Semantic Data Integration for Life Science Entities
Data fusion; Duplicate detection; LSID; Object identification
An entity is the representation of a (not necessarily physical) real-world object, such as a gene, a protein, or a disease, within a database. To integrate information about the same entities from different databases, these representations must be analyzed to uncover the corresponding underlying objects. This process is called entity identification. A variation of entity identification is duplicate detection, which analyses two or more entities to determine whether they represent the same real-world object or not. Finally, data fusion is the process of generating a single, homogeneous representation from multiple, possibly inconsistent entities that represent the same real-world object.
When entities have globally unique keys, such as ISBN numbers in the case of books, entity identification and duplicate detection are simple. However, in life science databases, one usually has only descriptive...
- 3.Gibson G, Muse SV. A primer of genome science. Sunderland: Sinauer Associates; 2001.Google Scholar
- 4.Karp P.D. Models of identifiers. In: Proceedings of the 2nd Meeting on Interconnection of Molecular Biology Databases; 1995.Google Scholar
- 8.Müller H, Naumann F, Freytag J.-C. Data quality in genome databases. In: Proceedings of the 8th Conference on Information Quality; 2003.Google Scholar