Local Chapter Sydney, Australia Chapter
Coordinated byAustralia ,
Status: Completed
Project Duration: 24 Oct 2022 - 28 Nov 2022
Open Source resources available from this project
Project background.
Australia has gone through several Disaster events in recent years , The floods in July 2022 Australia are some of the worst the country has ever experienced and have caused widespread devastation.. These are the same communities where we saw massive bushfires in 2019 and 2020 which resulted in the loss of shrubs and trees setting the scene for extreme flooding. Tens of thousands of Australians have had to evacuate their homes after devastating floods struck the eastern part of the country, resulting in several millions of dollars in damage
Experts say climate change is fuelling an increase in extreme weather across Australia, threatening to make bushfires, floods and droughts more common.
A report published last month by the United Nations Environment Programme (UNEP) and GRID Arendal predicts that wildfires will become more frequent and intense, with a global increase of extreme fires by 50 per cent by the end of the century. The increase in wildfires renders land barren, which leads to increased run-off and, therefore, floods and, later, drought.
AI and Machine Learning can play crucial roles in the forecast and monitor and manage these disaster events. While several technologies and mechanisms is already in place , one area that can significantly supplement and improve the efficiency of these managing these disaster events is leveraging Social media
The problem.
At a high level, the goal of this project is to extract useful information regarding the disaster detection events that arise from social media feeds and refine it further from additional information. Based on this information we can identify and tag the areas that need attention on the disaster extent map.
The project results will be made open source. The aim being to help connect and encourage organisations to use AI tools to understand and plan on how to manage these disaster events. We also hope to encourage citizen science by open sourcing the dataset and code.
Project goals.
Project plan.
Week 1
Collection of Dataset (Tweets, Images, Public Data sets related to Disaster events)
Week 2
Collection of Dataset (Tweets, Images, Public Data sets related to Disaster events)
Week 3
Data Pre processing
Week 4
Data Preprocesisng
Week 5
Model Building
Week 6
Model Validation & Fine tuning
Week 7
Reporting and final Packaging
Week 8
Wrapup
Learning outcomes.
NLP Processing and Segmentation, Twitter Text Extraction, Classification Models, Reporting
