Page 113 - JOURNAL OF LIBRARY SCIENCE IN CHINA 2015 Vol. 41

112 Journal of Library Science in China, Vol. 7, 2015

effectively obtain and utilize them on scientific research.
Recently, in the history domain, knowledge organization, semantic representation, and
knowledge discovery, etc. have been studied by many researchers. Several ontologies have been
developed to describe and organize historical knowledge, such as “Kuomintang-Communist
Cooperation” historical ontology (Dong et al., 2006), “Northeast Anti-Japanese Struggles”
historical ontology (Wu, 2012), “Zizhi Tongjian” historical ontology (Peng & Song, 2010), and
“Three kingdoms” domain ontology (Liao, 2011). Dong, Xu, Wang and Yu (2014) designed a
semantic process based on the Twenty-Four Histories semantic data to construct a knowledge
base, applied semantic techniques to discover implicit knowledge in historical records, and then
built a Basic Historical Analysis Platform. Their work provides important practice references
for subsequent research. Hyvönen, Alm, and Kuittinen (2007) built a historical event ontology
focusing on Finnish history and applied it in the semantic portal “CultureSampo - Finnish Culture
on the Semantic Web”. Corda, Bennett, and Dimitrova (2011) proposes a logical model of event
ontologies for exploring connections in the history domain. Ide and Woolner (2007) outlined a
model for historical ontologies, which is temporally contextualized and represents relations among
entities during different temporal intervals.
Based on the above mentioned related researches, this paper proposes a method, which
is called “Mining down, Organizing up”, to represent and organize historical knowledge on
contemporary China. Based on this idea, we developed a system to implement historical knowledge
representation, reorganization and other new applications.

1 Main idea

The aim of the project “Knowledge web of the history of the People’s Republic of China” is to
popularize the historical knowledge on contemporary China and facilitate historical education. The main
problem of this project is how to help historical experts extract and represent important knowledge of
more than 60 years of history using automatic information processing technologies, and organize related
text items, from historical information resources, such as reference books “Dictionary of the history
of the Chinese Communist Party”, “Encyclopedia of the national history of the People’s Republic of
China”, “Conspectus of Chinese modern history”, “Chronicle of the People’s Republic of China” and so
on.
To solve the problem, this paper proposes the method of “Mining down, Organizing up”. Based
on a contemporary Chinese historical ontology, this method extracts knowledge objects and facts
from unstructured historical texts to build a historical knowledge network on contemporary China,
and then organizes knowledge in multidimensional at a higher level based on the relations such as
time, subclass, hierarchy, and statistics. The framework of this method is shown in Figure 1.
Specifically, “Mining down” is a deconstruction process, which transforms knowledge in
historical texts into a series of historical knowledge objects, facts and text items which form