基于跨语本体重用的维语本体构建方法  

Uygur ontology construction based on cross-language ontology reuse

作　　者：李连倍 刘胜全[1,2] 刘艳[1] 杨倩倩[1] 

机构地区：[1]新疆大学信息科学与工程学院,乌鲁木齐830046 [2]新疆大学现代教育技术中心,乌鲁木齐830046

出　　处：《计算机工程与应用》2015年第11期104-108,共5页Computer Engineering and Applications

基　　金：新疆维吾尔自治区科技攻关项目(No.200931103);新疆大学自然科学基金(No.XY110121)

摘　　要：针对维英本体共性知识的获取问题,提出一种基于跨语本体重用的快速构建维语领域本体方法。该方法将初始维语本体转换为英语本体,通过本体选择、映射和合并等过程对其丰富,达到一定阈值,转换为维语本体。提出了数据源势、本体势等概念和构建维语本体的数据模型。基于该方法构造了一个旅游领域本体实例,转换率达到78.8%,充分验证了该方法的可行性与有效性。Concerning the Uyghur-English ontology common knowledge acquisition problem, this paper proposes a Uyghur domain ontology fast-construction method based on cross-language ontology reuse. It converts the initial Uygur ontology into an English ontology, enriches it via some processes such as ontology search, mapping and merging, transforming it into Uygur domain ontology when reaches certain threshold. The conception such as data source cardinality, ontology cardinality and the data model for constructing Uyghur are proposed. A travel domain ontology with transformation rate reaching78.8% demonstrates the feasibility and effectiveness is constructed.

关 键 词：跨语本体重用 维语本体 稀疏数据 本体映射 本体合并 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]