Information Technology Reference
In-Depth Information
such as names of buildings and events are more likely to turn up if referred
locations are close enough. Since the dictionaries and domain-topic repository
are reusable, the more pieces of information are stored to them, the easier the
process gets to extract spatiotemporal information from other sources.
The collection of fragmented sentences based on the narrative data from the
regional community members can be considered as a clue to understand collec-
tive culture in the region. As we discussed, even though the community members
did not meet each other those days, and did not go through the externalization
process as Valsiner proposes, it is highly likely that they share the similar expe-
rience, which means there exists the possibility of indirect externalization. For
example, a research participant has some story about his/her experience in the
elementary school, and another research participant went to the same elemen-
tary school in the similar time period. They might or might not have known each
other then, but it is likely that they had some friends, teachers, and school events
in common. Through such intermediates, each personal culture gets externalized
and shared by the community. In the end, the collection of personal stories based
on spatiotemporal information can be considered as collective culture like this
case.
3 Visualizing Collective Culture
The following sections introduce the KC system as a platform for visualizing
regional collective culture. It imports the contextualized fragments of sentences
generated by the procedure above, visualizes them onto a virtual 3D space based
on spatiotemporal information, and assists researchers to understand collective
culture shared by the community members in a region.
3.1 User Interface
The KC system provides a web-based virtual 3D space to its user (see Fig. 2).
The space includes a two dimensional map with timescale, and is used to store
and plot the contextualized fragments of sentences extracted from the narrative
data based in spatiotemporal information.
Fig.3showsa3Dspacefromtwodifferent angles. It consists of a map with
timescale. The user can rotate the 3D space to browse the fragmented sentences
or narratives from suitable angles.
3.2 Data Structure
Expecting KC to be sharable, we apply RDF (Resource Description Framework)/
OWL (Web Ontology Language) [12] to define the classes for storing narrative
data. RDF and OWL are Semantic Web standards that provide a framework for
sharing and reuse of data on the Web. Fig. 4 shows a part of the class definition
in KC. Under the
Document
class, the following classes are defined.
