Journal of Theoretical and Applied Information Technology
30
th
November 2021. Vol.99. No 22
© 2021 Little Lion Scientific
ISSN:
1992-8645
www.jatit.org
E-ISSN:
1817-3195
5460
BIG DATA ANALYSIS ON YOUTUBE WITH TABLEAU
1
JOHANES FERNANDES ANDRY,
2*
HENDY TANNADY,
3
ISABELLE IVANA LIMAWAL,
4
GLISINA DWINOOR REMBULAN,
5
RUSTONO FARADY MARTA
1
Department of Information System, Universitas Bunda Mulia, Jakarta, Indonesia 14430
2*
Department of Management Graduate Program, Institut Teknologi dan Bisnis Kalbis, Jakarta, Indonesia
13210
3
Department of Information System, Universitas Bunda Mulia, Jakarta, Indonesia 14430
4
Department of Industrial Engineering, Universitas Bunda Mulia, Jakarta, Indonesia 14430
5
Master's Degree of Communication Science Department, Post-Graduate Studies, Universitas Bunda Mulia,
Jakarta, Indonesia, 14430
E-mail:
1
jandry@bundamulia.ac.id,
2
*hendytannady@gmail.com,
3
isabelleivana@gmail.com,
4
grembulan@bundamulia.ac.id,
5
rmarta@bundamulia.ac.id
ABSTRACT
YouTube is the second biggest search engine on the web and a platform with features where users can post,
view, comment, and link to videos. YouTube has more than a billion active users where users can see
Recommended Channels, which are based on videos that users watch frequently. Furthermore, Trending is
based on video trending on the number of clicks the video gets every day. This study aims to provide insight
into how understanding and implementation of Big Data on YouTube. In the current era, Big Data is gaining
much interest because of the opportunities and benefits felt to be unprecedented. Big data analytics can
leverage business change, enhance decision-making by applying advanced analytic techniques on big data,
and reveal hidden insights and valuable knowledge. To better understand the changes brought about by big
data, this paper focuses on data analysis using data visualization. In analyzing YouTube trending videos, we
use tableau software to help visualize the data that has been obtained. The following are some of the charts
used in this analysis, such as pie charts, area charts, horizontal bars, highlight tables, treemaps, and mapping.
Data Visualization can help YouTube, and its users identify which areas need to be improved, which factors
affect users' satisfaction and dissatisfaction, and what to do and know to develop parts lacking and pay
attention to today's users' needs. Visualized data can give a better prediction of current trending topics or
videos and future growth. It can be used to influence our decision-making.
Keywords:
Big Data, YouTube, Data Analytics, Data Visualization
1.
INTRODUCTION
Social network data, audio, video in social media,
and social networking sites are ample data examples.
One example is YouTube. This platform was
founded in 2005 and developed rapidly to become
the most crucial video-sharing website on the web
[1]. Following its acquisition by Google, the location
has continued to grow in popularity. It is now one of
the top-visited sites globally and attracts 800 million
unique visitors per month [2]. Where users can
upload videos, watch various videos available on the
platform with various topics. This platform provides
various facilities to interact with one another and
answer certain video content. Using big data, users
can use the facilities provided, like users can see
Recommended Channels, where users can view
videos recommended supported videos that users
often watch. Also, there is Trending, where videos
that are being watched by many users are often seen
in Trending. There are also Recommended Videos,
which are videos that are recommended, supported
videos that users watch. YouTube has great potential
to succeed in potential consumers, media, and
interest groups. With numerous users using the
platform, indirectly creating new data are often
processed to supply a replacement analysis result. In
Internet usage, each user automatically collects large
amounts of knowledge a day. This data is stored
directly by the physical device. It is more feasible for
giant amounts of storage to store within the cloud
than a locally available hard disc. Because storing on
the hard disc, the danger of losing data is bigger than
storing within the cloud. Around 2016, active
internet users reached 3,424,971,237 worldwide.
