Molecular Connections’ Author & Affiliation Disambiguation service includes:
Author & Co-author knit development
RDF triple stores management and development
Name Entity Disambiguation
Molecular Connections uses a fuzzy logic based technology to disambiguate authors /institutions automatically and the “hybrid” workflow provides manual muscle to improve the accuracy of the disambiguation to more than 99%. The approach further utilizes the meta-data surrounding authors’ profile (First-, Middle- and Last- name) and institution/affiliations from various data sources. Author record meta-data is further complemented with entities such as the institutions’ address, Digital Object Identifier (DOI), co-authors in the publications, public profile identifiers (ORCID, Publons, etc.) and funding history (grants) to create a comprehensive database around the author and also to increase the efficacy of disambiguation.
Molecular Connections has generated an in-house proprietary database, which now contains a master list of more than 50,000 institutions (Academic and Commercial) from over 100 countries, and over 1,00,000 aliases enriched across multiple domains.
Proprietary Author & Affiliation Disambiguation System – MC Identify™
MC Identify™ facilitates identification of various related components including co-author profiles, net output of a research organization of interest etc., and helps in referee searches.
Select Case Studies
Molecular Connections partnered with a major life-sciences publisher to device a hybrid method to deal with author name and institution disambiguation – through its proprietary offering MC Identify™. The strategic partnership has helped the publisher create author profiles with more than 99% accuracy. The workflow also facilitates identification of various related components including co-author profiles, net output of a research organization of interest etc., and helps in referee searches.