Social Network Analysis: Understanding User Behavior in Threads

Abstract

The presence of threads has enhanced users' ability to share stories, information and views in greater detail and effectively in the online world. Social media platforms allow users to interact, share content, and connect online. Therefore, the purpose of this research is to create a Social Network Analysis (SNA) network visualization and categorize the data to contribute to the understanding of user interaction on this platform. The research method in this study is the Social Network Analysis (SNA) approach to analyze the relationship between individuals in social networks. The results show that user participation in the Threads platform has a significant positive impact. The data shows a high level of participation, providing an understanding of user engagement in various topics. Social network analysis revealed characteristics of user interactions, such as the number of nodes, average connectedness, and relationship complexity. In addition, word categories and network visualizations provide insights into relationship patterns and topics in conversations.

Full text

Kontigensi: Jurnal Ilmiah Manajemen
Vol. 12, No. 1, June 2024, pp. 115-124
ISSN 2088-4877
115
Kontigensi: Jurnal Ilmiah Manajemen
Management Science Doctoral Program, Pasundan University, Bandung, Indonesia
https://creativecommons.org/licenses/by-nc/4.0/
Social Network Analysis: Understanding User Behavior in Threads
1Diana Wahyu Lestari, 2*Rita Ambarwati
Manajemen, Universitas Muhammadiyah Sidoarjo, Sidoarjo, Indonesia
E-mail: ritaambarwati@umsida.ac.id
DOI: https://doi.org/10.56457/jimk.v12i1.505
Received: April 6, 2024
Accepted: May 16, 2024
Published: June 11, 2024
ABSTRACT
The presence of threads has enhanced users' ability to share stories, information and views in greater
detail and effectively in the online world. Social media platforms allow users to interact, share content, and
connect online. Therefore, the purpose of this research is to create a Social Network Analysis (SNA) network
visualization and categorize the data to contribute to the understanding of user interaction on this platform. The
research method in this study is the Social Network Analysis (SNA) approach to analyze the relationship between
individuals in social networks. The results show that user participation in the Threads platform has a significant
positive impact. The data shows a high level of participation, providing an understanding of user engagement in
various topics. Social network analysis revealed characteristics of user interactions, such as the number of
nodes, average connectedness, and relationship complexity. In addition, word categories and network
visualizations provide insights into relationship patterns and topics in conversations.
Keywords: Threads, Twitter, Media Social, Social Network Analysis
INTRODUCTION
The digital era has evolved and the presence
of social media platforms has transformed
society into an information platform where they
can share their views, experiences and opinions
on various topics (Samrin & Akbar, 2023). The
emergence of the internet provides a new space
for people to communicate to the current group
of Indonesians who use social media widely not
only to receive information, but also as a space
to discuss, share information, and communicate
(Rofidah, 2021). Social media users are widely
used by the general public as information,
expressing opinions and other things that attract
their attention, so it can be seen from the many
applications created to fulfill the wishes of the
community (Hadna et al., 2016). Applications
used to access information. One of them is
Threads, Instagram's new social media
platform. The threads feature includes a search
for the hottest and most talked about topics. In
addition, threads provide information about what
is happening (Rhein Rahmahsya Reshany &
Santi Indra Astuti, 2023).
In July, a new social media platform,
Threads: an Instagram app, took the internet by
storm. The emergence of Threads created a lot
of buzz. Threads are considered to be similar to
Twitter. Of course, it made many users curious.
Apparently, in less than a week, Threads
surpassed the 100 million user mark (Ragam et
al., n.d.). Threads are similar to Twitter in that
they allow users to share text-based posts.
However, the appearance and features of
Threads are slightly different from Twitter,
although some features are similar. Threads
has several features such as viewing profiles,
checking mentions, and more (Samrin & Akbar,
2023). Some of the reasons for Threads users
to review the description in the Threads app are
to find a collection of topics of conversation that
are very popular at the moment in Threads.
Threads is basically similar to Twitter, so it is a
microblogging platform (a content that contains
short information in the form of text). It allows
users to express their thoughts and feelings in
text form. Users will get feedback from other
users. The response given can be in the form of
re-sharing or commenting on the text given
(Ragam et al., n.d.).
