Bloomberg curates Twitter feed for algo traders; State Street reads the news

Bloomberg curates Twitter feed for algo traders; State Street reads the news

Bloomberg is launching a real-time feed of curated Twitter data, sifting through the deluge of social media sentiment to pick out the most useful content for its algorithmic trading clients.

Built on the back of Bloomberg’s natural language processing (NLP) techniques and available through the company’s Event-Driven Feeds (EDF) product, the data promises to allow financial firms to extract value by making sense of the over 500 million tweets per day.

Tony McManus, Bloomberg Enterprise Data CIO, says: "Our customers tell us that Twitter data is a vital part of their information-driven trading strategies, helping them uncover early trends and changes in sentiment.

"Our Twitter EDF feed will help quantitative traders to capitalise on Twitter’s influence on the markets through constantly evolving curation methodologies. These include proprietary NLP modeling, coupled with Bloomberg’s reputation for data quality and the expertise of a world-class news organisation."

Separately, State Street has launched a mobile application that uses big data, machine learning and NLP to help investors assess their portfolio exposure to breaking news.

The tool, called Verus, gathers coverage from thousands of major global, English-language news publications and combines machine learning algorithms with portfolio data from State Street’s end-to-end risk analytics platform to curate users’ newsfeeds.

Humans also play a part, with Verus incorporating the insights of a dedicated editorial team comprising former financial news editors and journalists continuously providing feedback to the algorithms so that the connections made are relevant and that the underlying algorithms are continually improving.

Once direct connections are identified, Verus analyses third-party relationship data in order to surface relevant indirect relationships within the portfolio. Connections are then ranked by a 'V Score' based on three main factors: article content, risk, and network connections. The V Score tells users whether they should care about the news and how much.

Stephen Marshall, head, State Street Verus, says: "About 150,000 articles are generated on business and finance topics each week, and investment and risk professionals struggle to reliably filter out noise and feel confident that they are not missing anything that could impact their portfolio.

"Our new solution highlights not only the direct connections but also the less obvious indirect connections between these news items and clients’ portfolios; shortening the amount of time between when a relevant news story breaks and when a client can comfortably answer the question, ‘What is my exposure?’"

Comments: (0)


Featured Job
All Jobs »