Information Technology Reference
In-Depth Information
Fig. 2.1 Screenshot of web interface
on New York in Google maps as it can be seen from the Fig. 2.1 . Furthermore, the
statistical indicators can be viewed by the system as well.
It is demonstrated that the Naïve Bayes classification andMaximumEntropy clas-
sification methods can be effectively used to perform sentiment analysis on tweets.
To the best of our knowledge, this is the first geographic location-based sentiment
tracking system for Twitter, which allows one to monitor the brands in line with the
views of people in different cities.
Section 2.2 starts with the concepts of blogging and microblogging. Afterward,
Twitter, Sentiment Analysis, Natural Language Processing, Maximum Entropy,
and Naive Bayes classification methods are briefly described. Section 2.3 provides
an informative introduction to the technologies used while developing the web-
based project. Python and Natural Language Tool Kit (NLTK) are introduced, fol-
lowed by Twitter API, Google Maps API, and hand-classified dataset. Section 2.3
also describes how the tracking sentiment for Twitter project was implemented.
Section 2.4 covers the evaluation of the system. Section 2.5 , finally, summarizes the
chapter and gives a brief outlook for future studies.
2.2 Background
2.2.1 Blogging and Micro-Blogging
Blogging can be described as a platform where people can share their hobbies and
personal experiences on the World Wide Web. It has become one of the social
Search WWH ::




Custom Search