Challenge Question
How can we discern a real tweet with important information from a prank or fake one in the aftermath of a disaster?
Project Source: Kaggle, via Lassonde
Project Summary
With the ubiquitous nature of smartphones enabling people to announce an event they are observing in real-time, Twitter has become an important communication channel in times of emergency. Because of this utility, agencies such as disaster relief organizations and news agencies are interested in monitoring Twitter via an automated program. However, it is not always clear whether a person’s words are actually announcing a disaster. Using a dataset of 10,000 hand-classified tweets, this project challenges a research team to build a machine learning model that predicts which tweets are about real disasters and which ones are not. Interested students may have backgrounds in data science, communications, computer science and technology, digital communication, the social sciences, and disaster and emergency management. This challenge originated from the Kaggle website, and team members will be working with a York University mentor rather than a mentor from the organization.
Want to learn more? Click here!Sustainable Development Goals
Organizational Profile
Kaggle is an online data science and machine learning community that also serves as a repository of code and data. Users can find and publish high-quality data sets, write code and build models, ask questions and collaborate, and take data science and machine learning courses through the website. Kaggle also runs competitions, where users have the opportunity to apply their knowledge to real-world machine learning problems and solve a number of complex global challenges.
Partner Website
Learn more about the kind of work the project partner does by browsing their website.
Additional Resources
Key Words
- Emergency Response
- Information Verification
- Machine Learning
- Data Science