ITECH1103 Big Data Analytics: YouTube Video Analysis Report 2018

This report analyzes a YouTube dataset from 2006 to 2018 using IBM Watson Analytics to identify trends and insights. The analysis explores various video attributes such as category, publish date, views, likes, and dislikes to provide recommendations for content managers. Key findings include identifying popular video categories, trends in video uploads across different countries, and factors contributing to video dislikes. The report concludes with suggestions for improving video quality and user engagement to maximize viewership and minimize negative feedback. Visualizations and graphs generated from IBM Watson are included to support the analysis. Desklib provides access to this and similar assignments for students.

Name of the Student:
Name of the university:

The requirement to use IBM Watson
The tool is a Business Intelligence tool which is used for
processing
An example is included here
Where was X born?
The birth place fo Einstien is included by the visualization.
Structured
Unstructu
red

A brief history of Watson
 Started in 2007,
 lead David Ferrucci
Initial goal: Natural
language processing
Knowledge Extraction
Need: Gathering
knowledge that would not
be possible by the usual
system.

Basic Architecture IBMBasic Architecture IBM
WatsonWatson

INTRODUCTION TO
CURRENT PROJECT
The selected dataset Youtube.xlsx contains the
information about different videos uploaded in the
period of 2006 to 2018. The data dictionary has the
following columns Video_id, Trending_date, Title,
Channel_title, Category_id, Publish_date, Time_frame,
Publish_day_of_week, Publish_country, Tags, Views,
Likes, Dislikes, Comments_count, Comments_disable,
Ratings_disabled, Video_error_or_removed. The
description of each columns are given in the question
file.

ADVANCED INSIGHTS
The advanced insights have been developed after the
dashboards have been developed.
The main goal for the insights are that they would be
able to provide the organizations with useful in
information so that they would be able to use it for
future endeavours.
The YouTube dataset has been used here for the
development of the insights on the topic and the
insights on dislikes have been developed here.

INSIGHT 1

INSIGHT 2

INSIGHT 3

INSIGHT 4

INSIGHT 5

RESEARCH
From the analysis it is found that, the
dataset contains data only about 18
categories which are listed below;
 1-Film & Animation
 10- Music
 29 – Non-profits & Activism
 20 – Gaming
 22 - People & Blogs
 23 – Comedy
 43 - Shows
 44 - Trailers
 24 - Entertainment
 22 - People & Blogs
 23 – Comedy
From the analysis it is found that, the
dataset contains data only about 18
categories which are listed below;
 43 - Shows
 44 - Trailers
 15 - Pets & Animals
 17 – Sports
 25 - News & Politics
 19 - Travel & Events
 26 – How to & Style
 2 Autos & Vehicles
 27 - Education
 28 - Science & Technology
 30 – Movies

RECOMMENDATIONS
The content manager would be advised to take the following
recommendations into considerations:
It has been suggested to the Content Manager to enhance the
quality of the videos in t he You tube. There have been
various music videos uploaded in the min t h and gaining the
most viewed category in the YouTube. The quality of the
music video might help in increasing the viewers in the
YouTube.
There has been slow growth in the video upload as compared
to GB than US, France, and Canada. Therefore, users need to
be encouraged to upload their videos over the Youtube.
Dislikes for the videos should be minimized as it reduces the
interest of the other users therefore, it is suggested to check
and remove the contents that can be disliked by the
audience.

CONCLUSION
For conclusion it can be said that data analytics is very
important for any type of business and the report
consists the details of the Youtude dataset. The
analysis has tried to gather information about the
current trends in the YouTube videos and the uploads
which are made in YouTube for the entertainment of
the audience. Visualizations an graphs from datasets
have been provided from IBM Watson for the
reference of the readers.

BIBLIOGRAPHY
Silverman, B. W. (2018). Density estimation for statistics
and data analysis. Routledge.
Agresti, A. (2018). An introduction to categorical data
analysis. Wiley.
Wickham, H. (2016). ggplot2: elegant graphics for data
analysis. Springer.
Ott, R. L., & Longnecker, M. T. (2015). An introduction to
statistical methods and data analysis. Nelson Education.
Schabenberger, O., & Gotway, C. A. (2017). Statistical
methods for spatial data analysis. CRC press.
Yan, C. G., Wang, X. D., Zuo, X. N., & Zang, Y. F. (2016).
DPABI: data processing & analysis for (resting-state) brain
imaging. Neuroinformatics, 14(3), 339-351.