Problem and Objective

During the gasoline shortage crisis in Florida in the onset of Irma, the following kinds of tweets were observed:

"The shelters are full, there is no gas. Tornados could happen, and storm surge is predicted. So what are people supposed to do? Irma "
"Insane..95 percent of Florida trying to leave at one time. Roads r slammed. No gas. No hotels available. Scared to see my neighborhood after irma"
"Gas stations out of gas, water shelves empty, stores and airports closed. Stocked up on food and wine, waiting on irma"

Other issues were also reported. We explore different approaches for situation awareness and social sensing using social media posts during disatser

Methodology

We tested the following approaches:

Traing different Classifiers - LSTMS, RNNs etc to detect specific issues like gasoline shortage.
Finetuned open source LLMS like LLama for the above task.
Developed a RAG approach for general situation awareness and ability to detect and deep dive into specific issues

Name		Name	Last commit message	Last commit date
Latest commit History 23 Commits
data		data
.DS_Store		.DS_Store
RAG_on_disaster_tweets.ipynb		RAG_on_disaster_tweets.ipynb
README.md		README.md
Testing_DL_architectures.ipynb		Testing_DL_architectures.ipynb
testing_LLMS.ipynb		testing_LLMS.ipynb
testing_LLMS_tweets.ipynb		testing_LLMS_tweets.ipynb