Skip to main content
ExLibris
  • Subscribe by RSS
  • Ex Libris Knowledge Center

    DataCite in CDI

    DataCite's unique content

    DataCite supports researchers by helping them to find, identify, and cite research datasets.  It assigns persistent identifiers to citable datasets. All DataCite DOIs resolve to a public landing page that contains information about the associated dataset and a direct link to the dataset itself.  A wide range of organizations contribute records to the DataCite Metadata Store. 

    Lately, CDI was enriched with new content from DataCite, but we have received many justified complaints that DataCite's metadata, in many cases, can be problematic and contains mistakes. When DataCite records are merged in CDI with records from other providers, then the wrong metadata can cause the OpenURL to fail in creating valid links, or give end-users false indications and information derived from incorrect metadata.

    To solve these problems, we have reduced DataCite content in CDI to include only content with a high level of uniqueness, meaning content that is not indexed from other sources in CDI. Since we filtered DataCite content by content type, there still might be in the index a small number of records that are not entirely unique to DataCite that came for content types that hold a high level of uniqueness. However, these records should represent only a small portion of DataCite's content in CDI.

    By doing that, merged records that are covered in CDI from other sources will no longer be merged with DataCite's records. As a result, the OpenURL won't fail due to erroneous metadata, and merged records won't provide false indications and information originated from DataCite's metadata. The unique content of DataCite which won't match and merge with other records in the index (e.g. DataCite's data sets records) can still be discovered in CDI, and will provide direct linking to the resource.

    Please note that all of DataCite removed content exists in CDI from other sources, so there won't be any effect on the search experience or the results.

    We are also aware that DataCite's content is not clean, and one might find records that should not refer to real resources and would be considered spam content. We are working closely with DatatCite to remove this content from the index constantly.

    Open-Access and A&I content

    DataCite content includes both open-access and full-text content, and also A&I content which do not include full-text but abstracts, citations, and bibliographic information. The DataCite records use direct linking, which means all links will redirect the end-user to DataCite platform, and from there another link will be offered to the target resource.

    However, when CDI records get merged, the direct link gets precedence over linkresolver records if the direct link is not open-access. In some cases, this can cause a direct link from DataCite A&I record to an abstract to obscure full-text OpenURL from other participants in the logical record. Although we reduce DataCite content to include only content with high level of uniqueness that is not likely to match and merge, If you still encounter this problem, then we would recommend activating in your Alma only the DataCite open-access collection and deactivate the DataCite general collection.

    Please activate the following collection:

    Collection ID Collection name     Provider Name     FullText Rights
    615300000000000136 DataCite (Open Access) DataCite OpenAccess


    And deactivate the following collection: 

    Collection ID Collection name     Provider Name     FullText Rights
    6114051530000041 DataCite DataCite Partially Open Access