The classic search in patent literature using keywords and patent classes is increasingly supplemented by novel search techniques, which apply semantic search concepts or artificial intelligence. Hypothetically, a combination of semantics and artificial intelligence will result in a better identification of potentially relevant patent documents. This is investigated on independent test cases. Potentially relevant State-of-the-Art documents are identified through a defined process consisting in the generation of a patent portfolio by means of a semantic search followed by a subsequent analysis of this portfolio using artificial intelligence. Results indicate a potential of the sequential search approach.
Patent Monitor from Averbis was used to test this approach using their semantic search technology.
Additional information on this tool can be found at: https://www.ml4patents.com/vendors/patent-monitor
More information on Averbis can be found at: https://averbis.com/