
A Method of Event Ontology Mapping
Xu Wang
1
, Wei Liu
1
, Yujia Zhang
1
, Yue Tan
1
, Feijing Liu
1
1
School of Computer Engineering and Science, Shanghai University, Shanghai
200444, China
{liuw@shu.edu.cn,wangx89@126.com}
Abstract. Ontology mapping is an important solution to ensure
interoperability while integrating heterogeneous and distributed data
sources. This paper proposes an approach for ontology mapping based
on event, which enable the mapping between event-based information
with more abundant semantics. Firstly, this paper gives the definition of
event ontology mapping, and then proposes an comprehensive semantic
similarity calculation model based on the similarity of events and event
structures. Experiments show that the proposed model can effectively
find the semantic relations of event in two event ontologies.
Keywords: Event Ontology Mapping, Semantic Similarity, Semantic
Neighbor
1 Introduction
Ontology is the core of semantic web, and it is widely used in the areas such as data
integration, metadata management and data exchange. However, due to various
cultural backgrounds and comprehensions of ontology, different usage habits of terms
leads to an ocean of ontologies with different structures, which causes the limited
integration and limited sharing of data. In order to accomplish interoperability of
semantics ontology, it is necessary to establish mapping relations among
heterogeneous ontologies. Whereas most current ontology matching methods are
based on calculation of concepts correspondence [1], it will result in semantic
information loss while processing event-based data because of “Tennis Problem” [2].
In recent years, research on the use of events as a key concept for representing
knowledge, organizing and structuring media on the web is surging, especially in
semantic web community. Event ontology is a shared, formal and explicit
specification of an event class system model that exists objectively, which has
become new paradigm for describing and reasoning event-based knowledge in web.
Event-based ontology matching is necessary for integrating heterogeneous data
sources in event-centered domain application, such as emergency response, public
opinion monitoring, history and cultural heritage, etc.
Aims at the event ontology mapping in semantic web application, this paper firstly
gives the definition of event ontology mapping, and then proposes a semantic
similarity calculation model based on the similarity of event classes and event class
structure. The similarity of event classes is calculated by the similarity of common
elements in different event. The similarity of event class structure is obtained by
calculating the similarity of semantic neighbor sets. The semantic neighbors of event
class could be found in a certain semantic radius in event ontology network structure.