Data mining algorithms in SQL Server 2008
Microsoft SQL Server 2008 Analysis Services ships with 9 built-in algorithms that can be used in data mining solutions. Apparently these algorithms are unchanged in SQL Server 2008 R2.
- Microsoft Association Algorithm
- Microsoft Clustering Algorithm
- Microsoft Decision Trees Algorithm
- Microsoft Linear Regression Algorithm
- Microsoft Logistic Regression Algorithm
- Microsoft Naive Bayes Algorithm
- Microsoft Neural Network Algorithm
- Microsoft Sequence Clustering Algorithm
- Microsoft Time Series Algorithm
Here’s my own summary of these algorithms, based on their descriptions and detail from the Data Mining Algorithms page in SQL 2008 Books Online.
Useful for generating recommendations. The classic example is people who buy beer and shampoo.
Group similar items together.
Predicts both discrete and continuous attributes based on relationships. Example is figuring out the common characteristics of old customers that determine whether they are more likely to purchase again.
Calculates a “line of best fit” for a series of data, and then allows prediction based on that line.
Variation of Neural Network, good for yes/no outcomes.
Classification algorithm using Bayes theorem. Good for quick results that may then be refined by other algorithms.
Analyse complex relationships with lots of inputs but few outputs.
Finds most common sequences. Good for identifying popular web page site navigation trends on a website.
Predict future values of continuous values over time. Forecast next year’s sales.
Working with the cloud
So [as mentioned yesterday]/2010/05/one-laptop.html), my laptop is sitting next to me, looking very shiny and pretty, but completely silent. I spoke to Dell support this morning and after managing to exhaust the call centre guy’s script of variations on “Have you tried turning it off and on again?”, he eventually logged a case for me, and I’m now waiting for a phone call from (presumably the local) support engineer.
One thing I’m not too worried about is the data sitting on the laptop. I know it’s there – just a matter of getting a new BIOS (and/or possibly motherboard) and it will be fine. I also know I have a full backup on my Windows Home Server, which is reassuring.
One of the first things I did when I joined LobsterPot was to create a Live Mesh folder to store business-related documents. So that means I’ve still got access to those documents via the web or my other machine at home – handy as [after experimenting with various online solutions]/2010/03/time-tracking.html), I ended up reverting to using an Excel spreadsheet to track my work hours too!
I’ve also recently been making use of some of the training content at Microsoft Learning. I download the content to the laptop and have been working my way through it on the bus to and from work (must be riveting for the lucky people who get to sit next to me!). There is an offline reader just for this purpose, though it does have its annoyances (eg. no keyboard shortcuts, doesn’t allow text to be enlarged) but one nice thing is that it synchronises your progress. So I can still log into the Microsoft E-Learning site and am able to pick up where I left off, using the “online” mode instead.
Then there’s email, IM etc, which all work fine from where ever you are. So, while having a dead laptop is a huge pain, it isn’t the end of the world.
Good news.. I just had a call from Dell and they’re coming out on Monday. Hope they can fix it quickly.
One 'bricked' laptop
This evening I thought I’d install the most recent firmware update A09 for my new Dell XPS 1645 laptop. I downloaded the installer and ran it, and it automatically started the update.
Then it got stuck.
I left it for a fair while, but the progress bar was not moving, and then I discovered the mouse and keyboard were also not responding. Holding down the power-button had no effect, so the only way to power off was to remove the battery.
A call to Dell support revealed that as it is a hardware problem I need to talk to their hardware support line which is only open 8am – 8pm (and it was already past 8pm).
So tomorrow morning I’ll be on the phone to see what can be done. Very frustrating.