Generating Tailored Classification Schemas for German Patents

Oliver Pimas, Stefan Klampfl, Thomas Kohl, Roman Kern, Mark Kröll

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review

Abstract

Patents and patent applications are important parts of a company’s intellectual property. Thus, companies put a lot of effort in designing and maintaining an internal structure for organizing their own patent portfolios, but also in keeping track of competitor’s patent portfolios. Yet, official classification schemas offered by patent offices (i) are often too coarse and (ii) are not mappable, for instance, to a company’s functions, applications, or divisions. In this work, we present a first step towards generating tailored classification. To automate the generation process, we apply key term extraction and topic modelling algorithms to 2.131 publications of German patent applications. To infer categories, we apply topic modelling to the patent collection. We evaluate the mapping of the topics found via the Latent Dirichlet Allocation method to the classes present in the patent collection as assigned by the domain expert.
Original languageEnglish
Title of host publicationInternational Conference on Applications of Natural Language to Information Systems
Pages230 - 238
ISBN (Electronic)978-331941753-0
DOIs
Publication statusPublished - 2016
Event21st International Conference on Applications of Natural Language to Information Systems: NLDB 2016 - Salford, United Kingdom
Duration: 22 Jun 201624 Jun 2016

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9612
ISSN (Electronic)0302-9743

Conference

Conference21st International Conference on Applications of Natural Language to Information Systems
Country/TerritoryUnited Kingdom
CitySalford
Period22/06/1624/06/16

Fields of Expertise

  • Information, Communication & Computing

Fingerprint

Dive into the research topics of 'Generating Tailored Classification Schemas for German Patents'. Together they form a unique fingerprint.

Cite this