Need help regarding implementing SSAS solution using microsoft decision trees algorithm

  • Hi All,

    I am new to SSAS and Data Mining techniques. I dnt have a good knowledge about data mining in SSAS. I have a requirement regarding predictive analysis and want to check whether i can implement SSAS for it.

    I have two tables namely Tree and Node. The Tree table has the columns namely NodeID, ParentID and Level. The ParentID in the tree table represents a node in the node table and at the same time the nodeid in the tree table also represents the nodeid in the node table.

    If you run the below query, it will list out all the possibilities of getting the nodename 'CTRYIMPACT - High - PRODIMPACT - Low - Question 6'

    select p.*

    from Node n, Tree t, Node p

    where n.NodeName='CTRYIMPACT - High - PRODIMPACT - Low - Question 6'

    and n.NodeId = t.NodeId

    and t.ParentId = p.NodeId

    I want to predict the various nodenames that are possible by inputting a NodeID using analysis services. If this is done, then i can implement it in my project.

    I went over this link to know about implementing decision tress but, they have used only a single table in the example. Since I have two tables i am struck in the middle. And also they have not specified clearly of what they need to achieve. They have just stated that on changing the play the value changes.

    http://www.mssqltips.com/sqlservertip/2965/classic-machine-learning-example-in-sql-server-analysis-services/

    Can anyone help me with this?

    I have attached the spreadsheet containing the values for the Node and Tree table.

Viewing 0 posts

You must be logged in to reply to this topic. Login to reply