A group of computer science professors and graduate students from the University of Maryland and University of California, Los Angeles have created an automated national database of food safety inspection information considered the largest of its kind in the U.S.
The InspectionRepo database uses data robots to automatically collect data from local government websites, according to UMD professor of computer science Ben Bederson.
The database represents a significant leap forward from local and state databases, which generally rely on manually collected data that can vary in the way information is correlated, coded and reported.
The new database allows food service businesses and consumers to monitor and compare food safety practices from outlets across the nation.
The national database was developed by Bederson, UMD professor of economics Ginger Jin, UCLA associate professor of business management Phillip Leslie, Ph.D. graduate Alexander Quinn and UMD Ph.D. graduate student Ben Zou, according to a report in Library Sciences.
“Our data robots cover a large number of local jurisdictions across the U.S., continuously detecting new data posted by each jurisdiction, and integrating them into a single, standardized, and cumulative database,” Bederson said. The new approach makes the database more robust, scalable and cost effective than manual databases.
A central challenge to the creating the national database was developing algorithms that normalize data that differ among jurisdictions. That meant writing custom data scraping scripts for some websites and interpreting data on other sites.
The research team also developed analytical tools that compare inspection outcomes by localities and by types of food outlets, which can help improve inspection efficiency, promote compliance, and reduce food-borne illnesses, according to Bederson.