Financial Events Recognition in Web News for Algorithmic Trading

Due to its high productivity at relatively low costs, algorithmic trading has become increasingly popular over the last few years. As news  can  improve  the  returns  generated  by algorithmic  trading,  there is a growing need to use online news information in algorithmic trading in order to react real-time to market events. The biggest challenge is to automate the recognition of financial events from Web news items as an important input next to stock prices for algorithmic trading. In this position paper, we propose a multidisciplinary approach to financial events recognition in news for algorithmic trading called FERNAT, using techniques from finance, text mining, artificial intelligence, and the Semantic Web.

If anyone want to port the code to MT4/MT5, here is the open source project : (https://code.google.com/p/umat/).

More information for Data Mining and News Trading Algorithms : (http://www.aphysicistinwallstreet.com/2010/11/example-rapidminer-r-for-trading.html)