Find. Extract. Predict.

Search for: "patents"

Konstantin Kruse ∙ 2020 March 09

Probably most of the scientific information is captured in tables – for example in US patents from 2001-2017 we have extracted more than 10 million tables containing interesting properties on materials and compou... read more

Lutz Weber ∙ 2020 January 15

A lot of scientific information is captured in images – we are using machine learning techniques such as deep neural networks to classify images.  For example, we have applied transfer learning to train a deep co... read more

Felix Berthelmann ∙ 2013 January 23

A first publication describing the classification of chemical patents using OntoChem’s OCMiner® has appeared in the new journal Pharmaceutical Patent Analyst, 2013, 2, 2046, “Ontology-based content analysis of US patent applications... read more