AI investigates electricity theft and bill errors

• Autores: Chris Baraniuk
• Localización: New scientist, ISSN 0262-4079, Nº. 3145, 2017, pág. 16
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • An algorithm tested on several million Brazilian households could help cut electricity theft and ensure bills are accurate. Patrick Glauner at the University of Luxembourg and his team looked at records of the energy used by 3.5 million Brazilian households over five years. Based on this, they developed an algorithm that could recognize when energy use at a property was suspiciously low. Because the researchers also had data on past inspections against which to check the AI's conclusions, they could verify that their tool was correctly highlighting potential cases of fraud or faulty metering.