.NET Tutorials, Forums, Interview Questions And Answers
Welcome :Guest
Sign In
Register
 
Win Surprise Gifts!!!
Congratulations!!!


Top 5 Contributors of the Month
Kaviya Balasubramanian
Sgraph Infotech
Imran Ghani
Post New Web Links

Cannot get recommendations after querying an association rules model for multiple customers

Posted By:      Posted Date: May 22, 2011    Points: 0   Category :
 
To carry out this data mining task, I am using the Data Mining(DM) add-in of Excel 2007, 
where one has to make use of SQL Analysis Server in order to utilize the add-in.
Basically, I have a three-column dataset: CustomerID,Product,Transaction. These are the steps I am following:
1. Build an Association Rules(AR) Model.
2. Query the AR Model with new data and obtain a recommended product.
Right now, if I supply an input of Products "A","B","C" for "Customer1" I get the recommendation e.g. product "D".
This is the DMX query I am using:
##############################################################################################################
SELECT
PredictAssociation([Associate ProductNo].[ProductNo_Table],INCLUDE_STATISTICS,5) AS [Output 1]  
FROM
[Associate ItemNo]
NATURAL PREDICTION JOIN SHAPE
{@InputRowset } APPEND ( { @InputRowset1 } RELATE [SessionNo] TO [SessionNo] ) AS [ProductNo_Table] AS
InputRowset = Microsoft.SqlServer.DataMining.Office.Excel.ExcelDataReader
InputRowset1 = Microsoft.SqlServer.Data


View Complete Post


More Related Resource Links

Extracting Association Rules from Data Mining Model

  
Hi I have succesfully created a Data Mining Model using the Association Model. I have deployed and processed it. I need to extract all the rules that is generated by the model. I know that you normally query the model using DMX prediction queries, but in this case I need to extract the rules to a separate table for further processing. I have tried the following approaches unsuccessfully: 1. Linked server in MSSQL Management Studio. DMX query using OpenQuery. The DMX query looks like this: SELECT FLATTENED NODE_CAPTION, NODE_SUPPORT, NODE_PROBABILITY, MSOLAP_NODE_SCORE FROM DataMiningModel.CONTENT WHERE NODE_TYPE = 8   On small models this method works. On larger models with many rules I receive an exception: "XML for Analysis parser: The XML for Analysis request timed out" before it was completed This always happens after 70min. I might have missed a timeout option? 2. Using SSIS to run the same DMX query as above. The method is presented here: http://www.sqlservercentral.com/articles/MDX/64697/ This method works on small models. On larger models that did not work with method 1, it return rules. The number of rules returned is sometimes different for the same package run multiple times. For the largest models it simply returns 0 rules. I suspect that the same XML Parser error happens under the hood of SSIS. I'm currently stuck, and need some inp

Association rules dimension error caused by setting the model algorithm

  

Hi,

I got this error meassage when I wanted to process a mined cube:

"Error (Data mining): An error occurred while the 'Bon_DMDim' data mining dimension with the 'Bon' source mining model was being processed. The algorithm of the source model returned data mining dimension content information that is not valid."

Before processing I modified only one algorithm parameter in the Association Mining model properties (AlgorithmParameters):

MINIMUM_SUPPORT = 10 (Default 0.0)

I need to have this adjustment for getting rules...

If I set back this field to 0 or empty, then the processing ends up with no error messages (and I would not have mined rules...).

I also get this message, if I use numbers between 0 and 1 (as a percentage).

How can I build a usable (but with adjusted algorithm parameter) mined dimension and apply it in an OLAP cube?

Thank you and BR,

Peter

 


How Does the Association Model Predict the Products for Individual Customers?

  

Dear All,

I have built an Association Model , now i am applying this model to the same data on which the model is built.

For many of the customers there is no products getting suggested , for few customers products are getting suggested, Please let me know how the model is suggesting products for few and nothing for most of the customers.

 

Thanks in Advance


How to get all the rules of an association model using "CALL System.Microsoft.AnalysisServices.Syste

  

Dear All,

I am using sql server analysis services to build an association model, once the model is built i am using "CALL System.Microsoft.AnalysisServices.System.DataMining.AssociationRules.GetRules"  procedure to get support,importance,probability for a rule  by passing the name of a model (1 st parametre), but this procedure can retreive only max of 2000 rules though there are more than this depending on the sorting condition( 4 th parametre) specified in the above procedure.

Is it possible to get all the rules of a model using the above procedure?

If not pls suggest some alternative to get all rules of a model with support,probability and importance.

Thanks in Advance.


How to get all the rules of an association model using "CALL System.Microsoft.AnalysisServices.Syste

  

Dear All,

I am using sql server analysis services to build an association model, once the model is built i am using "CALL System.Microsoft.AnalysisServices.System.DataMining.AssociationRules.GetRules"  procedure to get support,importance,probability for a rule  by passing the name of a model (1 st parametre), but this procedure can retreive only max of 2000 rules though there are more than this depending on the sorting condition( 4 th parametre) specified in the above procedure.

Is it possible to get all the rules of a model using the above procedure?

If not pls suggest some alternative to get all rules of a model with support,probability and importance.

Thanks in Advance.


C++ Rules: Power Your App with the Programming Model and Compiler Optimizations of Visual C++

  

Many programmers think that C++ gets good performance because it generates native code, but even if your code is completely managed you'll still get superior performance. In Visual Studio 2005, the C++ syntax itself has been greatly improved to make it faster to write. In addition, a flexible language framework is provided for interacting with the common language runtime (CLR) to write high-performance programs. Read about it here.

