Database Reference
In-Depth Information
Fig. A.1
A screenshot of the main screen of TweetTracker. The figure shows the Tweets generated
from the New York region during Hurricane Sandy. Tables below the map summarize the frequently
mentioned users and resources in the Tweets
resources and people in the Tweets is also presented to the user. Boolean search
capability is provided using a specific index built on tokenized text in MongoDB.
Search is made flexible by the combination of text with other parameters such as
the language of the Tweet and specific geographic regions. Figure
A.1
shows an
illustrative screenshot of TweetTracker.
TweetXplorer
©[
5
] is a visual analytics platform to analyze events using Twitter
as the data source. A user can analyze the data along four dimensions: temporal,
geospatial, network, and text discussed in Chap.
5
. A screenshot of the components
of the system can be observed in Fig.
A.2
. User analysis of the Tweets is organized
along specific themes comprised of keywords and hashtags. The network component
is implemented in a fashion similar to the one described in Chap.
5
. A user can
zoom-in to a specific region to visualize the text in the form of a word cloud. The
time series presents a global comparison of the traffic for the defined themes. The
network component is also implemented in a similar fashion to the one described
in this topic. Each node is associated with a specific theme and internal node color
instead of size is used to indicate the number of retweets received (darker colored
nodes are retweeted more often). The network and map visualizations are also
connected to each other. A selection in one causes the other to be automatically
filtered to show the corresponding information.
A.2
More Examples of Visualization Systems
Building Twitter-based systems to solve real world problems is an active area of
research.
Twitris
[
6
] is an online platform for event analysis using Twitter. The
system combines geospatial visualization, user network visualization, and sentiment
