Skip to Main content Skip to Navigation
Conference papers

Building Time-Affordable Cultural Ontologies Using an Emic Approach

Abstract : Recently, studies about culturally-aware systems have arisen to address digitized culture. Among these systems those enculturated driven by cultural knowledge embed culture in their design. To deal with the specifics of cultural groups, the development of machine-readable cultural knowledge representations can provide a substantial help. In this research we present a process to build time-affordable, emic, conceptually-sound and machine-readable cultural representations. These representations originate from Cognitive Anthropology. They follow a three steps methodology: ethnographic sampling, individuals’ personal knowledge elicitation and cultural consensus analysis. We use lexico-semantic relation extraction as a mean to automatically elicit knowledge structures. Their formalisation is achieved through Ontology Engineering.We conducted experiments to build three cultural ontologies in order to assess the whole process. It came out that with the lexico-semantic relation extraction technique, the best representations we can obtain are consensually-limited, incomplete and contain some errors. However, many clues indicate that these problems should be solved by using higher quality elicitation techniques.
Complete list of metadatas

Cited literature [13 references]  Display  Hide  Download

https://hal.inria.fr/hal-01626988
Contributor : Hal Ifip <>
Submitted on : Tuesday, October 31, 2017 - 3:43:31 PM
Last modification on : Wednesday, April 22, 2020 - 1:36:08 PM
Document(s) archivé(s) le : Thursday, February 1, 2018 - 1:22:57 PM

File

447996_1_En_8_Chapter.pdf
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Jean Petit, Jean-Charles Boisson, Francis Rousseaux. Building Time-Affordable Cultural Ontologies Using an Emic Approach. 3rd IFIP International Workshop on Artificial Intelligence for Knowledge Management (AI4KM), Jul 2015, Buenos Aires, Argentina. pp.130-148, ⟨10.1007/978-3-319-55970-4_8⟩. ⟨hal-01626988⟩

Share

Metrics

Record views

215

Files downloads

401