Open adambuttrick opened 3 months ago
Improvement to affiliation matching is consistent with possible degradation to our affiliation matching as a result of less location information in v2 vs. v1, as reported in https://github.com/ror-community/ror-roadmap/issues/243
Version v2
Describe the problem you would like to solve In developing new affiliation matching, OpenAlex found that the inclusion of postal codes associated with the organization's address can significantly improve the confidence scores of new matches for previously unmatched affiliation strings. However, postal codes are not currently included as a field in our location metadata, limiting their ability to leverage ROR for this improvement.
Describe the schema change that you would like in order to solve the problem OpenAlex proposes adding a "postal_code" field to the ROR schema for institutions. This field would contain the main address postal code for each institution.
Who would benefit from this change? This change would benefit:
Additional information ROR would need to investigate how to derive for all, most, or the most commonly referenced subset of organizations in our data. We would also need to change our approach to curating location metadata more generally to support. For all records now, the only value we identify and index is the Geonames ID for the city in which the organization is located.