My blog has moved! Redirecting...

You should be automatically redirected. If not, visit and update your bookmarks.

Data Mining Research - Data Analysis Trap

I'm a Data Miner Collection (T-shirts, Mugs & Mousepads)

All benefits are given to a charity association.

Monday, December 11, 2006

Data Analysis Trap

There is an interesting post on entitled Data Mining Goes Nuclear that deals with the danger of making easy conclusions on some data. In this post, the author gives a very good real-life example of easy and incorrect conclusions that can be made in the case of insurances. I think this can serve as a good example to illustrate the fact that people can easily incorrectly analyze data.

Sphere: Related Content


Will Dwinnell said...

I thought the point of the article was that the conclusions were correct and the technical analysis was valid, but the societal implications were sticky. It may very well be true that large houses incur less damage during hurricanes than small homes. Maybe they should pay less for insurance if they collect less in claims.

Sandro Saitta said...

I think you're definitely right. The conclusion they made is not incorrect, but if not socially interpreted can lead to wrong or let say strange decisions (i.e. to make people with big houses pay less than others).

Rohit said...

Somehow I think it is very reasonable to take decisions based on data, if u have the luxury to access and analyze it for your gain. No businesses are here for social service, so it makes sense for an Insurer to price the premiums in order of risk he sees in insuring someone / something. In long term everyone stands to gain from these “Right Decisions”
Rohit Sankhe

Clicky Web Analytics