CHAPTER certain requirements (e.g., comparing the publicCHAPTER certain requirements (e.g., comparing the public



We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

Social media provides many opportunities for organizations
such as businesses and political parties to spread their opinion. Twitter is a
leading and interacting micro-blogging network in social media networks. Users
of twitter generate a lot of information every day by the tweets and bursts
that are sent out and because twitter is used worldwide, there is a very large
market potential if the data generated are put to profitable use..

We are living in an era where a large amount of
digital information is generated on a basis through social networks, blogs, Web
logs, online communities, news sources, and mobile applications. This massive
amount of data holds valuable information that can be used for various
analytical purposes, e.g., to track social opinion on different topics, trends
of a certain event, interest of different social groups etc. There are many
ongoing efforts on using social media such as Twitter to analyze the public
political opinions on candidates in elections at different stages in various
countries. These results can not only help the campaign gain more insights
about the election, but also help political candidates decide how to allocate
campaign resources to maximize their winning chances. This type of analysis
requires a significant amount of effort, and tends to be performed in an ad-hoc
manner for certain requirements (e.g., comparing the public opinions on Hillary
Clinton and Donald Trump). Large organizations and Corporations are seemingly
employing more and more twitter analysts to run their Twitter accounts, with a
focus on the target of people they want to advertise to, this is shown in the
case of U.S Fastfood Giant, Wendy’s, where the twitter analysts running their
accounts have used twitter as a major advertising tool to young people, by tweeting
with popular and trending hashtags and pop-culture, using slangs that are
popular with young people and overall giving an image to the social media
account of a hip and up to date young person instead of a large franchise. This
has been shown to work for them where some of their tweets have gone viral and
this has brought in a lot of new customers. Increasing social media has led to
micro-blogging becoming the most importance tool for information feeding.

Twitter is one of the social media networks and micro-blogging service that
allows users to interact each other and sharing information such as sending,
reading and receiving messages these messages called tweets. In this year
Twitter users reached 271 million users and 500 million tweet sent per day, as
a result of these massive data generated on daily, twitter has become the main
focus on most of researchers involved data mining.

The twitter users mostly use this service as a
way of sharing of ideas and thoughts, posting news and discussing real time topics,
express their opinion get information from twitter, as election is big issue,
that users share their opinion on presidential candidate, and this become more
interest and popular to predict presidential election. Twitter is one of the
most popular social media in today’s modern world. It consists of information
about real events happening around the world.













The aim of our project is to present a tool for
visualizing Twitter data, and then use such tool to look at how tweets can
assist a business manager in BI by offering several different kinds of
visualizations that can pertain to a Twitter user or any keyword or hashtag
entered through the interface. In this project we present web tool that
visualizing twitter data that offers different set of kinds of visualizing
tools that can pertain to a twitter user or any hashtag or keyword entered in
the interface. We also look at a practical way of visualizing tweets and how
such visualization would affect or effect how business was done in an



















Some articles or websites that have
talked about the visualisation of tweets for business intelligence and how
using data from tweets to generate useful additives for businesses are shown,
were we explored the power of the programming language R for data mining. In
using R to visualize tweets as a word cloud to find out what people
are tweeting about the NBA using the hash tag (#nba){1}. A word cloud is a
visual representation showing the most relevant words (i.e., the more times a
word appears in our tweet sampling the bigger the word). The final result
should look similar to the following:


Load the Twitter authentication and
extract tweets using #NBA.


We have already been authenticated
and successfully retrieved the text from the tweets using #NBA. The first step
in creating a word cloud is to clean up the text by using lowercase and
removing punctuation, usernames, links, etc.


In the next step we will use the
text mining package tm to remove stop words. A stop word is a commonly used
word such as “the”. Stop words should not be included in the analysis.


Now we’ll generate the word cloud
using the wordcloud package. For this example we are concerned with plotting no
more than 150 words that occur more than once with random colour, order, and
position. This post highlights how easily R can extract and visualize Twitter
data as a word cloud. There are thousands of ways to represent data in R and
you’ll need to dig deeper to fully understand how all the words are related to
NBA. “Lakers” might be obvious, but “Paul” or “girlfriend” might require more contexts.

{2} TweetViz offers several
different interactive visualizations that can provide insight into user
interests and activity as well as information about certain keywords and Hashtags.

TweetViz can be divided into two separate modules. The first is user-centric
and focuses on analysing user behaviour from different aspects. The second
module, on the other hand, is more search term orientated, where a user can
explore Twitter activity surrounding a specific hashtag or keyword. Moreover,
TweetViz incorporates the LDA algorithm for visual representations of topic distribution,
from tweets, both from a specified user or tweets containing a search term. One
advantage of this tool is that it builds visualizations on up-to-date data,
unlike some approaches that use static, previously retrieved data.  In TweetViz we use few third party libraries.

