Information Technology Reference
In-Depth Information
2 Framework for Visualizing Changes in Relationships
between Historical Figures
The aims of the proposed framework are: to extract networks of historical
figures and characteristics of relationships between such figures from
historical documents; visualize these for ease of understanding; provide
interactive functions to explore networks of historical figures and temporal
changes in their relationships.
The outline of the proposed system is shown in Figure 8.1. A historical
document database (Figure 8.1 (a)) stores records of events that have
attribute values such as names of people, keywords, and dates related to
the events. Users can select records related to a specific viewpoint they are
interested in by specifying ranges of years, lists of people, and/or
keywords. The system then generates a network of historical figures from
the selected records by extracting co-occurring people in the same events
(Figure 8.1 (b)). If two people have a high co-occurrence ratio, they have a
connection. Users can extract networks related to specific people - for
example ODA Nobunaga, who was the best-known feudal lord in Japanese
history, or TOKUGAWA Ieyasu, who was the founder and first shogun of
the Tokugawa shogunate - by inputting their names. They can also extract
networks related to a battle in the warring states period in Japanese history
(called Sengoku Jidai) by inputting the range of years related to Sengoku
Jidai and keywords such as “battle”, “attack”, and/or “rebellion”. Moreover,
they can extract and compare networks related to various viewpoints.
The proposed framework allows users to extract a network of a
specific viewpoint with a specified time window, e.g. annual or decade.
The historical document database includes keyword fields that summarize
events. We can extract information about why two or more people have
connections by using events and keywords related to them. However,
keywords are too diverse for visualizing the reasons for the connections.
We therefore extract clusters of keywords and assign colours to extracted
clusters for easy recognition of characteristics of the relationships (Figure
8.1 (c)).
Our framework then visualizes a time-varying network extracted as in
Figure 8.1 (b) by assigning a person to a node of the network and a
connection between people to an edge (Figure 8.1 (d)). It also assigns
colours to the edge that has characteristics defined by co-occurring clusters
of keywords between two people (Figure 8.1 (d)). This enables visual
exploration of temporal changes in relationships between people.
Search WWH ::
Custom Search