Study: Text Mining Approaches for Postmarket Food Safety Surveillance Using Online Media
Researchers from San Diego State University, Virginia Tech, Loyola Marymount University, and Radford University are using text mining to try to pinpoint unique words and phrases in posts online, which identify consumers' experiences with hazardous food products.
The scientists say that they hope this will "provide a practical and inexpensive means for rapidly monitoring food safety in real time." They compiled a large data set of labeled consumer posts spanning two major websites, and were able to utilize a data set of over four million online reviews.
According to the study's abstract, the researchers say their results were 77 to 90 percent accurate in top-ranking reviews, while sentiment analysis was 11 to 26 percent accurate. The researchers suggest the use of these tools to "profile food items and assess risk, building a postmarket decision support system to identify hazardous food products."
Patrick Quade, founder of iwaspoisoned.com, provided access to food safety reports that made the study possible. His website has been able to help identify many high-profile foodborne illness outbreaks in the past few years.
View the full study here.
Looking for quick answers on food safety topics?
Try Ask FSM, our new smart AI search tool.
Ask FSM →
