The term automated feature engineering is particularly at home in the fields of artificial intelligence, big data, smart data and digital transformation. It plays a central role when companies utilise large amounts of data to teach machines to learn - for example for forecasts, pattern recognition or decision-making.
Feature engineering means extracting important characteristics (features) from raw data so that algorithms deliver the best possible results. Until now, data experts have done this laboriously by hand. Automated feature engineering utilises special software that takes over this process. This saves time and reduces human error. This means that efficient work can be done even with little knowledge of data.
An example: an online shop wants to predict which products will be particularly popular. Automated feature engineering takes all existing customer data, searches for meaningful characteristics such as purchase frequency or seasonal trends and prepares them for artificial intelligence. This enables the system to recognise trends much more quickly and accurately - without the need for time-consuming manual sifting through the data.
Automated feature engineering makes data projects simpler, faster and more successful - and enables companies to better utilise the potential of modern technologies.
Deutsch 
English (UK)