There are many sites trying to explain the SVM algorithm, but I find the following one the shortest and the easiest to understand without using a lot of equations: Link to Gist SVM Server
Problem #1: Automatically classify a stock as a buy, hold or sell (or strong vs. weak) based on a variety of fundamental and technical information.
Problem #2: Automatically classify a length of rock formation for hundreds of wells as sandstone, shale, and other lithofacies based on a variety of geological and log data.
In statistical machine learning - a branch of artificial intelligence, the two classification problems are exactly the same.
The free Java applet can generate a graphical view ("graph") of the hirachieal tag structure of a HTML web page. For example, the graph for this blog site (a WordPress blog) can be found at:
A seven-part series on economics of virtual trading, which happens to be popular among gamers in China also. Will read later.
Online fantasy games have booming economies and citizens who love their political systems. Are these virtual worlds the best place to study the real one?
Quote from the data mining blog:
I've added a new image of the blogosphere to the gallery. It is repeated below. The image is a hyperbolic projection of a graph of the largest 8 partitions (connected components) - the data is a combination of that from WWE2006 and from our upcoming ICWSM2007. The visualization is centered on DailyKos. The highly connected area below centre is the socio/political community. Above and to the right is the heart of the technical blogosphere with BoingBoing being the brightest node.
