Using text mining and other big data analytics to analyze public discussion of novel food technologies, using cultured meat as an example

Understanding public opinions is crucial for the food industry and related regulatory bodies. Scholars, companies, and regulatory bodies have explored public opinion on food-related issues, including food products/technologies and food safety outbreaks, through various methods such as surveys and calls for comments. However, these methods can be limited by high costs and low response volumes, potentially leading to inaccurate data and hindering informed decision-making. Conversely, the vast amounts of online text data provide a rich, yet underutilized, source of public opinion that could greatly enhance decision-making in the food industry and among regulatory bodies. To address this gap, this work aims to leverage conventional big data analytics and develop new text mining tools to provide stakeholders with rapid and reliable insights into the evolving public discussion on food-related issues, empowering them to take immediate actions and make more informed decisions. This study uses cultured meat as a case study to gain comprehensive insights into the longitudinal public discussions of novel food technologies. It will first analyze the usage frequency, information dissemination, and topics associated with different names for cultured meat in Twitter discussions through volume analysis, network analysis, and word cloud analysis. Subsequently, the study will examine public attitudes toward cultured meat by conducting sentiment analysis on cultured meat-related tweets, leveraging a novel generative large language model (LLM)-based method. This study contributes by (1) providing a reference for the use of various analytical methods on online food-related data, (2) offering insights to cultured meat industry to support more informed decision-making, and (3) proposing a novel generative LLM-based text analysis method, which has the potential to overcome the time and cost limitations of traditional machine learning models.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Food Science & Human Nutrition

Grantor

University of Illinois Urbana-Champaign

Year dc:date

2025

Author dc:creator

• Chen, Tianli

Contributors dc:contributor

• Su, Leona Yi-Fan
• Schmidt, Shelly J
• Stasiewicz, Matthew Jon
• Ng, Margaret Yee Man
• Wang, Yi-Cheng

• Nomenclature
• Public perceptions
• Twitter
• Generative large language models

Statement dc:rights

• Copyright 2025 Tianli Chen

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/129722

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F129722