Novelty detection is particularly at home in the fields of artificial intelligence, big data and smart data as well as Industry and Factory 4.0. This involves technical systems - for example machines, computer programmes or sensors - being able to automatically recognise when something unusual or new happens.
Imagine a production plant that manufactures thousands of car parts. Using novelty detection, a system with sensors constantly checks the parts and compares them with previous data. If a part suddenly appears that looks or has been processed differently than usual, the system sounds the alarm. This is particularly important in order to recognise errors at an early stage and avoid major damage or costs.
Novelty detection is also used in the area of big data. It helps companies to discover unusual patterns in huge amounts of data - for example in payment transactions to expose attempted fraud.
In short, novelty detection means that digital systems can automatically recognise deviations and new, unprecedented events. This helps companies to react in good time, manage better and therefore work more efficiently and safely.
Deutsch 
English (UK)