Information travels rapidly through social media. In times of crisis, it can be difficult to keep informed about the people affected by a disaster. Crisis Map uses tweet data to reach beyond the limits of news coverage to show the impact and response of people near the crisis.
What it does
Crisis Map raises public awareness of recent areas of crisis and how others are reacting to the event by visualizing disaster data. In addition, Crisis Map has the potential to provide a tool for emergency responders. Within the disaster’s region, Crisis Map may be extended to show them where the most people have a negative reaction, signaling places that still need aid.
How we built it
Crisis map pulls disaster data from Doina Caragea’s research with Kansas State University’s Machine Learning and Data Science Lab. This data, provided to us by ACM-W, is comprised of Twitter data from the time span of the event, such as a hurricane. Crisis Map then plots tweet locations using the Google Maps API to show what geographic areas people are responding from. Utilizing artificial intelligence, Crisis Map visualizes the twitter data through most common hashtags, the sentiment of the actual event, and frequently mentioned words from the tweets. Common hashtags are located using Python and sentiment analysis is calculated using the Google Cloud Natural Language API. The sentiment analysis determines how positive or negative people feel regarding an event. Frequently mentioned words are found using Compromise and individual words can be searched for their number of instances and sentiment analysis. Crisis Map is run using Node.js.
Challenges we ran into
Our team had some challenge adjusting to working with a variety of API’s. Initially, it was difficult determining how to sort through the large amount of disaster data that we were provided.
Accomplishments that we're proud of
We are proud to have developed a working product based on the researcher’s data.
What we learned
Our team gained more knowledge about text analysis along with more experience with developing a web application.
What's next for Crisis Map
The next stage for Crisis Map would focus on real time data streaming along with more data and flexibility integrated into the map. Also, Crisis Map may include twitter integration to connect the data with a user’s followers.
Plus, we are looking to add in a donations feature, where people can view relevant organizations and relief efforts based on the disaster. This way, Crisis Map can help charities and non-profits expand their impact when responding to crisis.