Symptom
- When searching for an applicant’s full name in Chinese, Japanese, or Korean, the system returns partial matches (individual characters) rather than only exact full-name matches.
- Candidates whose full name exactly matches the query are not being prioritized at the top of the search results.
Environment
SAP SuccessFactors Recruiting Management
Reproducing the Issue
- Navigate to the latest applicant workbench.
- Search for a name, e.g., "XYZ" (where "XYZ" represents three CJK characters).
- The search results include any combinations that contain X or Y or Z, such as "XYA" or "BCZ".
Cause
It is related to how CJK (Chinese, Japanese, Korean) text is processed by the underlying Solr search engine.
Resolution
For CJK languages, Solr uses specialized analyzers that are designed to handle the text by breaking words or names like "XYZ" into individual searchable characters. For example: ["X", "Y", "Z"].
Because of that tokenization, any name containing one or more of those characters will appear in the results. This explains why partial matches like "XYA" or "BCZ" are also returned.
If you would like to request a change for this behavior, please consider submitting an enhancement idea to the Customer Influence page. This platform enables users to vote for each other's ideas, post comments, and receive direct feedback from the Product Management team. Refer to the KBA 2090228 for more information.
See Also
Keywords
rcm, application, candidate, applicant, solr, search, accuracy, find, name, fuzzy, partial, exact, match, Chinese, Japanese, Korean, CJK, language, character , KBA , LOD-SF-RCM-APP-LWB , Latest applicant WorkBench , How To
SAP Knowledge Base Article - Public