Kang Su Gatlin

MSDN Magazine January 2005


Multiple forms on the same page MVC2: how to handle (including validation from model)

  

Hi,


It should be simple; although I'm having a hard time to figure out the best way to do it. I've MVC2 pages with multiple forms on the same page having different functionalities.


What's the best way to handle this; including the MVC2 way of validation from the model?


Thanks.


Custom List Row association to multiple documents - Design choices?

  
I want to associate custom list entries to muliple documents. Thinking about options.   1 row will have many docs, but one doc will only have one row. Possible? Trying to keep it simple with minimal development. I was thinking I could have a unique column or the id column in the list act like a key (somehow).. and  add a required column to the document list requiring it to tie back to the list entry.  Then a view could drive listing all the docs for a particular row from a link in the list built in a custom dataview. All the docs could be in the same library. Is the the best approach? Ideas/suggestions? Thanks.

Problems with multiple validation rules

  

Hi!

I've added some custom validation rules programatically like this:

 

 

            BindingOperations.GetBinding(txtTitulo, TextBox.TextProperty).ValidationRules.Clear();

            BindingOperations.GetBinding(txtTitulo, TextBox.TextProperty).ValidationRules.Add(new ReglaDatoRequerido(Idioma.RecursoIdiomaLocal(this, "msjErrorDatoRequerido")));

            BindingOperations.GetBinding(txtTitulo, TextBox.TextProperty).ValidationRules.Add(new ReglaRangoDatos(Idioma.RecursoIdiomaLocal(this, "msjErrorDatoRequerido"), 1, 5, false));


these are the rules:
public abstract class ReglaValidacionBase : ValidationRule
    {
        /// <summary>
        /// Mensaje de error.
        /// </summary>
        public string MensajeError { get; set; }
    }

   public class ReglaDatoRequerido : ReglaValidacionBase
    {
        public ReglaDatoRequerido(st

Error While Applying a Data on Association Model Built

  

Dear All,

I  have built an association model containing arnd 800 transactions, now i am trying to apply this model on data containing 200 transactions (here each transaction can contain huge products which might have not been present while building the model)   using the query which i get from mining model prediction but i am getting the following error while parsing in sql server management studio as well in  mining model prediction (when i click on the result)

Executing the query ...

Parser: The end of the input was reached.

Execution complete


Issue with Report Model on a Cube (multiple measure groups)

  

Hi All, 

I have was asked to built a report model on the cube for some users and being new to report builder 3.0. 

I am having a weird issue. Lets Say I have following DIMS and FACTS. I have a cube built on top this Dimensional model. 

Two measure Groups/mearsures  : ClassFact/ClassHours and FacilitatorFact/FacilitatorHours.

used Many to Many relationship for FacilitatorDIM and ClassFact measuregoup. I now browse the cube it works excellent. 

The problem Starts

Now a report model is built on top of it, then in the query wizard when I pull both groups and then FName the Classhours(measure) Disappears. 

Does Report Modes supports Many to Many or am I missing something? Any help will be appreciated. 

Thanks

 

Dims

ClassDIM (ClassID, ClassName) 

FacilitatorDIM ( FID, FName) 

Facts

ClassFact( ClassID, Classhours) 

FacilitatorFACT(FID, ClassID, FacilitatorHours)

 

Sample Data: 

 

 

SELECT     ClassID = 1 , ClassDurationHours = 100.0

FROM       ClassFact

UNION 

SELECT     ClassID = 2 , ClassDurationHour

Multiple database support with Entity Framework

  
One of the features introduced in Entity Framework is being database independent. Which mean each database provider can support Entity Framework by implementing its provider.

This feature allows you build applications independent from the underplaying database provider. In this post I'm going to show how to build an application that support multiple databases using same conceptual model made by Entity Framework.

Binding WPF Controls to an Entity Data Model

  
In this interview, programming writer, McLean Schofield, demonstrates how to bind WPF controls to an entity data model, using Visual Studio 2010 Beta 1. You can also learn more in the topic: Walkthrough: Binding WPF Controls to an Entity Data Model.

Binding the multiple drop down boxes in GridView using JQuery in ASP.NET

  
Here is the code for binding the multiple drop down boxes in GridView using JQuery in ASP.NET

Upload Multiple File using JQuery in asp.net

  
In this article I have explained how to upload multiple files using file upload control. I have used jQuery plugin for uploading multiple files.

I have also explained how to check for file type, file maximum size limit using jQuery & also using server side validation code.

Model View Controller (MVC) for ASP.NET

  
Model View Controller (MVC) architectures, immediately wonder how the MVC pattern can be applied in an ASP.NET world. This article will demonstrate how the MVC pattern can be applied to an ASP.NET web application and still maintain the basic Microsoft supplied architecture.

Adding Multiple Columns and Rows in GridView without using a Database

  
This article describes on how to add multiple columns and rows in GridView without using a database. Basically the GridView will be populated with data based on the values entered in each TextBoxes on Button Click and retain the GridView data on post back.
Categories: 
ASP.NetWindows Application  .NET Framework  C#  VB.Net  ADO.Net  
Sql Server  SharePoint  Silverlight  Others  All   

Hall of Fame    Twitter   Terms of Service    Privacy Policy    Contact Us    Archives   Tell A Friend