Augmented intelligence for state-of-the-art patent search

Ana Hafner, Nadja Damij*, Dolores Modic

*Corresponding author for this work

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

3 Downloads (Pure)

Abstract

The volume of patent data is increasing, which is a big challenge to patent examiners as well as to all inventive companies and individuals. In this paper we take the view of individual inventors who believe they invented something new. Artificial intelligence brings a promise to support their prior art search for existing (similar) inventions with machine learning and deep learning algorithms. We discuss the potential of artificial intelligence in prior art searching. We present an experiment, based on a real-life invention, comparing relevant patents we got from Boolean keyword searching with those from the semantic search supported by artificial intelligence. We can confirm that artificial intelligence has great potential in this field. However, presently it is not yet able to make traditional patent search engines obsolete, hence it still fits better with the notions of augmented intelligence or expertise.
Original languageEnglish
Title of host publication2022 IEEE Technology & Engineering Management Society Conference - Europe (TEMSCON-EUR)
Subtitle of host publicationSocietal Challenges: Technology, Transitions and Resilience Virtual Conference
EditorsAlexander Brem, Tugrul Daim, Yogesh K. Dwivedi, Jelena Španjol, James Fleck
Place of PublicationPiscataway, US
PublisherIEEE
Pages61-66
Number of pages6
ISBN (Electronic)9781665483131
ISBN (Print)9781665483148
DOIs
Publication statusPublished - 25 Apr 2022
Event2022 IEEE Technology and Engineering Management Conference (TEMSCON EUROPE) - Virtual, Turkey
Duration: 25 Apr 202227 Apr 2022

Conference

Conference2022 IEEE Technology and Engineering Management Conference (TEMSCON EUROPE)
Country/TerritoryTurkey
Period25/04/2227/04/22

Fingerprint

Dive into the research topics of 'Augmented intelligence for state-of-the-art patent search'. Together they form a unique fingerprint.

Cite this