Nathan Jacobi
Computer Science Department, University of Georgia

Ivan Mo
Computer Science Department, University of Georgia
Linguistics Department, University of Georgia

Albert You
Computer Science Department, University of Georgia

Krishi Kishore
Computer Science Department, University of Georgia

Zane Page
Computer Science Department, University of Georgia

Shannon P. Quinn
Computer Science Department, University of Georgia
Cellular Biology Department, University of Georgia

Tim Heckman
Public Health Department, University of Georgia

Abstract

Human languages' semantics and structure constantly change over time through mediums such as culturally significant events. By viewing the semantic changes of words during notable events, contexts of existing and novel words can be predicted for similar, current events. By studying the initial outbreak of a disease and the associated semantic shifts of select words, we hope to be able to spot social media trends to prevent future outbreaks faster than traditional methods. To explore this idea, we generate a temporal word embedding model that allows us to study word semantics evolving over time. Using these temporal word embeddings, we use machine learning models to predict words associated with the disease outbreak.

Keywords

Natural Language Processing, Word Embeddings, Bioinformatics, Social Media, Disease Prediction

DOI

10.25080/majora-212e5952-01a

Bibtex entry

Full text PDF

Copyright © 2022 Nathan Jacobi et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.