In previous research (Samrin & Akbar, 2023)
has discussed Sentiment Analysis of Threads

Kontigensi: Jurnal Ilmiah Manajemen
Vol. 12, No. 1, June 2024, pp. 115-124
ISSN 2088-4877
116
Kontigensi: Jurnal Ilmiah Manajemen
Management Science Doctoral Program, Pasundan University, Bandung, Indonesia
https://creativecommons.org/licenses/by-nc/4.0/
Application User Comments on the Google Play
Store Using the Multinomial Naïve Bayes
Algorithm. On the other hand, other studies
have focused on Twitter Interaction Networks in
Customer Engagement in E-Commerce and E-
Health Performance Competition in Indonesia,
using the Social Network Analysis (SNA)
approach in Qualitative Descriptive Research
(Navisha et al., 2023). In addition, there is also
previous research that discusses Twitter as a
medium for self-expression among millennials,
by conducting a qualitative analysis using Johari
Window self-disclosure theory (Mutiara et al.,
2020).
The research being conducted by the author
has similarities with previous research (Navisha
et al., 2023) because both use the Social
Network Analysis (SNA) method. However, the
main difference from the research conducted by
the author is the use of the Multinomial Naïve
Bayes algorithm to perform sentiment analysis.
Meanwhile, previous research focused more on
the Social Network Analysis (SNA) approach in
the type of qualitative descriptive research.
Therefore, the research conducted by the
author combines elements of sentiment analysis
with the SNA approach to provide a more
comprehensive insight into the data under
study. Meanwhile, previous research (Mutiara
et al., 2020) used Johari Window self-disclosure
theory. Thus, the main differences between the
two studies lie in the research methods, data
collection techniques, and theories that serve as
the basis for analysis. This research applies an
assumption that threads users give reviews to
others, so that two nodes with two edges are
formed and connected to each other (Susanto
et al., 2012). Therefore, a suitable method is
needed to analyze the interaction of a large
number of threads user reviews. Social network
analysis (SNA) was chosen because it is able to
process large amounts of data to provide an
overview of user interactions in the network, and
can quantify through network properties
consisting of size, edges, density, modularity,
diameter, average path length, average degree
(Alisya Putri Rabbani et al., 2020).
Although SNA has great potential, there has
been no significant use of Social Network
Analysis (SNA) in the field of information and
communication technology so far. Social media
platforms allow users to interact, share content,
and connect online. Therefore, the purpose of
this research is to create a Social Network
Analysis network visualization and perform data
categorization to contribute to the
understanding of user interactions on these
platforms.
Problem Statement : SNA network analysis
and categorizing data on Review Threads
Research Question : How to make SNA
network visualization and categorize data on
Review Threads?
SDGs Category : This research falls into the
17th category of partnerships for the goals,
namely increasing global partnerships in
achieving sustainable development goals
https://sdgs.un.org/goals/goal17
LITERATURE REVIEW
Social Media
Defined as a software that allows users to
exchange information, share photos, videos,
text, and the like online by utilizing forums or
blogs to create a virtual world space called
social media. Another term for social media is
social networking, where most people in the
world use this device to make it easier to share
and create content (Watie, 2016). Based on this
description, the conclusion that can be drawn
regarding the definition of social media is a
device that utilizes the internet network in
supporting communication activities,
collaboration, interaction, sharing, and
expressing themselves with other users to
create virtual social ties (Puspitarini & Nuraeni,
2019).
The features available in social media are
communication features (voice, video, chat,
images, tags, like, share, tweet/retweet, and
many more). Many people from various
backgrounds, ages, as well as individual and
organizational aspects, use social media to
exchange information. With the increasing
number of social media applications in society,

Kontigensi: Jurnal Ilmiah Manajemen
Vol. 12, No. 1, June 2024, pp. 115-124
ISSN 2088-4877
117
Kontigensi: Jurnal Ilmiah Manajemen
Management Science Doctoral Program, Pasundan University, Bandung, Indonesia
https://creativecommons.org/licenses/by-nc/4.0/
social media communication networks are also
increasing. Social interaction plays an important
role including communication in work,
education, business, and many other fields
(Ariyanti, 2022).
Google Play Store
A device that provides google digital content
in various categories ranging from books,
music, movies, games, applications, and the like
is called the google play store. Users can
access this software by using the android
application or website. Google play store also
features reviews and ratings from other users
(Herlinawati et al., 2020). Google Play Store has
various features. One of them allows users to
leave reviews about the application. Thus
allowing other users to get information from
these reviews (Nishfi et al., 2023). This feature
can influence potential users because the
tendency of users before downloading will see
the review section as a benchmark for whether
the product is good or not.
