Press Release: EU-Project FIRST uses Twitter & Co. for Financial Decision Making
The consortium has launched the first running prototype that can extract and analyze sentiment from social media networks in near real-time. The highly promising event detection and prediction capabilities have been enthusiastically welcomed by industry professionals in the financial domain.
In today’s world of rapidly changing and evolving financial markets, the sentiment of market participants – i.e. their feelings and opinions - can be tremendously valuable to detect and predict financial market events. The vast amount of dynamically changing, heterogeneous information makes it nearly impossible for financial market participants to manually focus on the relevant information sources. Thereby, the key challenge associated with sentiment extraction and analysis is to reliably and meaningfully interpret data sources such as social media networks, blogs, and bulletin boards in real-time.
The international project consortium of the EU-Project FIRST presented the first prototype version at the ABI Lab Conference in Milan, Italy. The attendees, financial industry professionals, highly welcomed a decision support model based on Web sentiment. Applied to the micro blogging service Twitter or textual data extracted from blogs, the prototype is presently capable of extracting sentiment and relating it to stock price movements. Results indicate a strong positive relationship between sentiment and trading volume. Such information can provide the end-user with important insights about financial market movements, and thus contribute to improved decision-making and overall financial market efficiency.
Another distinguished feature of the prototype is the ability to detect scenarios of financial market abuse, such as attempts to manipulate prices of financial instruments via false information release. This development may improve current state-of-the-art tools of market supervision and financial market regulation solutions. At a later stage, the focus of the large-scale information extraction and integration infrastructure framework will be extended to the identification and handling of reputational risks, which have proven to be a substantial threat to the stability of global financial markets.
The FIRST project started in October 2010 and will be conducted over the course of three years. The goal of the FIRST project is to use methods of artificial intelligence to support financial decision making. In a nutshell, FIRST harnesses the knowledge on the Web to support financial decision makers. Co‑funded by the European Union’s “Information and Communication Technologies” Theme of the Seventh Framework Programme, the project has an overall funding budget of nearly 4.6 million €. The FIRST consortium consists of 9 European partners from Germany, Italy, Slovenia, and Spain.
For more information and explanatory videos, please visit:
FIRST project Web site:
Videolectures.net Web site: