Saturday, February 9, 2019

Marquette’s FinTech Topics Course Week 4 Update: Introduction to Bloomberg Big Data Analytics Functions


Joe Furey, Bloomberg Consultant, visited Dr. Krause’s FinTech Topics course and the students learned about the powerful news and social media data scrapping tools within Bloomberg


On Wednesday, February 6, 2019, Dr. David Krause (AIM Program Director) and the students in the FinTech Topics course met with Joe Furey of Bloomberg.

Dr. Krause said, “With the increased news and social media volume and time sensitivity of understanding the implications of breaking media stories and posts, as financial analysts it is necessary to access actionable information quickly. Joe showed us various ways to access data on the Bloomberg and to automate the processing of extracting usable information.”

Using blogs, tweets and other unstructured textual information presents challenges that are can be best handled by machine-learning techniques. The students learned how Bloomberg has applied such techniques to identify a news story or tweet as being relevant for an individual stock ticker and to assign a sentiment score to each story or tweet in the feed.

The students will soon be receiving an assignment where they will access Bloomberg and utilize the news and social sentiment data from historical and real-time news flows. Mr. Furey explained Bloomberg’s process for scrapping and amassing data - and how their models automatically assign a probability of the news sentiment being positive, negative or neutral to each news story or tweet.

Bloomberg provides two types of sentiment analytics: story-level sentiment and company-level sentiment. The students had an opportunity to ‘test drive’ Bloomberg on specific company tickers and to view how the stock’s performance appears to follow the breaking of stories on blogs or social media.

 It will be interesting to see what the students learn from their assignments about the predictive power of big data and its relationship to future stock performance.