Users of social sites or social media are increasing a
day and leading to data collection. Google and
YouTube have about 15 exabytes of knowledge, and
Facebook and Twitter have tons of knowledge
collection. The info collected is getting bigger and
Journal of Theoretical and Applied Information Technology
30
th
November 2021. Vol.99. No 22
© 2021 Little Lion Scientific
ISSN:
1992-8645
www.jatit.org
E-ISSN:
1817-3195
5461
larger because it has increasingly collected
everywhere - sensing mobile information devices,
antennas, sensory technology, software logs,
cameras, microphones, frequency identification
readers, and wireless sensor networks. Supported
world technology, information storage capacity has
doubled every 40 months since the 1980s; as of 2012.
A day the info has been generated about 2.5 trillion
[3].
Big data provides organizations with insights into
their business processes and enables them to market
their products better by employing analytical
techniques like data mining [4]. The method of
extracting valuable information and data from large
ones is named "Data Mining." The platform uses big
data technology to take care of its filing system. The
quantity of knowledge generated from the start of
our time until the last decade is perhaps generated
every few days today [5]. Altogether probability, this
trend is about to continue because of the increase in
social media applications. Simple access to data
through mobile devices exponentially increases the
quantity of knowledge captured every second [6].
The platform has billions of individuals who
connect, inform, and encourage the planet by making
videos a day. Therefore, it is not surprising that
YouTube now features a significant influence on the
web but has severe problems. Archiving, processing,
and efficient analysis of sizeable data is extremely
difficult. The info generated by the billions of
YouTube videos is appropriate. Consistent with
statistics published by Google, YouTube has quite
one billion customers. Also, as many as 33% of
individuals spend their time watching YouTube
because the platform has around 300 hours of video
a day and may even exceed the amount published by
Google [7]. Social media's exponential growth in
contemporary society makes them necessary tools
for communication, content creation, sharing, and
business growth [8].
The authors will analyze daily statistics for
trending videos where the dataset comes from
Kaggle.com. To work out which videos are trending
and are on the highest list this year, YouTube uses
various
factors,
including
measuring
user
engagement (trending dates, total videos, region, and
category). The info to be analyzed is within the USA,
France, Russia, India, and Japan, with up to 200
trending videos registered per day. Each region data
is during a separate file. Data includes trending
dates, total videos, region, and category. The authors
will visualize data that has been taken from
Kaggle.com.
Data visualization is the use of natural human
skills to enhance processing and organization
efficiency. Data visualization represents data or
information during a graphic, chart, or other visual
formats. Visualization can help us affect more
complex information and enhance memory [9]. It
communicates the relationships of the info with
images. This is often vital because it allows trends
and patterns to be more easily seen. With the increase
of massive data, we would like to interpret
increasingly larger batches of knowledge. Machine
learning makes it easier to conduct analyses like
predictive analysis, which may then function helpful
visualizations to present. In this sense, data
visualization is not only crucial for data scientists
and data analysts, but all people in any career got to
understand data visualization.
It is often used interchangeably with information
graphics,
statistical
charts,
and
knowledge
visualization. We want data visualization because
that visual summary information makes it easier to
spot patterns and trends than to look through the
thousands of rows on a spreadsheet. That is how the
human brain works. Since the aim of knowledge
analysis is to realize insight, data is far more valuable
when visualized. Albeit a knowledge analyst can pull
insights from data without visualization, it will be
tougher to speak the meaning without visualization
[10]. Data visualization may be a big field with many
disciplines.
It
is
precisely
due
to
this
interdisciplinary nature that the visualization field is
filled with vitality and opportunities. To research and
visualize the info, the authors are going to be using
Tableau Software.
Tableau is an end-to-end data analytics platform
that permits us to prep, analyze, collaborate, and
share our significant data insights. Tableau excels in
self-service visual analysis, allowing people to ask
new questions of governed big data and share those
insights across the organization [11]. Tableau works
closely with the leaders during this space to support
whatever platform the customer chooses. Tableau
allows users to seek out this value in company data
and investments within the technology to get the
most out of their data. Tableau can help businesses
see and understand Big Data. During this time,
efficient analysis of business data and their storage
on appropriate devices is tedious. The YouTube
videos and, therefore, the multimedia data, which are
generated from it, are usually unstructured. Perfect
analysis of those semi-structured and unstructured
multimedia data may be a big challenge. This study
aims to supply insight into how understanding and
implementation of massive Data on YouTube
Journal of Theoretical and Applied Information Technology
30
th
November 2021. Vol.99. No 22
© 2021 Little Lion Scientific
ISSN:
1992-8645
www.jatit.org
E-ISSN:
1817-3195
5462
2.
