Date of Award
2021
First Advisor
Amanda Landi
Second Advisor
Michael Bergman
Abstract
Health departments have been deploying text classification systems for the early detection of foodborne illness complaints in social media documents such as Yelp restaurant reviews. For example, such system will classify each review as "Sick" or "Not Sick", meaning whether the writer indicates in the review that they experienced food poisoning from eating at the restaurant or not. Then by looking at the "Sick" reviews, the offcials can determine if the restaurant needs further investigation. Current systems have been successfully applied for documents in English and, as a result, a promising direction is to increase coverage and recall by considering documents in additional languages, such as Spanish or Chinese. This thesis explores one way of expanding the current English system to a multilingual one by only using labeled English reviews. The first chapter provides an introduction to the problem and outlines important contributions by the author. The second chapter summarizes previous works and explains key concepts related to the topics. Next, the third chapter describes the model structure and configurations for the proposed method. The fourth chapter presents experimental setup and results and highlights important findings. The final chapter ends with conclusion, discussion, and further work.
Recommended Citation
Liu, Ziyi, "Detecting Foodborne Illness Complaints in Multiple Languages Using English Annotations Only" (2021). Senior Theses. 1510.
https://digitalcommons.bard.edu/sr-theses/1510
Simon's Rock students and employees can log in from off-campus by clicking on the Off-campus Download button and entering their Simon's Rock username and password.