Google may soon tell you which restaurants could give you food poisoning.
The tech giant is working with Harvard University to develop an algorithm that analyzes Google searches to spot which restaurants might have food safety issues.
Researchers say it’s capable of flagging possible offenders in ‘near real time.’
Scroll down for video
Google is working with Harvard University to develop a machine learning-based algorithm that analyzes Google searches to spot which restaurants might have food safety issues
They created a machine-learning based algorithm to identify unsafe restaurants, training it to look for specific search terms and location data.
The model is called FINDER, or Foodborne Illness Detector in Real Time.
First, it classifies searches that contain certain terms, like ‘stomach cramps’ or ‘diarrhea.’
It then uses anonymized and aggregated location history data from smartphones of people who have opted to save it.
The algorithm uses this information to determine which restaurants people searching those terms had recently visited.
To test the algorithm’s effectiveness, researchers gave actual health inspectors a list of restaurants identified as having food safety issues, as well as those that had consumer complaints.
FINDER was more effective than customer complaints, which were discovered to be accurate only 38% of the time. This is likely because about 62% of consumers believe the last restaurant they visited in the culprit of their foodborne illness, when medical research shows otherwise
Health inspectors weren’t told which ones had been identified by the algorithm and which had received consumer complaints.
Researchers tested the system in Chicago and Las Vegas between 2016 and 2017 and received positive results.
They found the overall rate across both cities of unsafe restaurants detected by the model was 52.3 percent.
By comparison, the overall rate of unsafe restaurants detected by routine inspections was 22.7 percent.
‘[We] demonstrated that FINDER improves the accuracy of health inspections; restaurants identified by FINDER are 3.1 times as likely to be deemed unsafe during the inspection as restaurants identified by existing methods,’ according to the study.
To test the algorithm, researchers gave actual health inspectors a list of restaurants identified as having food safety issues, as well as those that had consumer complaints
FINDER was also more effective than customer complaints, which were discovered to be accurate only 38 percent of the time.
Researchers believe this is because most people assume the cause of their food poisoning was the last place they ate at, causing them to file a complaint at the wrong restaurant.
They point to medical studies that have shown foodborne illnesses can take 48 hours or even longer to become symptomatic after someone has been exposed.
‘The new model outperformed complaint-based inspections and routine inspections in terms of precision, scale, and latency (the time that passed between people becoming sick and the outbreak being identified),’ Harvard’s T.H. Chan School of Public Health said in a statement.
GOOGLE’S ALGORITHM CAN PREDICT WHEN PATIENTS WILL PASS AWAY
Google has created an AI that it claims is 95 per cent accurate in predicting whether hospital patients will pass away 24 hours after admission.
This is around 10 per cent better than traditional models.
To make its predictions, the software uses data such as patient’s ethnicity, age, gender, previous diagnoses, lab results and vital signs.
But what makes it so powerful is that it includes data previously thought out of reach of machines, such as doctor notes buried in PDFs or scribbled on old charts.
As well as death, AI can also predict unplanned re-admissions within 30 days and probable length of stay at a hospital.
Google has created an AI that it claims is 95 per cent accurate in predicting whether hospital patients will pass away 24 hours after admission
The system is still in its infancy, but Google believes it could someday be used to predict death far longer in advance.
To test the system, Google obtained de-identified data of 216,221 adults, with more than 46 billion data points between them.
After studying the data, the AI was able to identify which words were associated closest with outcomes.
While the results have not been validated, Google claims huge improvements over traditional models.
The biggest benefit, researchers claim, is the ability for the system to use all types of data.
Researchers believe the FINDER algorithm could be used in accordance with existing methods used by health departments to spot restaurants with foodborne illnesses.
As a result, it may allow them to ‘better prioritize inspections and perform internal food safety evaluations,’ Harvard explained.
‘In this study, we have just scratched the surface of what is possible in the realm of machine-learned epidemiology,’ Evgeniy Gabrilovich, senior staff research scientist at Google and a co-author of the study, said in a statement.
‘We can use online data to make epidemiological observations in near real-time, with the potential for significantly improving public health in a timely and cost-efficient manner.’