NLP improves emergency message routing in disaster simulations

This research paper explores the use of Natural Language Processing (NLP) to prioritize emergency messages during disasters, using the 2015 Nepal earthquake as a case study. The study employed TF-IDF with Logistic Regression and BERT to categorize social media messages into nine humanitarian categories, which were then mapped to four routing priorities. While BERT demonstrated higher accuracy in assigning priorities and better recall for critical P1 messages, these improvements did not translate into significant overall network performance gains in terms of delivery probability or latency. The benefits of semantic prioritization became most apparent under conditions of limited transmission and storage resources, where it helped reduce message loss in buffers. AI

IMPACT This research demonstrates how NLP can be applied to improve emergency communication systems, though its effectiveness is constrained by network limitations.

RANK_REASON Academic paper detailing a simulation study. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• April 2015 Nepal earthquake
• Bert
• Binary Spray and Wait
• Kathmandu
• logistic regression model
• Milena Radenkovic
• natural language processing
• Nepal
• ONE simulator
• Papyrus 1
• Persona 4
• tf–idf

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →