Minimum mentors the engine examines for predictability

When it determines whether it can make recommendations for a given user, the Preference Engine must examine the user's mentors. <eng_instance_name>.engine.titan.predictable.loop_check_cutoff specifies the number of mentors the Preference Engine will examine before it checks to see whether the user is predictable. The default is 10.

For example:

movie_pref.engine.tiuan.predictable.loop_check_cutoff = 12

If the user is not predictable, the engine will examine another group of <eng_instance_name>.engine.titan.predictable.loop_check_cutoff mentors, and then check to see whether the user is predictable. It will continue until it finds that the user is predictable or it runs out of mentors.

Parent topic: User predictability

Parent topic: User predictability

Related reference
Minimum number of Clickstream activities for a user
Recomputing Clickstream Engine predictions
Default Clickstream Engine recommendations


Home |