During disasters, social media like Twitter is an important platform for people to post content to report updates about injured or dead people, infrastructure damage, and missing or found people among other information types. The information contained in Tweets helps humanitarian organization save lives and reduce damage. Therefore, we aim to classify the humanitarian information based on tweets with textual messages and images. We apply finetuned BERT model for text classification, ResNet-50 model for image classification, and decision level functions for multimodal classification. The classes of humanitarian information of tweets with texts and images are predicted well by our model.
-
Notifications
You must be signed in to change notification settings - Fork 1
Given dataset consists of tweets with textual messages and images, the task is to determine what kind of humanitarian information the tweets convey to help alert humanitarian organizations what kind of aid is required to save lives.
qianxi5/Social-Media-Data-Classification-for-Disaster-Respnse
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
Given dataset consists of tweets with textual messages and images, the task is to determine what kind of humanitarian information the tweets convey to help alert humanitarian organizations what kind of aid is required to save lives.
Topics
Resources
Stars
Watchers
Forks
Packages 0
No packages published