利用语义技术实现武装冲突事件数据的快速集成

Exploiting Semantics for Conflict Event Data Integration

针对如何集成来源多样、数据异构、表示不同的大数据的问题,提出一种基于本体和Karma建模的大数据集成方案,将其应用于武装冲突事件数据的快速集成;分析武装冲突事件数据项目ACLED的数据集,构建武装冲突事件本体进行数据语义描述;使用数据集成开发工具Karma针对多源异构数据建立相应的Karma模型完成数据的语义映射、数据清洗和整理,实现多源异构数据的快速集成;编写SPARQL查询验证了集成发布的RDF数据和Karma建模的正确性,也进一步验证了基于本体和Karma建模实现多源异构数据集成方法的有效性。

Aiming at processing heterogeneous data sets, we provide an integration approach for big data,and apply this approach in integrating conflict event data. By Analyzing data of ACLED （Armed ConflictLocation ＆ Event Data） Project, we build conflict event ontology to describe semantic for ACLED data.We describe Karma, a system that implements our approach and show how Karma can be applied to integratedata in ACLED data. By building Karma models to implement data cleaning and semantic mapping,and integrate heterogeneous data to RDF data. Using SPARQL query verifies the validity of RDF data andKarma models, and it also verifies the validity of our integration approach for big data.