Geek Answers Handbook

The recommendation algorithm used on the e-commerce site and solved using the neo4j chart database

I plan to implement a product recommendation on my e-commerce site using the neo4j chart database.

The recommendation will be based on user action on the product. Actions will be

   - Product View , 
   - Rating ,
   - Read book
   - Download book , 
   - Purchase , 
   - Add to card , 
   - Review , 
   - Share
   - Some more action applicable to our site.

Chart structure will be

User (Node)

  • ID
  • Mark

Product (Node)

  • Name
  • Mark

Action (relationship between user and Node product)

  • Weight (indicated on the basis of the action, for example: purchase: 10, view: 1, etc.)
  • Timestamp (time at which the action occurred)

Later I will add social relationships between user nodes.

. , . ( , ).

 - Item-Item similarity   
       - k-nearest neighbors (k-NN) algorithm
       - Pearson correlation coefficient.   
 -  User-User similarity   
 -  Matrix Factorization    
       - Singular Value Decomposition (SVD)
       - Restricted Boltzmann Machines (RBM)
       - Non-Negative Matrix Factorization ( NNMF )
 -  Latent factor analysis   
 -  Co-visitation analysis   
 -  Latent topic analysis   
 -  Cluster model   
 -  Association rule    
       - Bi-gram matrix association rule
 -  Ensembles

, , , neo4j ( ).

+4

More articles:


All Articles