DMX practice for Tourist Recommendation System

Posted by on May 18, 2015 in Business Intelligence | No Comments
Facebooktwittergoogle_plusredditpinterestlinkedintumblrmailFacebooktwittergoogle_plusredditpinterestlinkedintumblrmail

 

there is a first experience with DMX

I employed  DMX in my .NET code last year to create a travel mining model for predictions travel budget, length of travel, destination.

I need to predict via different algorithms.

so I have  added a reference to Microsoft.AnalysisServices.AdomdClient to my C# project

using Microsoft.AnalysisServices.AdomdClient;

public AdomdConnection connection = new AdomdConnection(“Data Source=****;Catalog=****”);

it was created a dynamic model..that traveller or admin of system can change the predict attributes also the algorithm type.

so user now able to create a model even for travel length, budget,activity and destination. Also there is a possibility to choose any algorithm that traveller interested

from clustering, associative rules …

codedmx1

after creating model now I can select the actual data to load them into model

 

and finally I

and finally I use the select query in DMX to show the result of prediction. in this example I employed the clustering algorithm .

dmx3

Facebooktwittergoogle_plusredditpinterestlinkedintumblrmailFacebooktwittergoogle_plusredditpinterestlinkedintumblrmail
rssyoutuberssyoutube
Leila Etaati
Dr. Leila Etaati is Principal Data Scientist, BI Consultant, and Speaker. She has over 10 years’ experience working with databases and software systems. She was involved in many large-scale projects for big sized companies. Leila has PhD of Information System department, University of Auckland, MS and BS in computer science. Leila is Microsoft Data Platform MVP.

Leave a Reply

Your email address will not be published. Required fields are marked *