Prediction quality values

Portal, Express Beta Version 6.1
Operating systems: i5/OS, Linux,Windows

Prediction quality values

The "quality" value refers to the degree a user will like an item. The LikeMinds server presents predictions in decreasing order of quality. Recommendation quality is calculated using this equation:

quality = confidence^mpowx (pred rating —mean pred rating) + K x confidence^pow

where:

confidence is the confidence value for the recommended item.

pred rating is the predicted rating for the item.

mean pred rating is the mean recommended rating for all items for the user.

To specify ^mpow, set db.applic.param.BestBets.MultPower:

db.applic.param.BestBets.MultPower = 0.5

To specify K, set db.applic.param.BestBests.Coefficient:

db.applic.param.BestBets.Coefficient = 0

To specify pow, set db.applic.param.BestBets.Power:

db.applic.param.BestBets.Power = 1.0

Note: You can use these parameters for all of the Recommendation engines except for the Item Affinity Engine. Parent topic: Specifying recommendation behavior

Library | Support | Terms of use |