LITERATURE REVIEW
This section will provide the foundation and
support for a new insight that authors contribute.
However, a literature review's main target is to
summarize and synthesize others' arguments and
ideas without adding original contributions.
2.1
Big Data
The term "Big Data" appeared for the primary
time in 1998 on the slide deck Silicon Graphics
(SGI) by John Mashey under the title "Big Data and
therefore the NextWave of InfraStress." The word
Big Data is very relevant from the start, as the first
book to mention "Big Data" is a mining data book
that appeared in 1998 by Weiss and Indrukya. For
the first time, an academic paper with the inscription
"Big data" in the title appeared sometime later in
2000 in a Newspaper by Diebold. The origin of the
term "Big Data" is that we are currently creating a
file of data every day. Big data can also be
interpreted as data sets of sizes beyond ordinary
software capabilities used to capture, curate,
manage, and process data in considerable time [12].
Big data is somewhat unusual in that it was broadly
accepted in the commercial and public space before
the academic discourse has had time to catch up [13].
This may explain why most of the literature on big
data has increased during the past few years.
Big data is not new, but the rapid rate of adoption
in recent times may make it appear so [14]. Sensors,
mobiles, and social media networks are just a couple
of samples of modern digital technologies that have
penetrated our daily lives. An excessive amount of
digital data is being generated a day. Nowadays, an
increasing number of knowledge silos are created
globally, suggesting that this growth will never stop.
These data are not only voluminous but also
continuous, streaming, real-time, dynamic, and
volatile [15]. Data have grown exponentially in
"volume", variety' and "velocity" [16], [17]. This is
generally described by the "3 V" of big data (volume,
variety, and velocity), which are used to describe this
phenomenon. Literature indicates that big data can
unlock plenty of new opportunities and deliver
operational and financial value [18], [19]. Internet
advent has triggered a boom in information research.
Companies are flooded by the wealth of knowledge
that results from simple internet browsing.
2.2
YouTube
YouTube was founded in 2005 and developed the
platform rapidly to become the largest video-sharing
website on the internet [1]. Following its acquisition
by Google, the location has continued to grow in
popularity and now attracts almost half a billion
unique users per month. The location allows users to
upload a vast number of video clips, which may be
viewed and linked to anyone. YouTube provides
various facilities to permit registered users to interact
with one another and share their content responses.
One option is for users to post direct video responses
to what they need to be viewed. Other options are
afforded by the user profiles, or "channels." There
are personal messaging service and a more visible
commenting option on the profile, and users can
prefer to store and display lists of their favorite
videos.
They can also befriend other users on the site or
become "fans" by subscribing to receive alerts once
they post new material. The location provides
sharing, rating, and commenting options on
individual videos. The one that has uploaded the
video can change the comment settings. Only certain
users can comment and need that comments are
subject to moderation before they become publicly
visible or prefer to disable the comments facility
altogether. Users can "rate" comments using thumbs-
up and thumbs-down buttons, as they will for videos.
The comments that gather the foremost significant
number of positive votes are displayed above the
most bulk of comments. The comments facility was
intended to be "a section of text for users to supply
information associated with a video" and where
users could express their opinion on the video.
However, in practice, this is often not always what
happens. Technology provides affordances but is not
prescriptive; they have often put to uses aside from
that that it had been initially intended. This is often
the case with the comments facility: a user can write
things within the comments box that are not relevant
to the video [2].
2.3
Data Visualization
With the advent of technology, YouTube ought to
deal with numerous rows of data. To make a better
judgment, the data must be analyzed. When a large
amount of data is displayed to an audience dealing
with it, daunting and minor trends might always skip
the analyst's eye. Data visualization is a process
wherein the analyzed data is represented visually in
graphs and pictures, both attractive and functional.
Data is one among the foremost powerful and – often
overlooked – tools to speak a message. However,
those numbers alone would not necessarily make the
desired impact if the audience cannot interpret them.
That is why data visualization is the key [20].
