A data source I am using to discover potential ORCID ids may/may not have initials as given-names. If your Lucene index had an EdgeNGram on given-names, to help accommodate incoming queries that contain abbreviated given-names, this would be immensely useful. Alternatively, an additional, computed field like "given-name-initials" with EdgeNGram would be equally useful and would not change the behaviour of your existing Solr field, "given-names". A relatively simple regex-based method from https://github.com/berkmancenter/namae/blob/master/lib/namae/name.rb#L178 might prove useful.
Please sign in to leave a comment.