+

Search Tips   |   Advanced Search

Ratability parameters

The buildvisit background utility computes a ratability value for each item in the database. It is only used for the Preference Engine. The LikeMinds server presents items for rating in decreasing order of ratability; items with the same ratability are presented in random order.

The ratability value is computed as

ratability = Ai+Bi+Ci

where Ai is the popularity of item i:

Ai=Kx(log10(num_rating))rpow

num_rating is the number of ratings for the item.

To specify K, set db.ratability.num_rating.coefficient (default is 1.0):

db.ratability.num_rating.coefficient = 1.0

To specify rpow, the rating power, set db.ratability.num_rating.power (default is 1.0):

db.ratability.num_rating.power= 1.0

Bi is a measure of how controversial the item is:

Bi=Kx(rating_std_dev)sdpow

rating_std_dev is the standard deviation of the ratings for the item.

To specify K, set db.ratability.stddev.coefficient (default is 1.0):

db.ratability.stddev.coefficient = 1.0

To specify stdpow, the standard deviation power, set db.ratability.stddev.power (default is 1.0):

db.ratability.stddev.power= 1.0

Ciis a weight based on the age of the item:

Ci=Kxexp(-(current_year - release_year)) exp ^ apow

release_year is the value of the year field in the Lps_Item_Data table for the item.

To specify K, set db.ratability.age.coefficient. We can set this to zero if you do not want to consider age in determining ratability (default is 2.7):

db.ratability.age.coefficient = 2.7

To specify apow, set db.ratability.age.power (default is 0.5):

db.ratability.age.power= 0.5


Parent: Configure LikeMinds utilities