Overview
What is it and why is it important?
How does iSearch Analytics solve the problem?
Data linkages and disambiguation article (external link)
Disambiguation disclaimer
Overview
Disambiguation is the process of correctly merging name variations of individuals based on research topic, organization affiliation, and more. Disambiguation is an important part of data analysis when using names of grantees, authors, or other person-level data. iSearch Analytics includes rigorous name disambiguation methodologies to provide users with detailed person-level data and metrics.
What is it and why is it important?
There are many scenarios where the use of ambiguous names and traditional data linkages not based on name disambiguation could provide inaccurate results when performing analysis. A few examples include:
- People sometimes publish papers or apply for grants under different names, e.g. sometimes including middle initials or only a first name initial, etc. Alternatively, multiple people with the same name may publish papers or apply for grants, resulting in an over-merging of attributions. When names are not disambiguated, one individual may look like three different individuals, or three individuals with the same name may look like the same person.
- Another consideration in disambiguation is the question of funding versus attribution. For example, grants may be linked to a publication in which the Principal Investigator (PI) is not an author, however, their grant provided funding to perform the necessary research for the paper. Therefore, a PI may track and credit the paper’s influence from their support, despite the lack of authorship. You can always find publications linked to grants in this way, but it is important to keep in mind that the attributed impact of the paper would not be based on authorship.
- Timing is also an important factor to consider. Sometimes, a PI may link any grants they have received during their tenure to a publication. In these cases, one may find that a linked paper was published prior to a grant received funding; therefore, it is important to perform year-matched disambiguation to remove these illogical linkages.
Accounting for these examples and more, iSearch Analytics uses name disambiguation to improve the accuracy of authorship attribution.
How does iSearch Analytics solve the problem?
Disambiguation links are used in iSearch Analytics to link authors to grant awards they have received prior to the publication date of a particular article. They are also used when finding literature based on a grantee’s name. Name disambiguation can alleviate missing links traditionally made using other methodologies, e.g. SPIRES. You can view these links by clicking on the View Linked Data button in the relevant datasets (data are available in Grant Records and Literature only). You can find whether/how people disambiguation was used to link specific data in the Record View or by exporting such data.
View Linked Data button above Topic Explorer
Record view
Additionally, Person History and Person Profiles capture all grants and publications that can be attributed to an individual. The Person Profile also contains a list of all name variations we have captured for that individual. The profiles have been quality checked for accuracy to identify over-merging of prolific authors based on name similarities, and disambiguated data can also be found here. Lastly, you can find whether/how people disambiguation was used to link specific data in the Record View or by exporting such data.
Associated People (Person History)
Person Profile
Disambiguation disclaimer
Linking does not infer an intellectual property connection between the award and publication or clinical trial.