Information Technology Reference
In-Depth Information
space. The point indicates the Senbon-Sanjo area in the central Kyoto from
the 1911s to the 1927s. According to the storyFragments provided by research
participants, there used to be some movie theaters and an enterprise related with
movie industry in that area. Since there are no movie theaters or movie-related
firms in the area at the present day, the place and time can be considered as a
crucial point to explore or restore collective culture regarding the once-flourishing
film industry in Kyoto.
5 Related Works
Our approach is closely related to that of GIS-based narrative analysis. For
example, Kwan et al. [5] develop the method called “geo-narrative,” which ex-
tends current GIS capabilities to analyze and interpret narrative materials such
as oral histories, life histories, and biographies. In this method, researchers are
supposed to go over narratives line by line, interpret sentences, and then plot
them onto GIS space. Although it provides a precise way to examine narrative
materials, its process of data plotting cannot exclude the researchers' subjective
interpretation.
Other researchers are exploring the possibilities of automatic geocoding from
text data. For example, Yao et al. (see [13], [14], and [15]) have been working on
the fundamental research on “qualitative location (QL)” which refers to a spa-
tial location using linguistic terms such as qualitative spatial descriptions. Their
research proposes a way to process multiplicity and uncertainty in QL described
in natural language. Angel et al. [1] also advocate a methodology for the semiau-
tomatic geocoding of persistent Web pages in the form of collaborative human
intervention to improve on automatic geocoding results. While their techniques
are universally applicable, their approaches cannot handle various proper nouns.
CulturalSampo [3] is an integrated virtual museum with the combination of
semantic web technology and web-based geographical system. It allows its user
to browse cultural artifacts in a specific region in a certain time. The target
domain of this system, however, is the public history or cultural artifacts which
are regularly exhibited in real or virtual museums.
Different from the researches above, our approach uniquely combines text
mining and semantic web technology to extract spatiotemporal information from
the narrative data, and helps the researcher understand personal and collective
cultures with the visualization of narratives in the virtual 3D space.
6Conluon
Taking Valsiner's socio-cultural definition of culture, this paper focused on spa-
tiotemporal information to capture personal and collective cultures, since such
information regularly appears in people's life stories. To analyze the culture, we
propose a method to extract personal culture from their stories.
We analyzed oral histories, narrative data from three research participants
in Kyoto's Rakusai area, and stored extracted spatiotemporal information to
