclic
2. Acyclic [5] (shown in figure 3) It includes the group of people interconnected with
each other’s through some patterns of
communications.
Such as Facebook, Twitter, WhatsApp, Instagram and
business Networks etc.

2.2 Information Networks

Information Networks describes the connections

between information objects. Such as Semantics,
Word Wide Web.
Whereas the main focus of this paper is the social
networks particularly Facebook and Twitter.
There areFigure 3 Cyclic
several and Acyclic
applications Graph theory but in
of graph 2.3 Social Networks
social networking it is gaining popularity day by day.
A social network is a website that brings people
This paper reveals the graph with reference to social together to chat, connect or make new friends with
networking. ideas and interests. Social networking is referred to as
this form of communication and sharing.[6]
2. NETWORKS
Depending on the branch of mathematics used, there
are many ways of formally describing a network.

Figure 4 Networks
Figure 5 Social Networks

A social network is conceptualized as a graph, that Social networking sites contain content produced by
is, a collection of vertices (or nodes, units, points) hundreds or even millions of different people, unlike
representing social entities or objects and a set of conventional media created by no more than ten
lines representing one or more social connections individuals. Figure 6 shows the network of friends.
between them. The most common and versatile
meaning is derived from the theory of graphs. [5]

2436
Riaz Ahmed Shaikh et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 2435 – 2440

reference to graph theory the users are vertices of

graph whereas the tags, comments and likes are
known as edges of the graph.[6] Figure 8 shows the
graph of Facebook.

Figure 6 Network of Friends

3 GRAPH THEORY AND SOCIAL

NETWORKS
The graph theory covers the large area of Social
Networks. Where the users are considered as vertices
Figure 8 Graph of Facebook
or nodes whereas the relations such as common likes
and mutual friends are considered as edges.
[7][8][8][9][10][11] Figure 7 shows the graph of 3.2 Graph Theory in Twitter:
social networks.
The persons or users are considered as nodes and
their following is known as edges. Shown in figure 9.

Figure 7 Graph of Social Networks

Figure 9 Graph of twitter

3.1 Graph Theory in Facebook:
As Facebook is the major source of communication
4. MEASURING NETWORKS ON FACEBOOK
now a days. The Facebook provides an interesting
way to connect with your friends, family and other The size of Facebook networks is reflected by this
strangers by liking their posts, tagging them on case study. For the course of a month, few random
timeline and by commenting on their posts. With

2437
Riaz Ahmed Shaikh et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 2435 – 2440

users have been studied, and the relationship is researchers. The challenge of partitioning a graph is
categorized into four distinct network patterns [7] entirely algorithmic. It aids in the reduction of the
complexity of large graphs as well as the introduction
4.1 All Friends of parallelization.Graph partitioning is needed in a
variety of applications, including social networks,
This network is the list of all a user's friends, so this road networks, air traffic management, picture
is the largest of all representations. analysis, and so on. Scientific computing, dividing
4.2 Reciprocal communication various phases of VLSI design circuits, job
scheduling in multiprocessor systems, and other uses
This representation indicates the reciprocal of graph portioning issues are only a few examples.
communication between the two parties, where there The goal of graph partitioning is to separate nodes
is a mutual exchange of information between the two into many disjoint sections in order to minimize the
parties, creating this sort of network. [8] stated objective function. Although the issue of
optimum network partitioning is NP-complete,
4.3 One-way communication several approximation methods have been developed
to address it.
It is made up of individuals with whom a user has
interacted. There are two types of graph partitioning:
4.4 Maintained Relationships 5.1 Constraints Partitioning
This relationship trend consists of individuals whose This partitioning has portions that are all the same
profile to maintain commitment has been reviewed size.
by the consumer more than once. [9] [10] The blue
line shows the number of mutual relationships in the 5.2 Unconstrained Partitioning
Figure: 3, the blue line shows the one-way
relationships, and the blue line shows that This type of partitioning has sections of varying
yournetwork size is a function of passive sizes. For graph partitioning, a number of methods
relationships. have been developed. Geometric partitioning,
Spectral partitioning, and Multi-level graph
partitioning are three of the most important
algorithms. [14] Graph partitioning is shown in figure
11.

