How can I filter websites by language in Google Vertex AI search?

[OS-leonardoprat]

In the now deprecated Custom Search Engine there was a filed "lr" that would filter by the language. When using the advanced indexing there is no explanation or documentation on how I can filter by language to keep only english documents.
Documentation is lacking regarding Google Inferred fields when using advanced search.
Is the language being inferred by google automatically?
If yes whats the name of the field?
Do I need to declare the field in the schema so that its available??

[MJane]

Hi @OS-leonardoprat, 

Welcome to the Google Cloud Community!

Google Vertex AI Search advanced indexing does not provide a direct replacement for the "lr" parameter from the deprecated Custom Search Engine, which was used for language filtering. However, you can achieve similar functionality by creating and using custom metadata fields. While Google's search technologies may automatically infer the language of documents, this inferred language is not exposed as a filterable field named "language" unless explicitly configured.

You might consider enabling language filtering by defining a custom metadata field, like "language," in your data store schema. Set this field as "Indexable" and "Filterable." During data ingestion, you could populate it with the relevant language codes for each document. Once indexed, applying filter expressions in your search queries targeting this "language" field might help restrict search results to documents in your desired language.

For more information about filtering website search you may read this documentation. 

Was this helpful? If so, please accept this answer as “Solution”. If you need additional assistance, reply here within 2 business days and I’ll be happy to help.