JRC1995/BERT-Disaster-Classification-Capsule-Routing
Exploration of BERT-BiLSTM models with Layer Aggregation (attention-based and capsule-routing-based) and Hidden-State Aggregation (attention-based and capsule-routing-based).
This project helps disaster response organizations, emergency managers, and crisis monitoring teams quickly identify and categorize tweets during a disaster. By inputting raw social media posts, it classifies them as disaster-related or not, and further assigns detailed categories like 'infrastructure damage' or 'requests for help'. This allows responders to efficiently triage information and prioritize urgent needs.
No commits in the last 6 months.
Use this if you need to automatically sort and understand large volumes of social media posts to extract critical information during a crisis or emergency.
Not ideal if you require real-time, production-ready tweet classification with minimal setup, as this project focuses more on model exploration and requires some technical setup.
Stars
25
Forks
4
Language
Python
License
MIT
Category
Last pushed
Mar 24, 2020
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/JRC1995/BERT-Disaster-Classification-Capsule-Routing"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
prrao87/tweet-stance-prediction
Applying NLP transfer learning techniques to predict Tweet stance toward a topic
brunneis/ilab-erisk-2020
Repository accompanying the CLEF 2020 eRisk Workshop Working Notes for the iLab team (University...
blazejdolicki/bert-sarcasm-detection
Sarcasm detection with BERT
SuryaVamsi-P/Conflict-NLP-Topic-Modeling-Sentiment-Analysis-using-LLMs
Extracts insights from 26K+ protest events using BERTopic, Top2Vec, and LLMs for real-world...
tk-yasuno/disaster-question-answer
Japanese disaster-focused question answering system : Utilizing the bert-base-japanese-v3 +...