All Friends Maintained Relationship

Oneway Communication Mutual Communication Figure 11 Constraint and Unconstraint partitioning

Figure 10 Network patterns of Facebook Users

5. GRAPH PARTITIONING The graph is bisected in Geometric partitioning by

using the coordinates that are acquired if nodes of a
Graphs are frequently utilized to represent an graph are available in space.
application software by numerous scientists or

2438
Riaz Ahmed Shaikh et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 2435 – 2440

The vertices that are geographically close to each arcs, or edges. It is possible for a graph to be directed
other are grouped together in this division. The or undirected. G=(V, E), where V is a collection of
associability of the graph is determined by vertices and E is a set of edges, is a graph.
determining the Eigen vectors with regard to the
second lowest Eigen value of the Laplacian matrix L A graph is represented using adjacency lists,
corresponding to the graph in Spectral partitioning. adjacency matrices, incidence matrices, and other
This bisection approach is extremely time data structures.Individuals or organizations are
consuming, yet it is not practical for big graphs. represented as nodes in a social network, while
relationships between individuals or organizations are
Multi-level partitioning outperforms traditional graph represented by edges.A collection of tools for
partitioning techniques. This technique effectively studying the structure of whole social entities is
solves constrained graph partitioning issues. The provided by social networks.Different degrees of
basic concept is to divide huge networks into k analysis exist, and they aren't always mutually
sections, group the vertices into a group, and deal incompatible.
with this group of vertices instead of individual
vertices. Generally there are three levels:

It is divided into three stages: 6.1 Micro level

1. Coarsening, It usually starts with a person or a small group of

individuals at this level.
2. Initial partitioning,
6.2. Meso level
3. Partition refining.[14][15] (discussed in figure 12)
This level is halfway between micro and macro. It
demonstrates the link between the micro and macro
scales. Meso-level networks are networks with a low
density.
6.3. Macro-level networks
These are large-scale networks that track
relationships across enormous populations.
These are more complicated networks illustrated in
Figure 13.

Figure 13 Levels of graph

7. CONCLUSION

6. ANALYSIS OF STRUCTURE This paper addressed the different aspects of graphs

Figure 12 Multilevel Partitioning and their basic properties. Graphs are used in social
A graph is a mathematical structure that depicts the networks that are growing in sophistication with the
relationships between different items. The vertices, progress of social media. Different algorithms are
nodes, or points of a graph are connected by lines, still being developed by researchers to better

2439
Riaz Ahmed Shaikh et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 2435 – 2440

understand social networks and to partition graphs to 12. J. Levy Moreno, Who shall survive?:
reduce their complexity and to better understand Foundations of sociometry, group psychotherapy
networks. We have given a survey of the work done and sociodrama, Beacon House, 1953.
in the field of social networks that uses graph theory 13. M. McPherson, L. Smith-Lovin, J.M. Cook,
ideas in this paper. By evaluating graph Birds of a feather: homophily in social networks,
characteristics and selecting the correct mix of graphs Annual Review of Sociology, Vol. 27, 2001, pp.
based on their challenge, the contribution can help 415-444.
many academics from diverse viewpoints.Beginner 14. G. Robins, P. Pattison, J. Woolcock, Small and
researchers can benefit from the well-summarized Other Worlds: Global Network Structures from
possible applications of graph theory, which can help Local Processes. American Journal of Sociology
them understand its ideas and current use in a variety (AJS), Volume 110, Number 4 (2005), 894-936.
of SN areas.The application and implications of graph 15. Kaur, Chhinder, and Anand Sharma. "Sentiment
theory on social networking, social network analysis, Analysis of Tweets on Social Issues using
Interpersonal relations, and social network properties Machine Learning Approach." International
are focused in this paper. The main goal of his study Journal 9, no. 4 (2020).
was to relate the social networks with graph
theory.An Analysis of a few Facebook users, needed
for social media structural analysis, is also included.

REFERENCES

1. Marcin Mince, EwaNiewiadomska-Szynkiewicz,

“Application of Social Network Analysis to the
Investigation of Interpersonal Connections”,
February 2012.
2. Kimball Martin, “Graph Theory and Social
Networks”, April 30, 2014.
3. D. Easley, J. Kleinberg. Networks, Crowds, and
Markets: Reasoning About a Highly Connected
World. Cambridge University Press, 2010.
4. T.h. Cormen, C.E. Leiserson, R.L. Rivest, C.
Stein, Introduction to Algorithms (2nd ed.), MIT
Press and McGraw–Hill, ISBN 0-262-53196-8,
2001.
5. AlexandruCostan, “Graph Theory and Social
Networks”,
6. https://www.uva.nl/binaries/content/documents/p
ersonalpages/ n/o/w.denooy/en/tab-one/tab-
one/cpitem%5B26%5D/asset?1355372751494.
7. Daron Acemoglu and AsuOzdaglar, “Networks,
Lecture 2: Graph Theory and Social Networks”,
September 14, 2009.
8. Jayson Rome, “Graph Triangulation”, October
14, 2002.
9. S. Wasserman, K. Faust, Social Network
Analysis in the Social and Behavioral Sciences,
Social Network Analysis: Methods and
Applications, Cambridge University Press, 1994,
pp. 1–27.
10. J. Farganis, Readings in Social Theory: the
Classic Tradition to Post-Modernism, McGraw-
Hill, New York, 1993.
11. G. Simmel, Die Grosstädte und das Geistesleben
(The Metropolis and Mental Life), Dresden:
Petermann, 1903.

2440