Automated TRIZ Domain Mapping

Abstract

The automatic analysis of patents is still one of the main challenges in R&D, particularly in terms of establishing automatic states of the art. Indeed, this is still mostly done manually, which is very time-consuming. The progress of artificial intelligence allows us to go a step further in the understanding of patents and in particular of the issues they address. In this paper we present an end-to-end tool that allows us to map the main trends in term of research directions in a sector in a few minutes from a simple keyword search. To do so, we will rely on TRIZ formalization with contradictions and evaluation parameters.

BibTeX

@inproceedings{guarino:hal-04054703,
  TITLE = {{Automated TRIZ Domain Mapping}},
  AUTHOR = {Guarino, Guillaume and Cavallucci, Denis},
  URL = {https://hal.science/hal-04054703},
  BOOKTITLE = {{IFIP Advances in Information and Communication Technology}},
  ADDRESS = {Warsaw, Poland},
  EDITOR = {Robert Nowak and Jerzy Chrząszcz and Stelian Brad},
  PUBLISHER = {{Springer International Publishing}},
  SERIES = {Systematic Innovation Partnerships with Artificial Intelligence and Information Technology},
  VOLUME = {AICT-655},
  PAGES = {198-208},
  YEAR = {2022},
  MONTH = Sep,
  DOI = {10.1007/978-3-031-17288-5\_18},
  KEYWORDS = {Domain mapping ; TRIZ ; Deep learning},
  PDF = {https://hal.science/hal-04054703v1/file/527541_1_En_18_Chapter.pdf},
  HAL_ID = {hal-04054703},
  HAL_VERSION = {v1},
}