App Store
The media provider of applications needed
by the iPhone OS (IOS) based Operation
System is the App Store. The party that
manages and develops the app store is Apple
Inc. The existence of this service makes it easier
for Apple users to browse and download the
applications they need. The App Store has
applications organized by categories such as
games, learning, and entertainment (Sandag,
2020). In addition, there are free and paid
applications available. The categories on this
app allow users to easily search for the apps
they need (Hadi et al., 2020).
Social Network Analysis (SNA)
The scientific field that studies the
relationship between the application of theory
and humans is called SNA (Social Network
Analysis). SNA is described through a network
representation where information sharing in a
social network is based on two fundamental
elements, namely edges (relationships) and
nodes (actors). The relationships formed
between nodes can be analyzed in detail by
involving visualization to obtain accurate and
relevant information to the interests of users
(Akbar et al., 2022). The function of SNA is to
facilitate the mapping of relationships to
optimize the formation of knowledge
management. The components contained in the
SNA network include average path length,
diameter, average degree, edges, and nodes
(Bratawisnu & Alamsyah, 2019).
METHOD
Social Network Analysis (SNA) technique is
the method used in this research. SNA is
associated with understanding social
relationships that involve the correlation of
edges (correlation between users as lines) and
nodes (correlation between users as points)
(Bratawisnu & Alamsyah, 2019). One type of
qualitative research is research whose purpose
is to analyze events or events in terms of
actions, motivations, behaviors, perceptions,
and others in detail expressed in the form of a
series of words (Hukum et al., 2013).
Researchers obtained secondary data sources
in the form of data covering review content. This
research requires the use of several supporting
programs, including jupyter notebook, Snscrape
library, notepad++, wordij and gephi. The
concept of the research flow is as follows:

Kontigensi: Jurnal Ilmiah Manajemen
Vol. 12, No. 1, June 2024, pp. 115-124
ISSN 2088-4877
118
Kontigensi: Jurnal Ilmiah Manajemen
Management Science Doctoral Program, Pasundan University, Bandung, Indonesia
https://creativecommons.org/licenses/by-nc/4.0/
Figure 1. Research flow
Problem Identification
Observe and discover the Threads
phenomenon
Data Collection
This stage of the research includes data
collection. The purpose of data collection is to
gather information that will be used to analyze a
research. Running tools, setting up searchable
subjects, and using text codes to create articles
that are retrieved by the Google Colab app are
the steps involved in data collection. Next, the
researcher used crawling coding, written in
Python, and used Gephi tools to build a social
network model from the collected data. The
researcher searched for datasets through
Kaggle that were relevant to the subject.
Data Pre-Processing
Data preprocessing is an important part of
preparing raw data for social network analysis.
This process involves various methods to
transform the raw data into a more structured
and analyzable format. In addition, the
preprocessing process on social media data
such as Twitter also includes things like
changing the base word form, removing
unimportant words, removing affixes, and
conjunctions from the tweet document. The
process of preprocessing:
a. Case Folding
Case Folding is where all characters
are converted to lowercase. The purpose of
case folding is to eliminate the inconsistent
use of uppercase and lowercase letters in
user reviews. This process is done
removing unused characters from the data
such as (&,/,*,0,(),etc).
b. Tokenizing
Tokenizing is separating words in a
sentence for further text analysis process.
c. Stopword Removal
The stage of removing words that are
meaningless and unrelated but often
appear is called stopword removal.
d. Normalize
A series of activities that are part of the
research stages that require researchers to
change abbreviated words to the original
word form and make standard words from
non-standard words is called word
normalization. How to find the word is
classified as a standard word or not through
KBBI (Kamus Besar Bahasa Indonesia),
where researchers must check whether the
word is listed in KBBI or not. Then the word
change is based on the data set used in the
research.
Network Modeling
Related to the stage of reprocessing data
that has been processed by applying the Gephi
application. This application makes it easy to
visualize the network model by implementing an

Kontigensi: Jurnal Ilmiah Manajemen
Vol. 12, No. 1, June 2024, pp. 115-124
ISSN 2088-4877
119
Kontigensi: Jurnal Ilmiah Manajemen
Management Science Doctoral Program, Pasundan University, Bandung, Indonesia
https://creativecommons.org/licenses/by-nc/4.0/
undirected graph that ignores the direction of the
relationship.
Network Property Identification
Each network model processed by the Gephi
application has several properties that will be
calculated. The network properties that will be
calculated are: nodes, edges, average degree,
modularity, and modularity.
