cross_reference_connection_insights
The job allows you to find non-obvious connections between the contents in different documents based on the shared references of the documents.
You must first create an archive and index the internal pieces of the various documents. If using Text index paragraphs, sentences or words, if Images, Sounds, Videos or Point Clouds index those individual pieces. Keep track of which document the indexed content ids belong to.
You must supply the name of the archive, the dictionary that connects document id to content ids and the list of links between documents.
The job will return a list of insights, each insight contains the ids of the concerned documents and the content ids that form a candidate connection.
Required Account Privileges: "read"
Request JSON ["inputs"]:
"archive":
string (3 <= len <= 30) unique in account
null NOT allowed
A unique string identifier for the archive within your account.
"document_id_to_content_ids":
dict of int to list of ints
null NOT allowed
A dictionary where each key is an integer representing a document ID, and the corresponding value is a list of integers representing content IDs associated with that document.
"document_id_links":
list of lists of ints
null NOT allowed
A list of lists of integers representing the documents that are related.Response JSON ["results"]
"connection_insights":
list of dicts:
{
"source_document_ids": list of ints,
"candidate_connection_content_ids": list of ints
}