ITECH1103- Big Data and Analytics

This report provides insights and analysis of a dataset on YouTube videos, including the number of uploaded videos, categories, countries, and more. It also includes advanced insights and recommendations for the ABC online multimedia company.

Running head: ITECH1103- BIG DATA AND ANALYTICS
ITECH1103- Big Data and Analytics
Name of the Student
Name of the University
Authors note

1ITECH1103- BIG DATA AND ANALYTICS
Background information
For the ABC online multimedia company, this project is aimed to find out the
different insights from the dataset which is collected form the following URL;
https://data.world/iamdilan/youtube-dataset
The dataset contains total 161,470 rows, 17 columns. The rows contains data about the
videos different attribute on the YouTube such as video_id, trending date, title of the video
channel title that uploaded the video , a specific category_id of the video, publish date of the
video, time_frame for the uploaded video, publication day week for the video, publish
country for a specific video , tags used for the video, views, likes, dislikes, count of the
counts for videos, whether the comments or ratings are disabled for the video and finally
information about whether there us any video error or removed from the site. From the
analysis it is found that, in the dataset, it contains the detailed data about the viewership of
the audience in the countries USA, UK (used country code GB), Canada and France.
The following report contributes to the interpretation of the different dashboards that are
created using the IBM Watson Analytics tool, five advanced insights, justification about the
use of the Watson analytics tool and Recommendation for the ABC online multimedia
company.
Reporting / Dashboards
Answer 1
The total number of uploaded videos is 161470.

2ITECH1103- BIG DATA AND ANALYTICS
Answer 2
The dataset contains total 18 categories of videos uploaded in the dataset for the
given time period which are;
1: Film & Animation
2: Autos & Vehicles
10: Music
15 - Pets & Animals
17 – Sports
22 - People & Blogs
23 – Comedy
20 – Gaming
24 - Entertainment
43 - Shows
44 - Trailers
25 - News & Politics
19 - Travel & Events
26 – How to & Style
27 - Education
28 - Science & Technology
29 – Non-profits & Activism
30 – Movies

3ITECH1103- BIG DATA AND ANALYTICS
Answer 3
There are total 4 countries in the dataset.
Answer 4
There are total 12360 unique channels in the dataset.

4ITECH1103- BIG DATA AND ANALYTICS
Answer 5
The top three countries are France, Canada and US according the distinct number of
channels.
Answer 6
The country with the least number of channels on YouTube from the dataset is GB.

5ITECH1103- BIG DATA AND ANALYTICS
Answer 7
There are total 2207 unique YouTube channels in US.
Answer 8
Following is the list dashboard that depicts the top 10 viewed videos for the
Countries US, GB, CANADA and France.

6ITECH1103- BIG DATA AND ANALYTICS
From the above dash board, it can be stated that, there are similar 10 videos that
ranked between the top 10 videos, the list of videos is;
Nicky Jam x J. Balvin - X (EQUIS) | Video Oficial
Bad Bunny - Amorfoda | Video Oficial
Te Bote Remix - Casper, Nio GarcÃa, Darell, Nicky Jam, Bad Bunny, Ozuna | Video Oficial
Ozuna x Romeo Santos - El Farsante Remix
Childish Gambino - This Is America (Official Video)
Drake - God's Plan (Official Music Video)
YouTube Rewind: The Shape of 2017 | #YouTubeRewind
Ozuna - UÌnica (Video Oficial) 🻠A U R A
Ariana Grande - No Tears Left To Cry (Live On The Tonight Show Starring Jimmy Fallon)
Becky G, Natti Natasha - Sin Pijama (Official Video)
Answer 9

7ITECH1103- BIG DATA AND ANALYTICS
following is the list of bottom 10 videos for the countries;
the list contains the following
Famous Dimbler,JO2018, Moring live and others.
Answer 10
The data set contains data about the uploaded video for 13 years.

8ITECH1103- BIG DATA AND ANALYTICS
Answer 11
For the last month December, the number of the uploaded videos is given by 8525.
Answer 12
For the country GB, the maximum amount of video is uploaded in the year 2018.

9ITECH1103- BIG DATA AND ANALYTICS
Answer 13
The time frame in which the maximum number of videos are uploaded is 16:00 to
16:59.
Answer 14
The top 3 viewed categories in terms of number of uploaded videos is given by 10, 1, 29.
Answer 15
Least three video categories in terms of uploaded videos are 44,25,27.