RESULT and DICUSSION
The crawling results in Table 1 show the
number of tweets in threads, illustrate the level
of user participation and engagement in various
topics, provide insights into conversation trends,
communication patterns, and the potential
impact and dynamics of social networks; while
further analysis can provide an in-depth
understanding of the level of interaction, user
responses, and characteristics of conversations
on the platform.
Table 1. Data Retrieval Results
Focus
Source
Date
Amount of Data
Review
Threads
July-August 2023
8.140
Table 1 shows that the number of tweets
obtained from crawling threads with the
keywords to be researched focuses on koten
threads such as reviews. This study collected
data from July to August 2023, as much as at
least 1,000 data per year for each criterion,
researchers only got the data in 2023 because
the application became popular in July 2023, so
researchers explored related to the application
“threads” focusing on reviews. Then processed
using Jupyter Notebook software. The data was
obtained from the review keyword as much as
8,140 data in July to August 2023.
After collecting data (crawling), the next step
is to preprocess the data. the help of a dictionary
that aims to eliminate words that are not relevant
to the analysis process, this process is called
the filtering process, which removes unused
words and then processes them using wordij
(Anjani & Ambarwati, 2023)
Table 2. Data Preprocessing Results
Focus
Date
Amount of Data
Unique
Average
Review
July-August 2023
28.265
1.375
20,5564
Table 2 shows the results of data
preprocessing using Wordij. This data was
obtained from the preprocessing done using
Jupyter Notebooks, which was saved as a CSV
file. The next process involved reprocessing the
data using the Wordij application, which enabled
data visualization using Gephi. The Wordij
application produces various output files,
including files such as net, stp, stw, wrd, wtg, pr,
and so on (Chamila & Sukmono, n.d.). In the
visualization stage, the data used comes from
the stw.csv file. This file contains information
about the total number of words that appear
from the keyword “review” threads in 2023", with
a total of 28,265 words. Of these, there are
1,375 unique words, and the average word that
appears is 20.5564. Visualization of this data is
an important step after the preprocessing
process to provide a clearer picture of patterns
or trends that may be present in the data.
The analysis method applied in this research
is Social Network Analysis (SNA). SNA is used
to analyze social media activities, especially in
modeling interactions. The interaction is
represented as a network of relationships
between users, represented by nodes and
edges. This analysis is important because it
provides a deeper understanding of the social
interaction patterns of individuals or
communities. Some of the network properties
used in Social Network Analysis (SNA) involve
nodes, edges, average degree, average
weighted degree, diameter, modularity and
average path length (Azmi et al., 2021). These
properties help in mapping the relationships in
the network, thus making a significant
contribution in improving knowledge
management and understanding of the
dynamics of social interactions in the context of
reviewing “threads” applications in 2023.

Kontigensi: Jurnal Ilmiah Manajemen
Vol. 12, No. 1, June 2024, pp. 115-124
ISSN 2088-4877
120
Kontigensi: Jurnal Ilmiah Manajemen
Management Science Doctoral Program, Pasundan University, Bandung, Indonesia
https://creativecommons.org/licenses/by-nc/4.0/
Table 3. Review Network Properties
NO
Network Properties
Value (Review)
1
Nodes
276
2
Edges
1317
3
Average Degree
2,858
4
Average Weighted Degree
29,383
5
Network Diameter
7
6
Modularity
0,217
7
Average Path Length
2,356
Nodes are representations of user actors in
the review social network. The data in table 3
shows that there are 276 nodes that interact with
each other in the social network, with a total of
1317 edges or relationships. Average Degree
shows the average number of connections
between each nodes and other nodes. In the
social network review, nodes have an average
connectedness of 2,858 (Digital et al., 2022).
Average Weighted Degree which is the average
weighted relationship between nodes, reaches
29.383 on the review network (Putri et al., 2023).
Network Diameter is the farthest distance
between two nodes, in the review social network
it reaches 7. Modularity reflects the ability of
actors or users to form different groups in the
network. In the review social network, the
modularity value is 0.217. Furthermore,
Average Path Length, which is the average
number of nodes that must be passed to reach
a certain node, on the review social network has
a value of 2.356 (Bratawisnu et al., 2018).
Data Categories
The category analysis process in this study
was categorized based on certain criteria, and
the results of the analysis of the relationship
between words formed four main categories,
namely App, Action, Reaction, and Object. Then
grouped by these categories, reflecting the
general characteristics of the review threads
observed in this study.