Google Charts and d3 (Data Driven Documents) are used to build the different
types of visualizations. For generating topic distributions with LDA we used
the Python gensim 1 framework.

{3}Tweets are visualized in
different ways in each of the tabs at the top of the window.


 Each tweet is shown as a circle positioned by
sentiment, an estimate of the emotion contained in the tweet’s text. Unpleasant
tweets are drawn as blue circles on the left, and pleasant tweets as green circles
on the right. Sedate tweets are drawn as darker circles on the bottom, and
active tweets as brighter circles on the top. Hover your mouse over a tweet or
click on it to see its text.


Tweets about a common topic are
grouped into topic clusters. Keywords above a cluster indicate its topic.

Tweets that do not belong to a topic are visualized as singletons on the right.

Hover your mouse over a tweet or click on it to see its text.


Pleasure and arousal are used to
divide sentiment into a 8×8 grid. The number of tweets that lie within each
grid cell are counted and used to colour the cell: red for more tweets than
average, and blue for fewer tweets than average. White cells contain no tweets.

Hover your mouse over a cell to see its tweet count.


Common words from the emotional
regions Upset, Happy, Relaxed, and Unhappy are shown. Words that are more
frequent are larger. Hover the mouse over a word to see how often it occurred.


Tweets are drawn in a bar chart to
show the number of tweets posted at different times. Pleasant tweets are shown
in green on the top of the chart, and unpleasant tweets are shown in blue on
the bottom. Hover the mouse over a bar to see how many tweets were posted at
the given time.

Map. Tweets are drawn on a map of
the world at the location where they were posted. Please note most Twitter
users do not provide their location, so only a few tweets will be shown on the
map. Hover your mouse over a tweet or click on it to see its text.


Frequent tweets, people, hashtags,
and URLs are drawn in a graph to show important actors in the tweet set, and
any relationship or affinity they have to one another. Hover your mouse over a
node, or click on a node to see its tweets.


Selecting an anchor tweet of
interest from the tweet list displays a time-ordered sequence of tweets that
form conversations or narrative threads passing through the anchor tweet. Hover
your mouse over a node or click on it to see its text. Hover your mouse over a
link to see all threads that pass through the link, or click on it to see the
tweets in each thread.

Tweets. Tweets are listed to show
their date, author, pleasure, arousal, and text. You can click on a column’s
header to sort by that column.


To zoom in on the tweets in the
Sentiment and Topic tabs, click the zoom icon to the right of the Query button.

This displays a zoom lens that you can move around the visualization.


You can query multiple keywords at
once, and combine keywords in different ways. For example, to search for tweets
with the words “cat” and “dog”, enter: cat dog. To search
for tweets with the words “cat” or “dog”, enter: cat OR
dog. To search for tweets with the phrase “cat dog”, enter: “cat
dog”. To search for tweets with the words “cat” but not
“dog”, enter: cat -dog.

Do You Estimate Sentiment?

We use a sentiment dictionary to
estimate sentiment. We search each tweet for words in the dictionary, and then
combine the words’ pleasure and arousal ratings to estimate sentiment for the entire
tweet. When you hover your mouse over a tweet’s circle to see its text, the
words in our dictionary are shown in bold italics. You can click on a tweet’s
circle to bring up a dialog that gives even more information.

is the leading social media for people to get and express the information
related policies proposed by their favorite candidates in elections. The growth
of twitter in political campaigns has made it popular subject in research area.

In the winning of 2008 in American election with Barrack Obama, the role of
twitter in politics become clear; twitter as a platform of political thought
attracted the attention of many researchers and politicians. Robertsona et al.,
(2010) studied the use of social media such as twitter for candidates and
voters for the United States presidential election in 2008. He examined the
purpose of candidates’ use of social media such as twitter and how this tools
can be used for better communication, and how public engage and share
information. The researcher analyzed users’ pattern of communication of
political opinions. In terms of visualization, they mainly focused on
statistical analysis using one bubble chart and fourteen bar charts which are
simple and easy to recognize, common and traditional analytical visualization.

Since it used many bar charts, it is somewhat difficult to see the overall
trend and takes time to interpret.

the visualization mainly focus on statistical analysis of bar chart and bubble
chart which are easy and sample to understand, in common visualization some of
them is difficult to see and interpret.

studies examined how to predict election result and monitored political
sentiments using twitter data so they used to compare the predict and the
result in bar chart, and sentiments and series of time visualization they used
line chart.


of the studies related election campaigns used for twitter expressed on
statistical analysis visualization such as bar chart and line chart and took
time to obtain and process data to get final result. Since start opinion
related political campaigns and presidential elections differ real time due to
debates of candidates and the real time analysis is more effective. To gather
easily in information by region, visualizing public opinion on maps would be















4.1 Data

Twitter enables third party applications and
developers to get access to the enormous amount of data generated by users
every day. This is done using the Twitter REST API, which offers a lot of
different endpoints for retrieving this data. What caught our interest was the
endpoint that allows us to retrieve tweets from a single user. We also leverage
the search capabilities offered by the Twitter Search API. This is used to get
tweets which contain a given search term, both keyword and hashtag. Although it
offers extensive functionality, the Twitter API has certain limitations. First
of all is the rate limit window which limits the number of requests that can be
sent. Another deficiency is that the search service provided by Twitter does
not index all tweets and as a result, not all tweets are available for
retrieval from Twitter Search. Nevertheless, it can provide sufficient number
of tweets for the purposes of our web tool.

