smacawi.github.io

Public site for Social Media and Crowdsourcing Assessment of Weather Impacts at https://smacawi.github.io/.

Links to the project's key outputs:

Reports:

• Final project report
• Annotated bibliography
• Social network analysis report

Publications:

• Romascanu, A., Ker, H., Sieber, R., Greenidge, S., Lumley, S., Bush, D., Morgan, S., Zhao, R., & Brunila, M. (2020). Using deep learning and social network analysis to understand and manage extreme flooding. https://doi.org/10.1111/1468-5973.12311
• Mircea, A. (2020). Real-time Classification, Geolocation and Interactive Visualization of COVID-19 Information Shared on Social Media to Better Understand Global Developments. https://doi.org/10.18653/v1/2020.nlpcovid19-2.37
• Brunila, M., Zhao, R., Mircea, A., Lumley, S., & Sieber, R. (2021). Bridging the gap between supervised classification and unsupervised topic modelling for social-media assisted crisis management. https://www.aclweb.org/anthology/2021.adaptnlp-1.5/

Presentations:

• Sieber, R., 2020, Communicating Winter Storms via Natural Language Processing of Social Media, CMOS Congress 54. https://www.iclr.org/wp-content/uploads/2020/07/Renee_Sieber.pdf
• Sieber, R., 2020, Crowdsourcing Winter Storm Sentiment via Natural Language Processing, World Meteorological Organisation HIWeather Workshop Successful citizen science, http://hiweather.net/article/18/132.html

News coverage:

• McGill news releases: https://www.mcgill.ca/newsroom/channels/news/using-artificial-intelligence-manage-extreme-weather-events-327770
• Synced AI and Technology Review: https://syncedreview.com/2021/01/20/mcgill-researchers-use-ai-to-manage-extreme-weather-events/

Codebase:

• Twitter scraper: https://github.com/smacawi/twitter-scraper
• Tweet classifier: https://github.com/smacawi/tweet-classifier
• Topic modeling: https://github.com/smacawi/topic-modeler
• BERT hybrid models: https://github.com/smacawi/bert-topics
• Dashboard APIs: https://github.com/smacawi/dashboard_standalone_apis
• Dashboard front end: https://github.com/smacawi/dashboard_frontend

Demos:

• Dashboard with dummy data: https://dash-flask-ec.herokuapp.com/.