Interactive charts and figures help understand the
varied outcomes possible in each situation. This type
of visualization helps people add an extra dimension
Journal of Theoretical and Applied Information Technology
30
th
November 2021. Vol.99. No 22
© 2021 Little Lion Scientific
ISSN:
1992-8645
www.jatit.org
E-ISSN:
1817-3195
5463
to their presentations, and cohesively display
essential statistics.
Tableau is a business intelligence and data
visualization tool that helps chart out the data onto
graphs, maps, etc. and break it down into several
components on a dashboard for real-time analysis.
With the help of Tableau, YouTube can explore and
analyses data visually. The program also offers the
option to create a dynamic presentation on the app
itself; post data visualization sheets have been
generated. By presenting the data in a visually
engaging way, one can enhance one's own and the
audience's experience. Tableau quickly creates
graphs and visualizations that include generic filters,
using graphs as filters; tooltips (hover over pop-ups);
and embedded webpages. A Tableau data extract
will take a snapshot of our data and put it into the
Tableau format that makes it easy for the software to
compute. This comes in handy when one has a large
data set. Data visualization is not just the process of
creating a few charts. It is about helping the audience
understand the significance of the data that is being
presented. Tableau makes the data easier to
comprehend,
making
statistical
relationships,
patterns, and critical trends more exposed and
recognized [21].
3.
METHODOLOGY
In this section, we will discuss the step-by-step
process,
which
helped
achieve
this
paper's
completion.
Figure 1: Research Flowchart[22]
a.
Identify The Problem
The initial stage of research is to identify
existing problems. The authors want to analyze
and visualize the Trending YouTube Videos
dataset, where the data are taken in the USA,
Great Britain, South Korea, and Japan.
b.
Determine The Method To Be Used
After the problem has been identified, the
authors then determine the method that will be
used. The method that will be used is data
visualization.
Data
visualization
visually
represents quantitative data with or without axes
in schematic or diagrammatic forms, e.g., Table,
Line chart, Pie chart, Histogram, Scatter plot, etc.
[23].
c.
Data Collection
After determining the appropriate method, the
authors then collected the data. The authors used
the Trending YouTube Videos dataset, where the
dataset is taken from Kaggle.com.
d.
Data Selection
Next is the data selection stage. At this stage,
the data will be selected. The input variables
include video titles, channel titles, published
times, tags, views, likes and dislikes, descriptions,
and the number of comments.
e.
Data processing
After all the necessary data have been selected,
the next stage is data processing. At this stage, the
data will be processed using Tableau, where the
data will be transformed or change the data
attribute values into the appropriate data so that
the data can be processed using data visualization.
The complete dataset will be obtained, which is
used to process to the next stage [24].
f.
Testing Data
At this stage, the data will be tested where data
testing is a stage that determines whether the test
made is suitable for use or not. If the test has
produced an output that suits the authors' needs,
the authors can proceed to the next stage.
However, if the test has not produced an output
that matches the needs, we have to return to the
data processing stage; if it has entered the repeat
or failure stage, it is necessary to solve it by
analyzing data processing [25].
g.
Conclusion
Journal of Theoretical and Applied Information Technology
30
th
November 2021. Vol.99. No 22
© 2021 Little Lion Scientific
ISSN:
1992-8645
www.jatit.org
E-ISSN:
1817-3195
5464
This stage is the last stage where the authors
draw conclusions that refer to the formulation of
the problem and research objectives based on the
test results. Suggestions are used to develop
further research and are included to improve the
quality of research.
4.
RESULT AND ANALYSIS
Before using the Tableau application, make
sure the data to be processed first. The data chosen
is data from YouTube Trending Videos in the
USA, France, Russia, India, and Japan from 2017
to 2018. When all data is valid, data processing
can be done using the Tableau application. The
first step is to connect the data. Open Tableau
Desktop, and after that, we will choose a select
connector, which is what data we will use.
Because the data taken from Kaggle is CSV, we
choose a text file; after that, we will be directed to
choose what data to use. After the data has been
selected, we will enter the Data Source. In the data
source, we can see the dataset that has been taken
from Kaggle because we are using multiple data
sources; we can see them all listed in the
data
source. After that, we can combine the data we
want to retrieve, namely the USA, France, Russia,
India, and Japan.
