March 20, 2013 at 3:00 pm
hehehe.
Yeah. It is a good question.
The main idea about data mining is that it obtains non intuitive results.
Using simple queries and percentages it can take a long time to find patterns.
In other words, you can do it, but data mining is a more sophisticated and specialiced tool to find customers patterns.
March 20, 2013 at 3:52 pm
Thanks for the quick reply 🙂
Looking forward to going through your parts 2 and 3 now!
July 17, 2013 at 11:33 am
Nice job with clear demonstration, Daniel!
Just a little thing missing:
For part 1 about data mining, In the Specify Columns Content and Data Type(which is step 10 for Data Mining Model), press Detect and press Next.
Age there was defined as Continuous Type. While in part 4 for Microsoft Naïve Bayes algorithm, you did not mention to change age to discrete Type. Otherwise, age will be ignored.
October 1, 2013 at 1:59 pm
Nice article - a step stone leading to data mining.
One thing that troubled me the most is when you look for BI jobs, many of the descriptions do not even have the word 'statistics'. May be the hiring managers think you can just use sql to mine the data? Apparently not!
February 4, 2014 at 11:19 am
Daniel, thanks for adding!
The more you are prepared, the less you need it.
May 27, 2014 at 2:03 am
Nice article, but you're not exactly predicting the future!! That's misleading. You are arriving at the probability that someone is more likely than not to be a bike buyer. That is merely based on past purchases and, as anyone will tell you, the past is no guarantor of future certainty. That's the problem with off-the-shelf data mining tools. As soon as you start to use software like this to attempt to increase sales (for example), you'll probably be hard pressed to beat 50% accuracy (surprise surprise). There are maths kids out there who write there own algos and who can get great results, but black-box software such as this is unlikely to make you rich quick!
May 27, 2014 at 2:04 am
I was really interested though in what tool you were using to create your captions! They look nice!
May 27, 2014 at 4:48 am
Camtasia
July 9, 2014 at 1:49 am
Shark Energy (3/20/2013)
Hi,Long time Senior DBA here - but just starting down the Data mining journey. Great article - makes it seem you can get something good out with minimal knowledge (but clearly the more you learn the better your output!)
Data Mining excites me, as I think we can change the way we do business. BUT in this example - in a strange way I see it as a more convenient way of avoiding a basic query. I.e. If I have one Customer table in my database, why would I not just stick a join on the products table - and then query for buyers of a product where age between x and x, town = x, salary > x and so forth, and then take that as a percentage of the whole sales for that product. (Thats how it works out the percentage?)
Don't get me wrong - but please put me right 🙂
I just arrived to this ...
I would be interested to compare the two ways basic query vs mining tool.
That would a good opportunity to get a better understanding of the tool.
Thank you Daniel for putting this very nice article together.
Iulian
July 15, 2014 at 3:53 pm
Hi Sugzy,
In menu bar go to project option, last option -"your project" properties-appear window of properties, on left treeview select deployment, on right pane change Server option to your AS instance.
July 15, 2014 at 4:47 pm
Hi Daniel,
I got a problem with point 15-process mining Model, I started process and after a few seconds i got this messages :
Processing Mining Structure 'DTStructure' completed.Start time: 15/07/2014 05:35:59 p.m.; End time: 15/07/2014 05:35:59 p.m.; Duration: 0:00:00
Processing Dimension 'DTStructure ~MC-Customer Key' completed.
Start time: 15/07/2014 05:35:59 p.m.; End time: 15/07/2014 05:35:59 p.m.; Duration: 0:00:00
Processing Dimension Attribute '(All)' completed. Start time: 15/07/2014 05:35:59 p.m.; End time: 15/07/2014 05:35:59 p.m.; Duration: 0:00:00
Processing Dimension Attribute 'Address Line2' completed.
Start time: 15/07/2014 05:35:59 p.m.; End time: 15/07/2014 05:35:59 p.m.; Duration: 0:00:00
Processing Dimension Attribute 'Bike Buyer' completed.
Start time: 15/07/2014 05:35:59 p.m.; End time: 15/07/2014 05:35:59 p.m.; Duration: 0:00:00
Errors and Warnings from Response
Internal error: The operation terminated unsuccessfully.
Server: The current operation was cancelled because another operation in the transaction failed.
OLE DB error: OLE DB or ODBC error.
Errors in the high-level relational engine. A connection could not be made to the data source with the DataSourceID of 'datamining', Name of 'datamining'.
Errors in the OLAP storage engine: An error occurred while the dimension, with the ID of 'DTStructure ~MC-Customer Key', Name of 'DTStructure ~MC-Customer Key' was being processed.
Errors in the OLAP storage engine: An error occurred while the 'Address Line2' attribute of the 'DTStructure ~MC-Customer Key' dimension from the 'DataMining' database was being processed.
Thanks in advance.
September 26, 2014 at 7:12 am
Hi Daniel,
Great article! I hope to follow the rest of the series; but there is a single problem before moving on. In following this introductory article (several times), in the first query, the Expression result always appears with only 0, no 1. Apparently I am making some error, but after repeating the article several times, setting up a new a new instance of SQL Server 2008 R2, to repeat the article; I must be repeating the mistake, please advise. I added an attachment showing the issue.
Thanks,
Steven
May 27, 2015 at 1:19 pm
Thanks for the article and a good place for me to start learning about this.
May 27, 2015 at 1:33 pm
Check your connection properties in the Data Source.
Make sure to connect with a user with permission to the Adventureworks database.
May 27, 2015 at 1:34 pm
Hello, if you change your query values does the query changes ?
Viewing 15 posts - 31 through 45 (of 46 total)
You must be logged in to reply to this topic. Login to reply