4.2 User-orientated

In making a visualization that can be easily be
understood by any user, we have to look at the understandability of the
visualization we are trying to use, the visualizations intended for use in this
project is wordcloud and heatmap, where we use wordcloud to see the area of
preference and how much about the business is tweeted, retweeted, liked and
discussed about, and the heatmap is used to show the concentration of the users
of twitter and the areas where the users tweeting about it the most are
concentrated, this helps the business in deciding which areas to advertise to
and also areas to add incentives so that the business operations can improve



Case Study

To show as an example a practical way in which
using visualization of tweets for business intelligence, Insurance Companies in
the United States of America, are looking for where to invest and open up new
branches and get new customers, with a focus on Natural Disaster insurance to
properties and lives, so tweets from all 50 states in the U.S with hashtags or
keywords like natural disaster, evacuation, tornado, bushfires, heavy wind,
blizzard, lightning, landslides, destruction, hurricane, sinkholes, wildfires,
avalanche, famine and volcano are collected, stop words like severe, the, and
other prefixes are removed, also the hashtag sign is viewed as a stop word and
also removed, this is used to generate a heatmap of the locations where the
majority of the tweets are from, or where the twitter users tweeted or
retweeted things pertaining to it.

Figure 3: Heatmap showing concentration of users
tweeting about Natural Disasters.

The data generated from the tweets collected are
then used to create a wordcloud that shows the frequency of the type of natural
disaster and how much it appears in the data collected, this helps the
insurance company know what the range for premiums should be, what their major
focus should be on and how much to invest in the different departments they
wish to make in the new branches.

4: WordCloud showing the frequency of the Natural Disasters.







Data Visualization

If we are to use wordcloud and heatmap to see
where the majority of the users are situated, it is shown on the location
services of twitter users called “tweeps” which they can voluntarily choose to
show or not, but when the users agree to show their location or to make it
public, we can use that to ascertain the concentration of people, the hubs,
which could help businesses and business managers know where to focus. Because
now we can show where products do well and where it doesn’t do so well based on
people tweeting about their use of the products, retweeting things about it,
liking and also based on views, because the modern versions of twitter shows
the number of interactions of a tweet with the
option of tweet activity, which shows impressions- times people on Twitter saw
the tweets, and total engagement which is broken down into likes, Details
expand- times people viewed the details about the tweet, and profile clicks-
number of clicks on your name, @handle, or profile photo.

Organizations advertise and promote their business, they rely on viral videos
or tweets or finely choreographed videos and it does work. The modern
internet-savvy user is tired of a television advertisement. People today prefer
advertising that is faster, less intrusive, and can be turned on or off at
will… and Twitter is exactly that. If you learn how the nuances of tweeting
work, you can get good advertising results by using Twitter.

 There are different ways Twitter can be used
by businesses. First, some businesses generate revenue solely through advertisements.

These organisations can Tweet about the content they provide or the activities
they’re involved in to drive more people and traffic to their website,
ultimately generating more revenue for them. To build subscribers, the company
could use hashtags related to its content to find its audience members. Other
companies like business-to-business or business-to-consumer companies can
spread its content or product information through Twitter in the same way.

Content-based business like publishers who have a lot of written content on
their websites uses Twitter for search engine optimization (SEO) purposes.



















project’s major aim was to present a tool for visualizing Twitter data,
identify how twitter can assist a business or organization or even a political
campaign by offering several different kinds of visualizations that can pertain
to a Twitter user’s use of a keyword or hashtag entered through the interface
and then analyzed and used to create a visualization that can assist businesses
and business managers in making smart decisions and in helping the organization
gain better efficiency and better productivity.
























Intelligence: Concepts, Methodologies, Tools, and Applications 1st Edition by
Information Resources Management Association.

{6} Robertsona, S.P, Vatrapub, R.K and Medinaa, R. (2010) “Off the wall
political discourse: Facebook use in the 2008 U.S. presidential election”
Information Polity 15 (2010) 11–31


{8} Kaplan, A.M and Haenlein, M. (2010) “Users of the world, unite! The
challenges and opportunities of Social Media.” Business Horizons (2010) 53,