After combining the data, use a data interpreter
to help clean up the CSV data that has been
retrieved. After that, we can select some of the
data needed for data visualization. Data can be
classified into two parts, namely dimension or
measure. These data are data that have been
selected and processed to produce the data
visualization previously described. In Table 1, we
can see an explanation of the data that we used.
After the data has been selected, the data will be
processed.
Table 1: Type of Data
Category
A category is a group, a group of
various types of videos found on
YouTube. These categories consist
of Autos & Vehicles, Comedy,
Entertainment,
Education, Film & Animation,
Gaming, How to & Style, Movies,
Music, Nonprofits & Activism,
News & Politics, People & Blogs,
Pets
&
Animals,
Science
&
Technology,
Trailers,
Shows,
Sports, Travel & Events.
Region
The region is a part/territory of a
specific area such as the USA,
France, Russia, India, and Japan.
Trending Date
Trending Date is the date a video is
being considered attractive by
many YouTube viewers/users.
Total Videos
The total number of videos
uploaded to YouTube by users.
Figure 2: Trending History
Next, we will go to sheet 1, where we will
visualize the existing data. The dataset taken from
Kaggle starts from 14 November 2017 to 14 June
2018. After that, we will explain the results of the
data
visualization
research
that
we
have
examined. As can be seen in Figure 2, the chart is
an area chart. An area chart is a line chart in which
the area between the lines and axes is shaded
according to the part being represented. These
charts are typically wont to represent accumulated
totals over time and are the traditional thanks to
displaying stacked lines. The area chart may be a
combination between a line graph and a stacked
bar graph. By stacking the quantity beneath the
Journal of Theoretical and Applied Information Technology
30
th
November 2021. Vol.99. No 22
© 2021 Little Lion Scientific
ISSN:
1992-8645
www.jatit.org
E-ISSN:
1817-3195
5465
road, the chart shows the entire fields as their
relative size to every other.
Based on the results of the data visualization in
Figure 2, it can be seen that the Autos & Vehicles
category is the category with the highest number
of videos, and the Travel & Events category is the
category with the least number of videos. This
shows that YouTube users are more interested in
videos in the Autos & Vehicles category than
other categories.
Figure 3: Trending by country
Next in Figure 3 is a pie chart. The pie chart
represents data as slices of a circle with different
sizes and colors. The slices are labeled, and
therefore the numbers like each slice are
additionally represented within the chart. We will
select the chart option from the Marks card to
make a chart. It can be seen in Figure 3 that the
highest number of trending videos came from the
USA region, amounting to 80,758. A total percent
of 23.75%, followed by France with a total
number of trending videos of 75,404 and a total
percent of 22.18%, Russia with a total number of
trending video 73,694 and a total percent of
21.68% India with a total trending video of 73,372
and a total percent of 21.58% Japan with a total
number of video trending 36,762 and a total
percent of 10.81%.
Based on the horizontal bars in Figure 4, it can
be seen that total trending videos in 2017 came
from the USA with 18,918, followed by France
with 17,760 videos, India with 17,706 videos,
Russia with 17,276 videos. Furthermore, it can be
seen on the trending date for 2018 that the USA
still occupies the first position for the highest
number of trending videos with a total number of
videos of 61,840, followed by France with a total
of 57,644 videos, Russia with a total of 56,418
videos, India with a total of 55,666 videos, and
finally Japan with a total of 36,762 videos.
Figure 4: Trending date by country
Figure 5 shows the highlight tables that show
the total trending videos by category and country.
The highest number of trending videos uploaded
comes from India and is in the Entertainment
category, amounting to 32,924 videos. Based on
the category, it can be seen that the total videos
from the five regions that attracted the most users'
interest
were
the
Entertainment
category.
Whereas previously mentioned, India had 32,924
videos, followed by USA 19,638 videos, France
19,020 videos, Japan 11,734 videos, and Russia
11,692 videos. Furthermore, it can be seen that the
Trailers category, with a total of 4 trending videos,
is the category that attracts YouTube users the
least.
Figure 5: Trending by Country and Category
Next, we will analyze our data geographically.
