infoapps GmbH

For external and internal patent searches as well as classifying jobs, infoPatent deploys semantic processing algorithms. They base on so-called "invention clouds", i.e. term vectors containing a description of the essence of a petent. These vectors are generated by identifying in the English, German, French, and Spanish full-texts or abstracts those parts, which most clearly describe the invention from a technical point-of-view. Each patent is described by one or more invention clouds.

For the similarity search or a classifying task, invention clouds are also built for the textual search input, and these input invention clouds are then compared to all pre-existing ones (about 200 million) from the stored patents.

Thus the use of potential synonyms for the search text has become obsolete, and the quality of the identified hits is dramatically increased.