Introduction to Data Mining - 2011

Data mining is not a difficult thing to do.

In fact, you can start a data mining with only one tool: Google Refine. It is very straight forward to use, and worth your time in downloading and using the software.

What do you use data mining for? It is simple. First of all, you get an insight of what customers want and you can see a rising trend before it gets very big.

Let me tell you my personal experience with data mining. I usually do text mining information from the keywords that come in from my site called www.datamine.me . It is a site in Japanese where I put together tweets about Android (from Google). There are literally thousands of queries that come in for free, and using the keywords that visitors used to search can be easily be acquired by using statcounter or Google Analytics.

What I usually do is to get a sense of how much demand there is for Android accessories, and I stock whatever I find which is in demand. The internet is a very big market, and although I mainly focus on the Japanese market, it is worth every minute of your time to data mine or text mine for valuable information.

You can sell the treasure buried and discovered by data mining to interested parties, or you can keep it for your own. One method, although not really recommended if you do not want unwanted competition, is to make the treasure public to unlimited people by putting it on the internet by blog or web page.

This will get you traffic, and a good idea is to lead the traffic to paid intelligence you can provide.

Automating data mining is a topic I want to research on the future, as it can save valuable time. For now, however, a trained data mining engineer will have a lot more power than programmed data mining software. A combination of both is probably the best idea (hybrid model).Often times, you wish to get huge amounts of data. The best idea is to start your own niche web page, and start getting traffics. Google Analytics and statcounter is good combination of tracking visitors to your site. Both have good parts which is not available in the other.

I hope I have shown you some merits of data mining. Enjoy and good luck in finding the valuable treasure buried in your database or data set.

Tsubasa Kato 9/25/2011