10ITECH1103- BIG DATA AND ANALYTICS
Answer 16
Videos that has highest percentage of likes is given by Childish Gambino - This Is America
(Official Video)
Answer 17
Videos that has highest percentage of dislikes is given for the video title “So Sorry.”

11ITECH1103- BIG DATA AND ANALYTICS
Answer 18
Friday has the maximum number of uploaded videos.
Answer 19
The least number of videos are uploaded on Saturday.

12ITECH1103- BIG DATA AND ANALYTICS
Answer 20
Advanced Insights
Advanced insight 1
Top 10 tags that generated the most disliked videos on you tube is given by;

13ITECH1103- BIG DATA AND ANALYTICS
“Rewind|
Rewind 2017
youtube rewind 2017
#YouTubeRewind
logan paul vlog
logan paul
logan
paul
Olympics
logan paul youtube
Eurovision Song Contest|"2018"|
Advanced insight 2

14ITECH1103- BIG DATA AND ANALYTICS
Country with the audience with most disliked videos
Advanced insight 3
The following dashboard shows the factors that drives the likes for the videos. The
most influential one is comment count for a specific video and the category of the video.
Advanced insight 4

15ITECH1103- BIG DATA AND ANALYTICS
Maximum comments in a month is recorded in the month of May, 2018 in USA and
the Supporting dashboard is provided below;
Advanced insight 5
Trend in the upload of the videos in different countries;

16ITECH1103- BIG DATA AND ANALYTICS
From the above dash board it can be easily stated that, there is huge increase in the
number of uploaded videos in GB from the year 2016 compared to the other three countries
whose records are available in the dataset.
Research
Use of the IBM Watson Analytics is helpful in identification of different associations
and patterns in the related data in order to provide automatically created starting points for
the analyst. As the analyst change the data parameters in the visualization process IBM
Watson changes the starting points in order to reflect changes along with the Parameters.
From the analysis it can be stated that, the number of uploaded videos has
significantly increased in the year 2018 for the four countries compared to the last 12 years
(from the year 2006 to 2017).
From the advanced insights, the tags that drives dislikes for the uploaded videos are
the following;
Rewind|"Rewind 2017""|""youtube rewind 2017""|""#YouTubeRewind""|
logan paul vlog|"logan paul""|""logan""|""paul""|""olympics""|""logan paul youtube""
Eurovision Song Contest|"2018"|
Moreover, from the given dataset the it is also evident that, the maximum number of the
video uploads are in the months of December from different countries. From the Country
GB, lowest number of you tube channels which may lead to the lesser number of views from
that country compared to the other countries.

17ITECH1103- BIG DATA AND ANALYTICS
Recommendations for Content Manager
In order to improve the business, it is important to increase the number of channels as
well as viewership in the GB.
The users and the viewers are not interested in the YouTube rewind videos as well as
other videos that includes the tags related to the YouTube Rewind. Thus it is suggested to
avoid the rewind videos. Moreover, in the advanced insights it is found that, most of the
dislikes are from the country GB. The second highest number of dislike are from the US.
As it is found that, the entertainment is the most liked category and the comments on
this category of the videos. So that, the viewership of the videos can be enhanced as well as
the business of the organization through the increased number of viewers. As there is slow
growth in the upload of the videos, it is important encourage users to upload videos in USA,
CANADA, FRANCE.
Cover letter
To
The Content Manager
ABC online Multimedia Company
Respected Sir/Madam
I am writing this letter in order to convey the results of the analysis
of the dataset provided. From the analysis, it is evident that, the viewership as well as
number of channels is lowest in the country GB. Furthermore, for the most viewed video
category is 24 which is Entertainment related videos. Therefore, it is suggested to promote
the Entertainment related videos and thus will help in attracting increased viewer ship from
the different countries. The videos tagged with “logan paul vlog|"logan paul” or

