Abstract
This books analyses the state of the art of applied research in a challenging field: natural language understanding of financial news. Currently, thanks to the world-wide technological spreading, stock market traders are overwhelmed with financial information, both numerical and textual, which has to be analysed quickly in order to react before market conditions change again. While there are several well-known numerical techniques for quantitative data, textual information is usually manually examined investing a lot of precious human time. This book shows how information extraction can be successfully applied to this task, at the same time speeding up the process and freeing the trader from this work load.
The book's main focus has therefore a double identity: finance, especially intra-day trading with large amounts of news arriving at a too fast pace to be examined manually, and information extraction, especially real-time analysis of predetermined events. Both sectors bring new problems and innovative techniques, which are overviewed through many examples.