Personal Recommendation Using a Novel Collaborative Filtering Algorithm in Customer Relationship Management

Pseudocode 1

Algorithm CF-M: Calculating the similarity between users

begin

 get 𝐴;

 𝑛 = size(𝐴,2), 𝑚 = size(𝐴,1);

 parameter 𝑎𝑟;

 preference degree 𝑣();

 range of the rating score 𝑀;

 𝑆 = 𝑧𝑒𝑟𝑜𝑠(𝑚,𝑚), 𝑜 = sum(𝐴), 𝑢 = sum(𝐴^′);

 for 𝑖=1∶𝑚

  for 𝑗=1∶𝑚

   𝑥=(𝑢(𝑖)∗𝑢(𝑗))∧(−0.5);

   for 𝑧=1∶𝑛

    𝑦=𝑦+𝑎(𝑖,𝑧)∗𝑎(𝑗,𝑧)∗((1−abs(𝑣(𝑖,𝑧)−𝑣(𝑗,𝑧))/𝑀)/𝑜(𝑧))∧𝑎𝑟,

   end

   𝑆(𝑖,𝑗)=𝑥∗𝑦;

   𝑥,𝑦=0;

  end

 end

End