Package jcolibri.method.retrieve.NNretrieval.similarity.local.recommenders

Common local similarity functions for recommenders

See:
          Description

Class Summary
InrecaLessIsBetter This function returns the similarity of two numbers (or enums) following the INRECA - Less is Better formulae sim(c.a,q.a)= if(c.a < q.a) then 1 else jump * (max(a) - c.a) / (max(a) - q.a) jump and max(a) must be defined by the designer.
InrecaMoreIsBetter This function returns the similarity of two numbers (or Enums) following the INRECA - More is Better formulae sim(c.a,q.a)= if(c.a > q.a) then 1 else jump * (1- (q.a - c.a) / q.a)) jump must be defined by the designer.
McSherryLessIsBetter This function returns the similarity of two numbers following the McSherry - Less is Better formulae sim(c.a,q.a)= (max(a) - c.a) / (max(a)-min(a)) min(a) and max(a) must be defined by the designer. q.a is not taken into account.
McSherryMoreIsBetter This function returns the similarity of two numbers following the McSherry - More is Better formulae sim(c.a,q.a)= 1 - ((max(a) - c.a) / (max(a)-min(a))) min(a) and max(a) must be defined by the designer. q.a is not taken into account.
 

Package jcolibri.method.retrieve.NNretrieval.similarity.local.recommenders Description

Common local similarity functions for recommenders


GAIA - Group for Artificial Intelligence Applications
http://gaia.fdi.ucm.es