With map visualization, it is easier for us to see
our data. To create map visualization, we first
need to change Region to Latitude and Longitude
geographic roles. After that, we will enter
Longitude in columns and Latitude in rows. To
make it easier to view the map, each region's
colors can be changed
so that each region's colors
differ. In Figure 6, we can see the mapping of the
Journal of Theoretical and Applied Information Technology
30
th
November 2021. Vol.99. No 22
© 2021 Little Lion Scientific
ISSN:
1992-8645
www.jatit.org
E-ISSN:
1817-3195
5466
USA, France, Russia, India, and Japan regions
with the total number of trending videos from
2017 to 2018.
Figure 6: Mapping by Country
Figure 7 shows a treemap chart. Treemap in
Tableau may be a primary chart type that's
represented by nested rectangular boxes. This
chart is often useful for giant
datasets for
visualization.
This
graph
will
mark
the
hierarchical data for comparative analysis and a
crucial chart to research the data set's anomalies.
In addition to the scatter plot, treemaps are only
visualization types that reasonably allow us to
communicate and consume many marks on one
view. This makes it easier to identify patterns and
relationships that we would not rather be ready to
see. Colors and gradients are wont to group items
while still identifying individual items. This chart
works rather well for
a massive amount of
knowledge. When the info increases, it also
increases the matter of understanding the data.
Treemap charts are effective practices once we
want to speak and consume a fair number of
marks on one view. With this significant behavior,
a user can easily spot patterns and relationships
between them, otherwise impossible.
Figure 7: Videos Count by Category and Country
As shown in figure 7, the largest square size is
India, with the Entertainment category in 2018
with 25,156 videos trending. Meanwhile, for 2017
with the same country and category, there were
7,768 videos. Next is the USA in 2018 with the
same category as before, with 15,010 trending
videos. It can be seen that from 2017 to 2018, the
Entertainment category attracted the most interest
from YouTube users, and in 2018, the second
most popular category was the people & blogs
category. In 2017, the second most popular
category was the comedy category, followed by
the people & blogs category.
A dashboard is a collection of sheets, which we
can compare with various data simultaneously.
Journal of Theoretical and Applied Information Technology
30
th
November 2021. Vol.99. No 22
© 2021 Little Lion Scientific
ISSN:
1992-8645
www.jatit.org
E-ISSN:
1817-3195
5467
After creating several data visualization sheets on
Tableau, we can combine the data visualizations
made on these sheets. As we can see in Figure 8,
we can see the full data visualization result
simultaneously.
CONCLUSION
In this paper, we have analyzed trending
YouTube videos data using data visualization.
As the "Age of Big Data" kicks into high gear,
visualization is an increasingly crucial tool to
make sense of the trillions of rows of data
generated every day. To better understand the
changes brought about by big data, this paper
focuses
on
data
analysis
using
data
visualization, where this data visualization we
are easier to understand through the visual
world, where pictures speak louder than words.
Data visualization is especially important for
big data and data analysis projects because it
allows trends and patterns to be more easily
seen. Good visuals tell a story, remove any
distractions from data, and highlight useful
information. Furthermore, since visualization
is so prolific, it is also one of the most useful
professional skills to develop.
With the rise of big data upon us, we need to
interpret increasingly larger batches of data.
Nevertheless, data visualization is not only
crucial for data scientists and data analysts; it is
necessary to understand data visualization in
any career. Whether we work in finance,
marketing, tech, design, or anything else, we
need to visualize data. That fact showcases the
importance of data visualization. Complex
algorithms are easier to understand in a data
visualization format than lines and text lines
and numbers. With so much information being
collected through data analysis in the business
world today, we must have a way to paint a
picture of that data to interpret it. If we are a
YouTuber, we will indirectly think about how
our videos can be trending and what type of
videos
trending
among
users.
For
any
YouTubers who question this, we can use data
visualization in answering these questions.
In analyzing YouTube trending videos, we use
tableau software to help visualize the data that has
been obtained. The following are some of the
charts used in this analysis, such as pie charts, area
charts, horizontal bars, highlight tables, treemaps,
and mapping. Based on the results of the analysis
that has been done, we can conclude that the
Entertainment category is the category that most
attracts YouTube users. This can be proven in
Figures 5 and 7, where from 2017 to 2018, the
entertainment category was the category with the
most trending videos compared to other
categories. Furthermore, it can also be seen that
the number of trending videos in 2017 was
significantly less than in 2018. In 2018 trending
videos increased dramatically, and with this
comparison, it can be concluded that YouTube
users in 2018 also increased.
