forked from Lefteris008/EvS
-
Notifications
You must be signed in to change notification settings - Fork 0
01. Introduction
Lefteris Paraskevas edited this page Apr 24, 2016
·
1 revision
This tool attempts to combine a given Event Detection technique with the sentiment information from the tweets of the dataset.
In particular, it combines the sentiment information of the tweets that belong to a certain event and exports them for further research.
The tool is comprised by seven separate modules; these are the following:
- Pre-Processing module, for the process of tokenization, stemming and any language-oriented procedure on the text of the tweets.
- Event Detection module, in which one or more ED algorithms are implemented and executed (one at a time).
- Sentiment Analysis module, for the SA part.
- Experimenter module, in which the main experiments of every technique are executed and the generated results are exported in separate files.
- Evaluator module, for the process of evaluating the generated results of the aforementioned techniques, by comparing them with predefined ground-truth data.
- EvS module, a module which binds the previous five in unique classes and methods and performs Event Detection via Sentiment Analysis.
- Utilities module, for secondary tasks like retrieving the tweets from the database etc.