Presentation at the Summer School on Data and Algorithms for Science, Technology & Innovation Studies 2022

Standard-essential patents are the fundament of technical standards and technologies like 5G. Their declaration as standard-essential is based on the honesty and knowledge of their assignees. While some assignees declare all of their standard-essential patents as standard-essential, other declare some that are not standard-essential or even refuse from declaring them.

On September 23, Mr. Andre Herzberg presented a paper about this topic at the Summer School on Data and Algorithms for ST&I Studies in Leuven. The paper is co-authored by Prof. Dr. Martin G. Möhrle, Nils Denter and Eliza Stenzhorn. By conducting unsupervised and supervised machine learning in form of topic modeling and deep neural networks, the authors are able to distinguish standard-essential and non-standard-essential patents. Their findings show that specific assignee characteristics positively influence the likelihood of not declaring potential standard-essential patents as standard-essential.

For more information, please visit the Summer School on Data and Algorithms for ST&I Studies homepage (https://feb.kuleuven.be/drc/MSI/sti).