Aim of this project is to classify tweets for the COVID19 pandemic using Machine Learning and Natural Language Processing.
In a time of crisis, social media such as Twitter tends to be people’s go-to place for sharing information quickly. This generates a significant amount of data, some of which are extremely valuable to be used for relief work.
These messages disperse among multiple categories like available facilities, asking for help, information about affected, recovered, or dead people, etc. Extracting important situational updates from this huge data is a difficult yet important task.
Our project automatically classifies these tweets into four main categories as Treatment, Awareness, Reports, and Irrelevant. The classified data will be highly helpful for health organizations and NGOs that are seeking to provide help.
The manually annotated dataset is available in the TweetsDataset.csv file. And the code of this project is available as a Jupyter Notebook as TweetSense.ipynb
This project is licensed under the MIT LICENSE.