Table 4. Data Category Identification
Topic1:App
Topic2:Action
Topic3:Reaction
Topic4:Object
app
10,6%
Delete
0,85%
nice
1,44%
people
1,02%
twitter
6,62%
Follow
0,83%
love
1,06%
time
0,66%
instagram
3,20%
Option
0,76%
bad
0,07%
features
0,56%
threads
2,74%
Post
0,72%
amazing
0,62%
data
0,51%
account
2,13%
download
0,50%
experience
0,53%
bugs
0,35%
thread
0,79%
Login
0,45%
hope
0,39%
phone
0,34%
application
0,69%
Review
0,42%
cool
0,39%
log
0,33%
meta
0,67%
Feed
0,38%
easy
0,35%
lot
0,31%
social
0,53%
Worst
0,33%
properly
0,27%
platform
0,31%
insta
0,48%
Posts
0,33%
awesome
0,27%
version
0,30%
Table 4 shows 10 words for each topic. "App"
(Topic 1), "Action" (Topic 2), "Reaction" (Topic
3), Object (Topic 4). Then, "App" (Topic 1) is
categorized as a word because it refers to a
specific field or topic, in this case related to
technology or software. For example, the words
"app" (10.6%), "twitter" (6.62%), "instagram"
(3.20%), "threads" (2.74%) appear most
frequently. Meanwhile, "Action" (Topic 2) is
included in the category of words because its
meaning is related to activities or actions,
including words such as "delete"(0.85%),
"follow"(0.83%), "option"(0.76%),
"post"(0.72%). The use of these words provides

Kontigensi: Jurnal Ilmiah Manajemen
Vol. 12, No. 1, June 2024, pp. 115-124
ISSN 2088-4877
121
Kontigensi: Jurnal Ilmiah Manajemen
Management Science Doctoral Program, Pasundan University, Bandung, Indonesia
https://creativecommons.org/licenses/by-nc/4.0/
a deep understanding of the activities that
dominate conversations and the extent to which
user interactions involve actions.
Furthermore, "Reaction" (Topic 3) belongs to
the category of words as it refers to a response
or response to a stimulus or event. It can be
used in various contexts, such as in science,
psychology, and everyday conversation, to
describe how someone or something responds
or reacts to a situation or change. For example,
in the words "nice"(1.44%), "love"(1.06%),
"bad"(0.07%), "amazing"(0.62%). Then,
"Object" (Topic 4) belongs to the category of
words because it can refer to a physical object
or concept that is the focus of attention. This
word has many usages involving various
contexts, such as in programming languages,
math, or everyday conversation, where "object"
can refer to something concrete or abstract that
is the topic of discussion or attention. For
example, "people"(1.02%), "time"(0.66%),
"features"(0.56%), "data"(0.51%)
Network Model Visualization
The next process is data visualization, the
concept used in Social Network Analysis (SNA)
is graph theory, consisting of nodes (points)
connected by edges (lines). The visualization
process requires supporting tools through Gephi
software.
Figure 1. Visualization of Review Network
The network visualization in Figure 1
provides insight into the review topics in the
Threads application. The graph is structured
with five large nodes providing a strong visual
representation of the centers of conversation
among Threads users. These main clusters,
marked with large nodes, include key words
such as (“app”, “twitter”, “instagram”, “threads”,
and “account”). This gives an idea of the main
focus of the conversation, which is clearly
centered around social media platforms and app
features. The different colors of the nodes and
edges play a key role in unlocking additional
layers of information. With five different colors:
purple, gray, green, blue and orange. This
visualization signals specific groups and their
relevance to the discussion. The largest group,
colored purple, shows a primary focus on social
media platforms, with words like (“app”, “twitter”,
“meta” and “application”). This suggests that the
discussion will include consideration and
comparison of various features and functions of
social media applications, such as Threads.
The second group, marked in green,
emphasizes how important millennials are in the
conversation. Words like (“instagram”,

Kontigensi: Jurnal Ilmiah Manajemen
Vol. 12, No. 1, June 2024, pp. 115-124
ISSN 2088-4877
122
Kontigensi: Jurnal Ilmiah Manajemen
Management Science Doctoral Program, Pasundan University, Bandung, Indonesia
https://creativecommons.org/licenses/by-nc/4.0/
“threads”, “account”, “options”, “login”,
“password”, “profile”) show that these
millennials are active in using the app and
exploring the options available. The third group,
which focuses on interaction and response, is
represented by the color blue and includes
words such as (“people”, “follow”, “feed”,
“posts”, “timeline”) This shows that the
conversation is not only limited to technical
aspects, but also involves social and inter-user
interactions. The significance of grey and
orange on nodes and edges is that despite the
similarity in colors, there is no interaction or
relationship talked about between grey and
orange. This shows that nodes are connected
not only using the same color, but also using
different colors. For example, gray (“mode”) and
orange (“platform”) nodes indicate different
topics in a given context.
