Researchers generate new technique for detecting environmental offences

Researchers in the US say environmental regulators across the world can benefit from a computer-based technique to double the the number of offences detected.

A study from Stanford University has found that machine learning - training computers to automatically detect patterns in data - can be “a powerful and cheap new weapon” for cash-strapped regulators.

The study, published in Nature Sustainability, looked at the US Clean Water Act, and using data from past inspections, the researchers deployed a series of models to predict the likelihood of failing an inspection, based on facility characteristics, such as location, industry and inspection history. They then ran their models on all facilities, including ones that had yet to be inspected.

Please sign in or register to continue.

Sign in to continue reading

Having trouble signing in?

Contact Customer Support at
or call 020 8267 8120

Subscribe for full access

or Register for limited access

Already subscribe but don't have a password?
Activate your web account here