Name: Public defence: Maliheh Farrokhnia
Start: 2019-05-21
End: 2019-05-21

Maliheh Farrokhnia will defend her thesis "Searching for cultural heritage information: Ontology-based modeling of user needs."

10.00–10.45: Trial lecture: "Enhancing information retrieval using CIDOC-CRM: implications for user-side tools."
12.00: Public defence

The ordinary opponents are:

First opponent: Associate professor Koraljka Golub, Linnæus University
Second opponent: Associate professor Christian-Emil Smith Ore, University in Oslo

Leader of the committee: Associate professor Michael Preminger, OsloMet.

Leader of the public defence: Vice-Dean Ivar Harsløf, Faculty of Social Sciences, OsloMet.

The main supervisor is Professor Nils Pharo, OsloMet.

The co-supervisor is Professor Trond Aalberg, OsloMet.

Abstract

This PhD dissertation investigates users’ actual information needs when they search for cultural heritage information and examines how their needs map to ontologies such as CIDOC Conceptual Reference Model (CRM).

With the invention of the semantic web and increasing demand for global search across cultural heritage data, ontologies such as CIDOC CRM have been developed to provide an extensible semantic framework to formally describe the implicit and explicit relationships of cultural-heritage concepts.

This ontology provides a basis for meaningful integrated access to semantically heterogeneous data. Currently, ontologies are developed by domain experts and system developers. In this project, qualitative research incorporating user’s perspective and actual needs is used to complement our understanding of how well the ontologies model user’s needs.

With a naturalistic approach, this study has conducted contextual inquiry to explore user’s information needs. Analyzing the qualitative data reveals that the information needs can be grouped into four emergent categories of Fact-finding, Entity-finding, Resource-finding, and Context-finding. As the result of analysis, the model of semantic information needs has been introduced that can be considered as an extension to Taylor’s levels of information needs.

Based on the ontological analysis of information needs, some general patterns have been discovered and projected onto CIDOC CRM.

The findings of this research show that ontology-based information systems can be built upon a limited types of entities and properties to meet a broad range of information needs.