As such, this visualization is not just a
network map, but also a visual narrative of how
Threads users engage in complex and layered
discussions around the app. It provides deep
insights into various aspects that include user
experience, social interactions, and feature
comparisons of social media platforms. The
Threads app, as a social media sharing product
from Facebook designed specifically for
Instagram users, takes the concept of sharing
photos, videos, text messages and stories of
close friends to a more intimate level. While the
ability to share multimedia content is a common
feature across platforms, Threads stands out
with its focus on more targeted and personal
interactions between users.
Content review threads, threads are
basically similar to twitter, threads are content
that contains short information in the form of
text. It was revealed that this application
received significant attention from the millennial
generation. The rapid increase in users,
reaching more than 100 million in less than a
week, is a clear indicator that Threads is able to
accommodate the needs and desires of
millennial users. The phenomenon of Threads
not only creates an impressive number of users,
but also stimulates a discussion space full of
social dynamics and interactions (Ragam et al.,
n.d.). In the context of visualization “threads”
indicate the intensity of related discussions,
which means that social media platforms to
thread content that often appear and are
related. The two nodes “threads” and
“Instagram” indicate how often “threads” and
“Instagram” are used in related applications.
This can provide insight into user trends and
interests in conversations on these social media
platforms. Thus, the threads application has
significant value in various aspects of business
and product development. This is in line with
research that has stated that users provide an
in-depth view of the product (Samrin & Akbar,
2023).
In social network analysis, the nodes that
often appear are the platforms “Twitter” and
“Instagram”, which are frequently used by
millennials. The focus is on understanding how
information and interactions spread among
users, including community identification,
influence, retweet or share patterns, and
analysis of shared content. The research being
conducted by the author has similarities with
previous research (Navisha et al., 2023)
because both use the Social Network Analysis
(SNA) method. However, the main difference
from the research conducted by the author is the
use of the Multinomial Naïve Bayes algorithm to
perform sentiment analysis. Meanwhile,
previous research focused more on the Social
Network Analysis (SNA) approach in a
qualitative descriptive research type. In the
previous study (Navisha et al., 2023), the main
emphasis was on using the Social Network
Analysis (SNA) method to understand
interaction patterns and user behavior on social
media platforms that identify user communities
of “twitter” and “instagram” applications,
measure the influence between users, analyze
retweet or share patterns, and conduct analysis
of shared content to understand certain trends
or patterns in online interactions.
Along with previous studies, the research
being conducted by the author also uses the
Social Network Analysis (SNA) approach to
understand the dynamics of social networks on
the “Twitter” and “Instagram” platforms.

Kontigensi: Jurnal Ilmiah Manajemen
Vol. 12, No. 1, June 2024, pp. 115-124
ISSN 2088-4877
123
Kontigensi: Jurnal Ilmiah Manajemen
Management Science Doctoral Program, Pasundan University, Bandung, Indonesia
https://creativecommons.org/licenses/by-nc/4.0/
However, the main difference lies in the focus of
the analysis. While the previous study
highlighted more on network description and
mapping as well as user behavior. In general,
the author's research adds the dimension of
sentiment analysis by using the Multinomial
Naïve Bayes algorithm. This allows the authors
to not only understand the structure and
dynamics of the network, but also to explore
how users respond to and perceive the content
they use or share on the platform.
Thus, while there are similarities in the use
of Social Network Analysis (SNA) methods, the
authors' research adds value by expanding the
scope of the analysis into the sentiment domain,
which can provide more comprehensive insights
into user behavior and preferences on the
"Twitter" and "Instagram" platforms.
CONCLUSION
The results show that user participation in
the Threads platform has a significant positive
impact. The data shows a high level of
participation, providing an understanding of user
engagement in various topics. Social network
analysis revealed characteristics of user
interactions, such as the number of nodes,
average connectedness, and relationship
complexity. In addition, thematic word
categories and network visualizations provide
insight into the patterns of relationships between
keywords and topics in conversations. In light of
this, the Threads platform has rapidly gained
popularity, exceeding 100 million users in a
short period of time, demonstrating the potential
for user growth and becoming the first choice for
users to interact by sharing social content.
