The term \u201einformed ML models\u201c originates from the fields of Artificial Intelligence, Big Data and Smart Data, as well as Industry and Factory 4.0. These are Machine Learning (ML) models that not only learn from data but also utilise additional knowledge or expert information. Such knowledge can, for example, come from manuals, expert statements, or existing research findings.<\/p>\n
This additional information helps the ML model to make better and more accurate predictions because it doesn't start with a \u201eblank slate\u201c during analysis. Instead of just learning from vast amounts of data, relevant knowledge is specifically incorporated.<\/p>\n
A vivid example: In a factory, a machine is to use machine learning to predict when a defect might occur. Instead of just relying on the collected machine data, the model is also fed the expertise of technicians. They know, for instance, that certain noises often indicate bearing damage. With this expert knowledge, the ML model detects potential problems earlier and more accurately.<\/p>\n
Informed ML models are particularly useful when a lot of data is available, but additional specialist knowledge makes the decisive difference. They combine the best of human and machine.<\/p>","protected":false},"excerpt":{"rendered":"