In 2017 the categories that received much
attention from users were entertainment, people &
blogs, news & politics, comedy, music, and film
& animation. Simultaneously, the less attractive
categories to users are the categories of movies,
trailers, nonprofits & activism, shows, travels &
events. In 2018 the categories that received much
attention from users were entertainment, people &
blogs, music, news & politics, comedy, and
sports. As we can see in Figure 3 Trending by
Country, it can be concluded that the most
trending videos are from the USA, followed by
France and Russia.
With this analysis results, Data Visualization
can help YouTube and its users identify which
areas need to be improved, which factors affect
users' satisfaction and dissatisfaction, and what
to do and know to develop parts lacking and pay
attention to today's users' needs. Visualized
data can give YouTube and its users a better
prediction of current trending topics or videos
and future growth. It can be used to influence
our decision-making. It is especially important
when analyzing and implementing strategies to
improve YouTube; visualization makes it easy
to ensure any platform is as optimized as
possible. We need data visualization because
the human brain is not well equipped to devour
such a lot of raw, unorganized information and
switch
it
into
something
usable
and
understandable. We want graphs and charts to
speak data findings to spot patterns and trends
to realize insight and make better decisions
faster.
The problem lies in too much data that is
difficult to analyze. In this paper, we suggest
using data visualization in analyzing data to
make it easier to read and understand. By
applying such analytics to big data, valuable
information can be extracted and exploited to
improve
decision-making
and
support
informed decisions. Finally, if appropriately
implemented, any new technology can bring
with it some potential benefits and innovations,
Journal of Theoretical and Applied Information Technology
30
th
November 2021. Vol.99. No 22
© 2021 Little Lion Scientific
ISSN:
1992-8645
www.jatit.org
E-ISSN:
1817-3195
5468
let alone big data, which is an extraordinary
field with a bright future if approached
correctly. We believe that big data analytics is
essential in this era of data glut and can provide
unexpected insights and benefits for decision-
makers
in various
fields.
If adequately
exploited and applied, big data analytics can
provide a basis for progress at the scientific,
technological, and human levels.
REFERENCES
[1]
P. Gill, M. Arlitt, Z. Li, and A. Mahanti,
“YouTube traffic characterization: A view
from
the
edge,”
2007,
doi:
10.1145/1298306.1298310.
[2]
A. Madden, I. Ruthven, and D. McMenemy,
“A
classification
scheme
for
content
analyses of YouTube video comments,”
J.
Doc.
, 2013, doi: 10.1108/JD-06-2012-0078.
[3]
F. E. Gunawan, J. F. Andry, H. Tannady and R.
Meylovsky,
“Designing
Enterprise
Architecture Using TOGAF Framework in
Meteorological,
Climatological,
and
Geophysical Agency,”
Journal of Theoretical
and Applied Information Technology
, vol. 97,
no. 20, pp. 2376-2385. 2019.
[4]
E. W. T. Ngai, L. Xiu, and D. C. K. Chau,
“Application of data mining techniques in
customer
relationship
management:
A
literature review and classification,”
Expert
Systems
with
Applications
.
2009,
doi:10.1016/j.eswa.2008.02.021.
[5]
J. F. Andry, H. Tannady and F. E. Gunawan,
“Purchase Order Information System Using
Feature Driven Development Methodology,”
International Journal of Advanced Trends in
Computer Science and Engineering,
vol. 9, no.
2, pp. 1107-1112
.
[6]
D. Korte, “Business Intelligence in the
Hospitality Industry,”
Int. J. Innov. Manag.
Technol.
,
2013,
doi:
10.7763/ijimt.2013.v4.435.
[7]
L. C. S. Reddy and Dr. D. Murali,
“YouTube: big data analytics using Hadoop
and map reduce.,”
Int. J. Eng. Technol.
, vol.
7, pp. 12–15, 2018.\
[8]
H. Tannady, J. F. Andry, F. E. Gunawan and J.
Mayseleste, “Enterprise Architecture Artifacts
Enablers for IT Strategy and Business
Alignment
in
Forwarding
Services,”
International Journal of Advanced Trends in
Computer Science and Engineering,
vol. 9, no.
