Mathematical Problems in Engineering / 2014 / Article / Alg 2

Research Article

Predicting Click-Through Rates of New Advertisements Based on the Bayesian Network

Algorithm 2

KBN inference.
Input:   , the DAG of the KBN;
   : the evidence nodes in (the new ad’s keywords);
: the non-evidence nodes in (all keywords except for the new ad’s keywords);
   : the set of values of the evidence (set 1 as the value);
   : the query variable (the keyword will be inferred);
   : the total times of sampling.
Output: The estimates of
Variables:
: the set of the values of the non-evidence;
;
: the current state of the KBN;
: the number of the samples when the value of is 1
Steps:
    (1) Initialization:
values of in Z
    (2) For to Do
 (i) Compute the probabilities of the selected variables based on the state ,
    Randomly select one non-evidence variable from :
    , where is the set of the values in the
    Markov chain of in the current state
  (ii) Generate a random number and we determine the value of :
     (5)
    
  (iii) Count
     If Then
  End For
    (3) Estimate