Structuring the Cultural Domain with an Upper Ontology of Culture

Introduction

Study of cultural similarities and differences is an important research topic for many disciplines such as psychology, sociology, anthropology, archaeology, museology, communication, management and business to cite but a few. This presents many potential opportunities for IT specialists to develop culturally-aware technology.

However the various coexisting and competing discipline-specific approaches and methods that have been developed, the genuine fuzziness of folk language that people use to discuss cultural matters and the ill-defined nature of cultural questioning lead us to the statement that cultural awareness is particularly difficult to address consistently in information technology.

As mentioned by Lane and his colleagues, “ill-defined domains, in contrast to those that are well defined, are characterized by problems that tend to lack consistent, unambiguous, and generalizable solutions” (Lane et al., 2007). Indeed, culture is an easy-to-use concept in everyday discussions. But it becomes much more difficult to deal with it when the time comes to give it a proper and consensual definition, to determine its constituents or to describe its specifics: in other words to consider it in a scientific manner.

Even when this is the case, cultural notions and terminologies sometimes differ from one discipline to another. The research focus of disciplines may also vary. For instance, some disciplines such as anthropology may be interested in discussing cultural artefacts, whereas others such as psychology may not be. Moreover, a huge amount of data is annually produced by research that could nurture the development of culturally-aware systems. But the computerization of such cultural data as well as the interoperability and centralization of resulting collections are currently limited. Finally, mastering the various research initiatives on culture is a difficult and highly time-consuming task, and the process of knowledge acquisition may frequently be limited by commercial realities and constraints such as deadlines. This could potentially result in ill-designed systems that would extensively rely on ethnocentric views of their development team, implying cultural misconceptions and stereotypes. This may thus affect the credibility of the resulting application, potentially increasing users’ misconceptions about a targeted culture, or reducing the efficiency of human-computer interactions.

Using formal ontology engineering techniques, the Upper Ontology of Culture (UOC) project aims to develop a generic conceptualization of the cultural domain, neutral and interdisciplinary, by identifying the cultural backbone common to culture-related disciplines and activities. Such theory-driven conceptualization has many interests for the development of research on artificial cultural awareness:

• 1.

  To allow development teams to consider cultures in a scientifically-sound and cross-disciplinary way, i.e. to propose appropriate guidelines on what development teams should focus on when addressing a specific cultural issue,

• 2.

  To propose ways of appropriately computerizing cultural aspects of a given problem by suggesting templates for theory-driven data structures and data management processes,

• 3.

  To promote interoperability by enforcing the consistency of cultural data modelling between systems, thus facilitating reuse of computerized cultural data,

• 4.

  To promote cultural automatic reasoning, thus allowing systems to take culturally-informed decisions that may impact on their internal processing as well as on human-computer interaction.