Database Reference
In-Depth Information
Appendix A
Additional Information
In this appendix, we provide more information on building practical applications
using the techniques discussed in the chapters of this topic. In Sect.
A.1
, we discuss
two systems built utilizing many of the techniques discussed in this topic. In
Sect.
A.2
, we discuss various academic and commercial systems built using Twitter
data. In Sect.
A.3
, we provide more information on the libraries used to construct
the examples described in this topic and provide links to other resources.
A.1
A System's Perspective
Throughout this topic, we have discussed techniques to build the necessary com-
ponents for a system which collects information from Twitter and facilitates and
analysis and visualization of the data. For those interested in examples of systems,
which can be built using the techniques discussed in this topic, we will present two
systems, which demonstrate how the techniques can be used to build a solution to a
real world problem.
TweetTracker
©[
1
], is a platform to collect and analyze Tweets in near real-time.
The objective of the system is to facilitate near real-time Tweet aggregation and to
support search and analysis of the collected Tweets. In TweetTracker Tweets are
organized into events to facilitate the study of related Tweets. Each event can be
described as a collection of hashtags/keywords, geographic boundary boxes, and
Twitter user IDs. Users create and edit events, which TweetTracker then uses to
collect Tweets using the Streaming API. Tweets are indexed and stored in MongoDB
using the techniques discussed in Chap.
3
.
Analysis of the data is supported by various visual analytics. Temporal informa-
tion of keywords and hashtags can be compared using the technique described in
Chap.
5
. Geospatial visualization is performed using the “dots on map” approach
discussed in Chap.
5
. Geographic location of the Tweets is obtained by using the
contents of the Twitter profile location field in the absence of geotagging. Tweet
text is summarized using a word cloud and a summary of the frequently mentioned