However, this study has limitations in the
data available, the analysis methods used, and
the overall research results. Although data
mining was done accurately, the limited data
may not cover all topics in Threads, affecting the
representation of the research results. The use
of Social Network Analysis (SNA) method, and
Multinomial Naïve Bayes algorithm may have
limitations in capturing the complexity of social
interactions and it is difficult to generalize the
research results. Therefore, the development of
more modern analysis methods and wider data
collection and validation of research results are
required. This reinforces the importance of
understanding the limitations of the study and
encourages further research to provide deeper
insights for decision makers in designing more
effective social media strategies.
REFERENCES
Akbar, M. A., Masniarara Aziza Balfas Amril, Raiza
Syahira, Fahrein Rachel Latisha, & Noor Jihan.
(2022). Analisis Struktur Jaringan Komunikasi
#Seagames2022 Di Twitter Menggunakan
Pendekatan Social Network Analysis (Sna).
Jurnal Studi Komunikasi Dan Media, 26(1), 1–
16. https://doi.org/10.17933/jskm.2022.4780
Alisya Putri Rabbani, Andry Alamsyah, & Sri
Widiyanesti. (2020). Analisa Interaksi User Di
Media Sosial Mengenai Industri Fintech
Menggunakan Social Network Analysis (Studi
Kasus: Gopay, Ovo Dan Linkaja). Jurnal Mitra
Manajemen, 4(3), 341–351.
https://doi.org/10.52160/ejmm.v4i3.352
Anjani, S. P., & Ambarwati, R. (2023). Analysis Of The
Credibility Of Utilizing Brand Ambassador To
Compete Between Shopee And Tokopedia
Using Social Network Analysis On Twitter.
Jurnal Manajemen, 14(3), 391–404.
https://doi.org/10.32832/jm-uika.v14i3.13777
Ariyanti, Y. D. P. (2022). Analisis Centrality Aktor pada
Penyebaran Informasi Kuliner di Media Sosial
dengan menggunakan Social Network Analysis.
Journal of System, Information Technology and
Electronics Engineering (J-SITEE), 2(1), 23–31.
Azmi, N. A., Fathani, A. T., Sadayi, D. P., Fitriani, I., &
Adiyaksa, M. R. (2021). Social Media Network
Analysis (SNA): Identifikasi Komunikasi dan
Penyebaran Informasi Melalui Media Sosial
Twitter. Jurnal Media Informatika Budidarma,
5(4), 1422.
https://doi.org/10.30865/mib.v5i4.3257
Bratawisnu, M. K., & Alamsyah, A. (2019). Social
Network Analysis Untuk Analisa Interaksi User
Di Media Sosial Mengenai Bisnis E-Commerce.
Sosiohumanitas, 21(1), 63–69.
https://doi.org/10.36555/sosiohumanitas.v21i1.
1000
Bratawisnu, M. K., Putra, M. R. D., & Ignatio, W.
(2018). Penentuan Top Brand Menggunakan
Social Network Analysis pada e-commerce
Tokopedia dan Bukalapak. Journal of

Kontigensi: Jurnal Ilmiah Manajemen
Vol. 12, No. 1, June 2024, pp. 115-124
ISSN 2088-4877
124
Kontigensi: Jurnal Ilmiah Manajemen
Management Science Doctoral Program, Pasundan University, Bandung, Indonesia
https://creativecommons.org/licenses/by-nc/4.0/
Information Engineering and Educational
Technology, 2(1), 1.
https://doi.org/10.26740/jieet.v2n1.p1-5
Chamila, A. N., & Sukmono, R. A. (n.d.). Analysis Of
Twitter User Interactions Using Social Networks
At Payday Sale Events On The Marketplace [
Analisis Interaksi Pengguna Twitter
Menggunakan Social Network Pada Event
Gajian Sale Di Marketplace ]. Universitas
Muhammadiyah Sidoarjo, 1–12.
Digital, P. P., Kualitas, P., Citra, D. A. N., Terhadap,
M., Pembelian, K., & Lokal, S. (2022). Journal of
Applied Business Administration. September,
90–100.
Hadi, S. W., Julianto, M. F., Rahmatullah, S., & Gata,
W. (2020). Analisa Cluster Aplikasi Pada App
Store Dengan Menggunakan Metode K-Means.
