Mining non-lattice subgraphs for detecting missing hierarchical relations and concepts in SNOMED CT.
Mining non-lattice subgraphs for detecting missing hierarchical relations and concepts in SNOMED CT.
复制标题
DOI：
10.1093/jamia/ocw175
复制
发表时间：
2017-07-01
期刊：
影响因子：
--
通讯作者：
Zhang GQ
中科院分区：
文献类型：
--
作者：
Cui L;Zhu W;Tao S;Case JT;Bodenreider O;Zhang GQ
目的：大型本体系统（如SNOMED CT）的质量保证是术语管理生命周期中不可或缺的一部分。我们介绍了一种混合结构-词汇的方法，可扩展的和系统的发现丢失的层次关系和概念在SNOMED CT。 材料和方法：使用可扩展的MapReduce算法穷举提取SNOMED CT中的所有非格子子图（结构部分）。四个词汇模式（词汇部分）被确定在提取的非格子子图。表现出这种词汇模式的非格子子图通常表示缺少层次关系或概念。每个词汇模式都与潜在的特定类型的错误相关联。 结果如下：将结构-词汇方法应用于SNOMED CT（2015年9月美国版），我们发现了6801个与这些词汇模式匹配的非格子子图，其中2046个可以进行目视检查。我们评估了100个小型子图的随机样本，其中59个由领域专家详细审查。经审查的所有子图都含有经专家确认的错误。最常见的错误类型是由于概念建模不完整或不一致而导致的is-a关系缺失。 结论：我们的混合结构-词汇方法是创新的，并证明有效的，不仅在检测SNOMED CT中的错误，但也建议补救这些错误。
Objective: Quality assurance of large ontological systems such as SNOMED CT is an indispensable part of the terminology management lifecycle. We introduce a hybrid structural-lexical method for scalable and systematic discovery of missing hierarchical relations and concepts in SNOMED CT. Material and Methods: All non-lattice subgraphs (the structural part) in SNOMED CT are exhaustively extracted using a scalable MapReduce algorithm. Four lexical patterns (the lexical part) are identified among the extracted non-lattice subgraphs. Non-lattice subgraphs exhibiting such lexical patterns are often indicative of missing hierarchical relations or concepts. Each lexical pattern is associated with a potential specific type of error. Results: Applying the structural-lexical method to SNOMED CT (September 2015 US edition), we found 6801 non-lattice subgraphs that matched these lexical patterns, of which 2046 were amenable to visual inspection. We evaluated a random sample of 100 small subgraphs, of which 59 were reviewed in detail by domain experts. All the subgraphs reviewed contained errors confirmed by the experts. The most frequent type of error was missing is-a relations due to incomplete or inconsistent modeling of the concepts. Conclusions: Our hybrid structural-lexical method is innovative and proved effective not only in detecting errors in SNOMED CT, but also in suggesting remediation for these errors.
登录
查看更多内容
影响因子：
4.5
作者：
Wang, Yue;Halper, Michael;Wei, Duo;Gu, Huanying;Perl, Yehoshua;Xu, Junchuan;Elhanan, Gai;Chen, Yan;Spackman, Kent A.;Case, James T.;Hripcsak, George
通讯作者：
Hripcsak, George
DOI：
10.1109/bigdata.2014.7004301
发表时间：
2014-10
期刊：
Proceedings : ... IEEE International Conference on Big Data. IEEE International Conference on Big Data
影响因子：
--
作者：
Zhang GQ;Zhu W;Sun M;Tao S;Bodenreider O;Cui L
通讯作者：
Cui L
影响因子：
4.5
作者：
Agrawal, Ankur;Elhanan, Gai
通讯作者：
Elhanan, Gai
DOI：
10.1136/amiajnl-2014-003151
发表时间：
2015-05-01
影响因子：
6.4
作者：
Ochs, Christopher;Geller, James;Wei, Zhi
通讯作者：
Wei, Zhi
DOI：
10.1007/978-3-642-17749-1_18
发表时间：
2010
期刊：
LECTURE NOTES IN ARTIFICIAL INTELLIGENCE
影响因子：
--
作者：
Zhang, Guo-Qiang;Bodenreider, Olivier
通讯作者：
Bodenreider, Olivier