PhD Defence: News Analytics for Financial Decision Support

News Analytics for Financial Decision Support

The newly emerged, growing body of scientific work on the use of news analytics in finance addresses the issue of designing automated financial trading systems that are able to incorporate information from news in trading strategies. News analytics extend trading algorithms by including information extracted from textual messages, and translating it into actionable, valuable knowledge.

In his PhD dissertation entitled <link doctoral-programme phd-in-management phd-projects detail>News Analytics for Financial Decision Support, Viorel Milea focuses on news-based trading using an interdisciplinary approach. His dissertation presents three systems that are able to use news for financial trading. The first system deals with structured text, the monthly statements of the European Central Bank. Another system introduces a framework for using information from news messages, in the form of events, for share trading. Finally, Milea introduces a more complex system, able to deal with time and time-related information as extracted from news. This system stands at the foundation of the time-aware framework for financial decision support.

A considerable part of Milea’s dissertation deals with developing technology that is necessary for news analytics in finance. Milea presents a temporal web ontology language, which is proposed as a building block of all systems that deal with temporal knowledge. The language is able to represent abstract domain knowledge as well as more concrete (temporal) facts, such as the information contained in news messages.

Viorel Milea defended his dissertation on 7 February 2013. His supervisor was Professor Uzay Kaymak. His co-supervisor was <link people flavius-frasincar>Flavius Frasincar. Other members of the Doctoral Committee are <link people rommert-dekker>Professor Rommert Dekker (Erasmus University Rotterdam), Professor Geert-Jan Houben (Delft University of Technology), and Professor Witold Abramowicz (Poznan University of Economics).

About Viorel Milea

Viorel Milea was born on April 27th, 1982 in Bucharest, Romania. After finishing high school in Romania, he moved to the Netherlands for his undergraduate studies. He obtained his MSc degree in Informatics & Economics from Erasmus University Rotterdam, the Netherlands, in 2006.

Afterwards, he joined the Econometric Institute for his PhD research within the TOWL project, a project co-financed by the European Union. Currently, he is an Assistant Professor at the Econometric Institute of the Erasmus School of Economics at Erasmus University Rotterdam. 

His research interests cover the use of Semantic Web technologies for enhancing automated trading, Semantic Web theory, management information systems, content analysis, and nature-inspired classification and optimisation techniques.

Abstract of News Analytics for Financial Decision Support

This PhD thesis contributes to the newly emerged, growing body of scientific work on the use of News Analytics in Finance. Regarded as the next significant development in Automated Trading, News Analytics extends trading algorithms to incorporate information extracted from textual messages, by translating it into actionable, valuable knowledge.

The added-value of this dissertation can be divided roughly in two contributions. On one side, we contribute to the design and implementation of systems that use news for trading. On the other side we deliver more theoretical contributions related to the technology that is required for such systems. 

The dissertation presents three systems that are able to use news for financial trading. The first system we present deals with structured text, in this case the monthly statements of the European Central Bank. Another system that we present introduces a framework for using information from news messages, in the form of events, for share trading. Finally, we introduce a more complex system, able to deal with time and time-related information as extracted from news. This system stands at the foundation of the time-aware framework for financial decision support that we introduce.

A considerable part of this dissertation deals with developing technology that we deem necessary for News Analytics in Finance. Here, we present a temporal web ontology language. We envision this language as a building block of all systems that deal with temporal knowledge. The language is able to represent abstract domain knowledge as well as more concrete (temporal) facts, such as the information contained in news messages.

Finally, despite its high relevance for business and the “next big thing in automated trading” status, News Analytics, especially in a financial context, is still in its academic infancy. Next to the concrete contributions of this dissertation, our work also delivers a proposal. Our discussion identifies the disciplines that can contribute to this emerging field. Finally, we provide possible focus points for future research.