2, pp. 1465-1472
.
[9]
E. D. Madyatmadja, L. Liliana, J. F. Andry and
H. Tannady, “Risk Analysis of Human
Resource Information Systems Using Cobit 5,”
Journal
of
Theoretical
and
Applied
Information Technology
, vol. 98, no. 21, pp.
3357-3367
.
[10]
Nikhat Akhtar, Nazia Tabassum, Asif
Perwej, and Yusuf Perwej, “Data analytics
and visualization using Tableau utilitarian
for COVID-19 (Coronavirus),”
Glob. J. Eng.
Technol.
Adv.
,
2020,
doi:
10.30574/gjeta.2020.3.2.0029.
[11]
“Big Data Analytics: Transform terabytes
into
insights.”
https://www.tableau.com/learn/articles/big-
data-anal ytics#:~:text=Tableau is an end-
to,those insights across the organization.
[12]
J. R. Mashey,
Big Data and the Next Wave of
InfraStress
Problems,
Solutions,
Opportunities
. 1998.
[13]
A. Gandomi and M. Haider, “Beyond the
hype: Big data concepts, methods, and
analytics,”
Int.
J.
Inf.
Manage.
,
2015,doi:10.1016/j.ijinfomgt.2014.10.007.
[14]
J. F. Andry, H. Tannady and F. Nurprihatin,
“
Eliciting Requirements of Order Fulfillment
in A Company
,” The 2
nd
International
Conference on Engineering and Applied
Sciences
2019
(InCEAS
2019),
IOP
Conference Series: Materials Science and
Engineering
.
[15]
J. L. Monino and S. Sedkaoui,
Big Data,
Open Data and Data Development
. 2016.
[16]
S. V. K. Jagadish
et al.
, “Genetic Advances
in Adapting Rice to a Rapidly Changing
Climate,”
Journal of Agronomy and Crop
Science
.
2012,
doi:
10.1111/j.1439-
037X.2012.00525.x.
[17]
A. McAfee and E. Brynjolfsson, “Big data:
The management revolution,”
Harv. Bus.
Rev.
, 2012.
[18]
F. J. Ohlhorst,
Big Data Analytics: Turning
Big Data into Big Money
. 2012.
[19]
Resdiansyah, J. Darmawan, A. H. Wijaya, L.
Hakim and H. Tannady, “
Comparing Freeman
Chain Code 4 Adjacency Algorithm and LZMA
Algorithm in Binary Image Compression,”
Annual
Conference
on
Science
and
Technology Research (ACOSTER) 2020,
Journal of Physics: Conference Series
.
[20]
Hubspot & Visage, “How To Use Data
Visualization To Win Over Your Audience.”
[21]
S.
Ashish,
Varma
&
Sakshi,
“Data
Journal of Theoretical and Applied Information Technology
30
th
November 2021. Vol.99. No 22
© 2021 Little Lion Scientific
ISSN:
1992-8645
www.jatit.org
E-ISSN:
1817-3195
5469
Visualization: Using Tableau for Decision
Making,”
Res. Rev. Int. J. Multidiscip.
, vol.
4, no. 4, 2019.
[22]
P. Pathak, S. Gupta, and G. S. Dangayach,
“Sustainable Waste Management: A Case
From Indian Cement Industry,” Brazilian J.
Oper.
Prod.
Manag.,
2015,
doi:
10.14488/bjopm.2015.v12.n2.a7.
[23]
S. Sara Banu, D. Uma, M. Waseem
Ashfaque, andQ. S. Ali Ahmed, “A review
on Visualization Approaches of Data mining
in heavy spatial databases,”
IOSR J. Comput.
Eng. Ver. V
, 2015, doi:10.9790/0661-
171595105.
[24]
M. Khan and S. Shah Khan, “Data and
Information Visualization Methods, and
Interactive Mechanisms: A Survey,” 2011.
[25]
A. S. Nagdive and D. R. M. Tugnayat,
“Overview
on
Performance
Testing
Approach in Big Data,”
Int. J. Adv. Res.
Comput. Sci.
, 2014.