Notes - SN

Social Network refers to the use of internet-based social media websites to stay
connected with each other. People like friends, relatives or family members,
professional actors, etc share their ideas, media like audio, video, live video,
documents, voice notes, video calling, audio calling on social platforms to stay
connected with each other.
Types of Social Networking

1. Social Media - facebook, instagram, twitter
2. Professional - LinkedIn
3. Discussion forum - Quora, Reddit
4. Educational - YouTube
Famous social networking websites :

1. Facebook
2. Instagram
3. Twitter
4. LinkedIn
5. YouTube
6. Snapchat
7. Tumblr
8. Tiktok
9. Tumblr
10. Quora
NetworkX is a graph package which is used to create and modify different types
of graphs. It provides a rapid development environment for collaborative,
multidisciplinary projects.
Basic inbuilt graph types :

1. Graph - stores nodes and edges and is undirected
2. Multi-graph - it acts as a base class for undirected graphs.
3. Di-graph - it acts as a base class for directed graphs.
4. Multi Di graph - type of a directed graph class that can store multi edges.
Network effects can be found throughout social networking websites. For
example, as more users post content on Twitter such as links and media, the
more useful the platform becomes to the public. The network effect has created
exponential growth rates for networking platforms such as Facebook, YouTube,
Instagram, etc.
The three primary sources of value for network effects :
1. Exchange - users can communicate with each other over social media.
2. Staying power - Network with greater number of users suggests stronger
staying power.
3. Complementary benefits - Complementary benefits are those products or
services that add additional value to the network.
Social Embeddedness - It refers to extent to which organisations are connected

to other actors via linkage of social networking media sites.
Network dataset stores the structural information of peoples who are connected
with each other by edges. Dataset allows user to know what kind of relation do
the people have, for eg. friendship, relationship or can be extracted using social
network sites.
Types of Datasets :
1. Ingredients - May have many dishes made of some same ingredients but
it is not possible those ingredients will be the same in every dish. For eg. a
dish may have ingredients like sugar and salt but it is not possible for a
dish with sugar, chilli powder, and chillies in the same dish. So a dish with
different ingredients cannot be linked with another dish with different
ingredients.
2. Synonym - A word’s synonym can be related with another word’s

synonym. For eg. passion and happiness can have a similar synonym but
it can be directly linked with the sad word. Happy’s synonym can be related
with another word which may be related with another word and so on. So
indirectly happiness’s synonym can be linked with Hate’s synonym
indirectly.
3. Zachary Karate Club Dataset - This is a very well-known and most used
dataset. Each node represents the member and every edge represents the
link between two members. Zachery used this dataset to explain the
splitting of groups following their disputes among the members.
4. Web graph - Web graph is a social diagram which shows the

interconnection between the individuals, group of people in a social
network. In the social graph diagram, the individual acts as a network node
and relations between two nodes are called edges.
Formats :
1. CSV (Comma separated value) :

An .csv file may have .txt or .csv as their extensions. Further .csv
has 2 more types, it can be either adjacency list or edgelist. .csv files
have data in records at every line, followed by commas.
2. GML (Graph Modelling Language) :

It is most commonly used for network dataset. It is a text file which
supports Network data with very easy syntax. It is mostly used by
NetworkX, Pajek, yEd, etc.
3. Pajek NET : Pajek (The Solevian word for the spider) is an excellent
computer program for the analysis and visualisation of graphs. It has
extension of .NET or .Paj
4. GraphML : Each GraphML file is written in XML format and defines a

graph.
Emergence of Connectedness : The emergence of connectedness is used to

check whether the graph is connected or not. Assuming a graph with n nodes,
we just need NLogN edges to make the graph connected.
Approach :
1. Taking any random graph of N number of nodes.

2. Let’s assume any node as a starting node.
3. Now add a random edge to the graph and check whether the graph is
connected or not.
4. If the graph is connected, then check if the number edges added and this
will be equal to NlogN.
5. If the graph is not connected, repeat steps 2 and 3.
Eg. people N - 100, even if one edge is removed, the graph is connected
until and unless one edge remains and so on till there is no edge remains. It is a
disconnected graph.
Strong Tie : Strong ties occur between the peoples who have strong
connections, and are in contact on a regular basis. Eg. They are closed ones like
close friends, family members or relatives.
Weak Tie : Weak ties usually have rarely to no contact between two or more
persons on a regular basis. Eg. They mostly are acquaintances, strangers with
common cultural background, colleagues or office mates.
Strong and Weak Relationships are the most important part of one’s social
networking interactions.
Homophily : The Term itself defines the people who tend to get attracted to or
seek out other people who share common interests and who are similar to each
other. It is a principle that a contact between similar persons attracts towards
each other at a higher rate than the dissimilar people.
For eg. If a tourer has a youtube channel who shares his travel recorded videos,
then it is more likely the people who have more interest in traveling and
adventure, tend to attract towards his lifestyle or youtube channel. The people
can admire his videos and may change their lifestyle accordingly.
FatMan Evolutionary Model : There are 3 main concepts which make an
evolutionary model, they are as follows :
1. Homophily : people tend to ….

2. Social influence : People tend to change their lifestyle according to the
influence of other people who share common ideas, behaviour or
properties.
3. Closure : There are 3 kinds of closures
a. Triadic : if a is friend of B and B if of C then A can be friend of C
b. Membership closure : If A and B are in the same club, there are
chances they may become friends.
c. Foci closure : It is the probability of becoming friends when they
have the same foci (commonness).
Clustering Coefficient : Clustering coefficient is a measure of the degree to

which nodes in a graph are likely to group together.
Diffusion : Spreading of information from one place to another through the

medium of interaction.
Modelling Diffusion : Modelling diffusion is the process of the node who decides
to spread information in a network based on payoffs associated with them.

Notes - SN

Uploaded by

Copyright:

Available Formats

Notes - SN

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Notes - SN

Uploaded by

Copyright:

Available Formats

Social Network refers to the use of internet-based social media websites to stay

Types of Social Networking

Famous social networking websites :

Basic inbuilt graph types :

The three primary sources of value for network effects :

Social Embeddedness - It refers to extent to which organisations are connected

2. Synonym - A word’s synonym can be related with another word’s

4. Web graph - Web graph is a social diagram which shows the

1. CSV (Comma separated value) :

2. GML (Graph Modelling Language) :

4. GraphML : Each GraphML file is written in XML format and defines a

Emergence of Connectedness : The emergence of connectedness is used to

1. Taking any random graph of N number of nodes.

1. Homophily : people tend to ….

Clustering Coefficient : Clustering coefficient is a measure of the degree to

Diffusion : Spreading of information from one place to another through the

You might also like