Difference between revisions of "Data integration"
Jump to navigation
Jump to search
(Created page with "<div id="CSB"> <div class="b1"> Data integration </div> : Loose coupling, combining datasets to improve annotation quality, evidence combination; Ontologies __TOC__ ==Cont...") |
m |
||
| Line 4: | Line 4: | ||
</div> | </div> | ||
| − | + | ||
| + | {{dev}} | ||
| + | |||
| + | |||
| + | Data integration discusses how data elements are connected (tightly or loosely); combining datasets to improve annotation quality; principles of evidence combination; ontologies. | ||
__TOC__ | __TOC__ | ||
| Line 16: | Line 20: | ||
===Evidence combination=== | ===Evidence combination=== | ||
Annotation quality can be improved if evidence can be used from varying sources. But how can such evidence be weighted and combined? How can confidence scores be constructed? | Annotation quality can be improved if evidence can be used from varying sources. But how can such evidence be weighted and combined? How can confidence scores be constructed? | ||
| + | |||
===Ontologies=== | ===Ontologies=== | ||
Revision as of 12:18, 18 January 2012
Data integration
This page is a placeholder, or under current development; it is here principally to establish the logical framework of the site. The material on this page is correct, but incomplete.
Data integration discusses how data elements are connected (tightly or loosely); combining datasets to improve annotation quality; principles of evidence combination; ontologies.
Contents
Contents
Cross referencing
Biological identifiers can cross reference each other IF their semantics allow mapping from one to the other...
Evidence combination
Annotation quality can be improved if evidence can be used from varying sources. But how can such evidence be weighted and combined? How can confidence scores be constructed?
Ontologies
Ontologies describe the semantics of a field of knowledge. They are indispensable for data integration across non-uniform databases.
Introductory reading
Exercises
References
