Sentiment and Semantics

The Sentiment and Semantics online app (Android version to follow), allows you to view tweets containing the search phrase you enter when presented with a popup after pressing one of the update buttons. The tweets you view are classified according to their sentiment and can be analyzed with respect to their semantic content. Duplicates have been removed along with possible spam.


During the initial classification phase, the sentiment of each tweet is estimated by looking for the presence of words with known sentiment within each tweet. A score is then generated which allows a tweet to be classified as either vpos, pos, neutral, neg and vneg. Classified tweets are then passed through a learning algorithm which estimates the sentiment of all words occurring in each tweet independently of the words used during the initial classification phase. The POS, NEG and NEUT tabs contain the classified tweets. The LANG tab contains the words classified by the learning algorithm.


Many tweets contain references to articles which can be analyzed further. By clicking on a link, the corresponding article will be opened in one tab along with a semantic analysis in another. The semantic analysis is also contained within the SUM tab.

The semantic analysis contains the following information at present:

  • Topics
  • Entities
  • Keywords
  • Tags
  • Categories

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s