Performing Score Functions
- Former user (Deleted)
The following sections explain how to perform score functions for specific attributes when carrying out searches, and how to persist data to enable such searches to be carried out.
Persisting attributes
The attributes to be used for score functions should be persisted as index columns and score parameters at the time of persisting attributes. In the example given below, the Price and Discount attributes are persisted for the purpose of performing score functions.
Generating Data
There are no specific guidelines to follow at the time of entering data in order to allow this search operation.
Searching for data
e.g., The following cURL command outputs the result os the score function Price - Discount for all the available titles.
curl -X POST -H "Content-Type: application/json" -H "Authorization: Basic YWRtaW46YWRtaW4=" -vhttps://10.100.5.73:9443/analytics/facets -d '{"tableName" : "BookStore","fieldName" : "correlation_Title","categoryPath" : [],"query" : "*:*","scoreFunction" : "correlation_Price * ((100 - correlation_Discount ))"}' -kThis generates a response similar to the following.
{"categoryPath":[],"categoryCount":10,"categories":{"SampleTitle":6.305843089461677E-42,"Sinatra: The Life":2.802596928649634E-42,"Sherlock Holmes: The Complete Novels and Stories":2.5223372357846707E-42,"The Invisible Man: A Grotesque Romance":1.6815581571897805E-42,"The Origin of Species":1.6815581571897805E-42,"Oliver Twist":1.6815581571897805E-42,"Gone With the Wind":1.6815581571897805E-42,"A Study in Scarlet":1.401298464324817E-42,"Childhood's End":1.401298464324817E-42,"Kim":7.006492321624085E-43}}