18ITECH1103- BIG DATA AND ANALYTICS
“Rewind|"Rewind 2017""|""youtube rewind 2017” are mostly disliked by the viewers. The
rate of upload of videos has been increased from the October, 2017 and continuously
increased for the December month of the year 2018.
Therefore with the help of the above mentioned insights and recommendations it will
be helpful for the organization in order to improve its business by acquiring increased
number of viewers from the countries.
Sincerely,
[Please Fill your name]
Reflection
While using the Watson analytics tools for the analysis of the tool we found that,
with the use of the starting points it becomes easier to find out the insights.
With all the above mentioned advantages there are some flaws in this AI based NLP
tool. Watson Analytics tool is not capable of combining the multiple formats of data
altogether. More than one excel data files cannot be merged and used for the analytics even if
the data files contains a common column along with the same column name. Along with that
the Watson analytics tool provides limited number of mathematic operations on the data set
imported to the platform. Compared to the other database and data analytics tools we need to
store moderately higher amount of redundant data on cloud platform of the IBM Watson. In
addition to that, IBM Watson Analytics uses different types of statistical algorithm to
uploaded and selected data in order to discover numerous insights from the different starting
points using different patterns as well as correlations in data.

19ITECH1103- BIG DATA AND ANALYTICS
Bibliography
Akhigbe, O., Heap, S., Amyot, D., & Richards, G. S. (2017). Exploiting IBM Watson
Analytics to Visualize and Analyze Data from Goal-Based Conceptual Models.
In Proc. 36th International Conference on Conceptual Modelling 2017 (pp. 352-355).
Anshari, M., & Lim, S. A. (2017). E-Government with Big Data Enabled through
Smartphone for Public Services: Possibilities and Challenges. International Journal
of Public Administration, 40(13), 1143-1158.
Bakkar, N., Kovalik, T., Lorenzini, I., Spangler, S., Lacoste, A., Sponaugle, K., ... & Bowser,
R. (2018). Artificial intelligence in neurodegenerative disease research: use of IBM
Watson to identify additional RNA-binding proteins altered in amyotrophic lateral
sclerosis. Acta neuropathologica, 135(2), 227-247.
Chen, Y., Argentinis, J. E., & Weber, G. (2016). IBM Watson: how cognitive computing can
be applied to big data challenges in life sciences research. Clinical
therapeutics, 38(4), 688-701.
Forrest, E., & Hoanca, B. (2015). Artificial Intelligence: Marketing's Game Changer.
In Trends and innovations in marketing information systems (pp. 45-64). IGI Global.
Gandomi, A., & Haider, M. (2015). Beyond the hype: Big data concepts, methods, and
analytics. International Journal of Information Management, 35(2), 137-144.
Herleman, H., & Ryan, J. (2015). A Big Data Platform for Workforce Analytics. In Big Data
at Work (pp. 33-56). Routledge.
Hoyt, R. E., Snider, D., Thompson, C., & Mantravadi, S. (2016). IBM Watson analytics:
automating visualization, descriptive, and predictive statistics. JMIR public health
and surveillance, 2(2).

20ITECH1103- BIG DATA AND ANALYTICS
Neto, L., Batista, O., Albuquerque, J., Souza, W. V., Cesse, E., & Cruz, O. G. (2017).
Disruptive innovations and transformations in public health in the digital
age. Cadernos de saude publica, 33(11).
Plevoets, C., & Fernandes, R. (2016). Exploring the role of Big Data and Analytics: Creating
Data-Driven Innovation.
Sammour, T., Cohen, L., Karunatillake, A. I., Lewis, M., Lawrence, M. J., Hunter, A., ... &
Thomas, M. L. (2017). Validation of an online risk calculator for the prediction of
anastomotic leak after colon cancer surgery and preliminary exploration of artificial
intelligence-based analytics. Techniques in coloproctology, 21(11), 869-877.
Saxena, A., Sharma, N., Saxena, K., & Parikh, S. M. (2017, August). Financial Data Mining:
Appropriate Selection of Tools, Techniques and Algorithms. In International
Conference on Smart Trends for Information Technology and Computer
Communications (pp. 244-251). Springer, Singapore.
Senthilkumar, S. A., Rai, B. K., Meshram, A. A., Gunasekaran, A., &
Chandrakumarmangalam, S. (2018). Big Data in Healthcare Management: A Review
of Literature. American Journal of Theoretical and Applied Business, 4(2), 57-69.
Steude, D. H. (2017). Change and innovation leadership in an industrial digital
environment. Management of Organizations: Systematic Research, 78(1), 95-107.
Zhang, X., Pérez-Stable, E. J., Bourne, P. E., Peprah, E., Duru, O. K., Breen, N., ... & Denny,
J. (2017). Big data science: opportunities and challenges to address minority health
and health disparities in the 21st Century. Ethnicity & disease, 27(2), 95.