Bianglala Informatika, 8(2), 86–90.
https://doi.org/10.31294/bi.v8i2.8191
Hadna, N. M. S., Santosa, P. I., & Winarno, W. W.
(2016). Studi Literatur Tentang Perbandingan
Metode Untuk Proses Analisis Sentimen Di
Twitter. Seminar Nasional Teknologi Informasi
Dan Komunikasi, 2016(March), 57–64.
Herlinawati, N., Yuliani, Y., Faizah, S., Gata, W., &
Samudi, S. (2020). Analisis Sentimen Zoom
Cloud Meetings di Play Store Menggunakan
Naïve Bayes dan Support Vector Machine.
CESS (Journal of Computer Engineering,
System and Science), 5(2), 293.
https://doi.org/10.24114/cess.v5i2.18186
Hukum, K., Kampung, M., Memiliki, M. U., Atas, S.,
Ulayat, H., Metodologi, B. I., Pendekatan, P. A.,
& Penelitian, M. (2013). Metodelogi Penelitian.
66–79.
Lolo, P. D., & Nurdiansyah, N. (2023). Analysis of
Consumer Behavior Towards Product Groups at
Indomaret Sudiang Raya. Kontigensi: Jurnal
Ilmiah Manajemen, 11(2), 491-499.
Mutiara, Z., Antonius, B., & Leviane, J. H. L. (2020).
Twitter Sebagai Media Mengungkapkan Diri
Pada Kalangan Milenial. Fakultas Ilmu Sosial
Dan Politik, Universitas Sam Ratulangi, 1–8.
https://ejournal.unsrat.ac.id/index.php/actadiurn
akomunikasi/article/view/31979/30381
Navisha, A. N., Ambarwati, R., & Hariasih, M. (2023).
Twitter Social Network Interaction As Customer
Engagement In Competition For E-Commerce
E-Health Performance In Indonesia. Jurnal
Manajerial, 10(02), 303.
https://doi.org/10.30587/jurnalmanajerial.v10i02
.5279
Nishfi, D., Huda, I., & Prianto, C. (2023). Analisis
Sentimen Layanan Jasa Pengiriman Pada
Ulasan Play Store : Systematic Literature
Review. 04(02), 87–98.
Puspitarini, D. S., & Nuraeni, R. (2019). Pemanfaatan
Media Sosial Sebagai Media Promosi. Jurnal
Common, 3(1), 71–80.
https://doi.org/10.34010/common.v3i1.1950
Putri, F. A., Ambarwati, R., & Kusuma, K. A. (2023).
Marketing Strategy of Fast-Food Restaurants in
Indonesia - Sosial Network Analysis on Twitter.
Jurnal Manajerial, 10(3), 561–576.
https://doi.org/10.21070/ups.2461
Ragam, E., Digital, K., & Rosadi, N. (n.d.). The
Existence of Inlis Variety in Digital
Communication in Threads : an Instagram App.
Rhein Rahmahsya Reshany, & Santi Indra Astuti.
(2023). Media Sosial Twitter sebagai Sarana
Mendiskusikan Kasus Kekerasan Seksual.
Bandung Conference Series: Journalism, 3(1),
34–43. https://doi.org/10.29313/bcsj.v3i1.6224
Rofidah, L. (2021). Analisis Komunikasi Konflik
Netizen dalam Pemaknaan Feminisme pada
“Thread Bekal Buat Suami.” Jurnal Interaksi :
Jurnal Ilmu Komunikasi, 5(2), 133–149.
https://doi.org/10.30596/interaksi.v5i2.5899
Samrin, N., & Akbar, M. N. (2023). Analisis Sentimen
Komentar Pengguna Aplikasi Threads Pada
Google Play Store Menggunakan Algoritma
Multinomial Naïve Bayes. 3(2), 1–9.
Sandag, G. A. (2020). Prediksi Rating Aplikasi App
Store Menggunakan Algoritma Random Forest.
CogITo Smart Journal, 6(2), 167–178.
https://doi.org/10.31154/cogito.v6i2.270.167-
178
Susanto, B., Lina, H., & Chrismanto, A. R. (2012).
Penerapan Social Network Analysis dalam
Penentuan Centrality Studi Kasus Social
Network Twitter. Jurnal Informatika, 8(1).
https://doi.org/10.21460/inf.2012.81.111
Watie, E. D. S. (2016). Komunikasi dan Media Sosial
(Communications and Social Media). Jurnal The
Messenger, 3(2), 69.
https://doi.org/10.26623/themessenger.v3i2.27
0