Service: Machine Learning in Text Classification Projects - Relieving staff from coding tasks

Responsible organisation: Federal Statistical Office Germany (Central-Government)

The Federal Statistical Office Germany currently works in several projects to employ machine learning to text classification problems. In many statistics, text information is collected. For statistical purposes, the texts need to be coded into a set of predefined categories. This is a time consuming task. The main goal of the text classification projects is to relief staff from coding tasks and to improve timeliness of statistical results at the same time. Often these approaches are imperative, because digitization leads to larger data sets that need to be processed in a timely manner.

Additional information

Source Open Innovation Regione Lombardia
Web site https://www.statistischebibliothek.de/mir/servlets/MCRFileNodeServlet/DEMonografie_derivate_00001972/PoC_MachineLearning.pdf
Start/end date 2017 - 2018.0
Still active?

Related cases