Is it possible to find out more information about how OpenSearch performs its indexing or retrieval when it comes to indexing longer documents?
(That is, longer than a scientific abstract, for example).
Longer documents can contain information about a diverse set of subjects and therefore, more accurate results could be returned if the document was to be broken down in overlapping text passages that are smaller in extent. Another approach would be to store a long document as the individual sections that compose it.
Are any of these “subdivision” approaches handled automatically by OpenSearch or a relevant plugin to store and search long form documents?