Aarhus University Seal

Data-driven Inter-Stock Predictive Analytics (DISPA)



Due to the complex dynamics of financial markets, analyzing huge number of transactions needs more effective and systematic methods. The development of both software and hardware infrastructure has enabled the extensive collection of trade data. Therefore, there is a significant potential for machine learning (ML) solutions to make the systematic analysis possible by greatly reducing manual effort and the associated costs.  

The project DISPA will design entirely new methods which, using machine learning, to reach following goals: 

• Conduct systematic analyses of stock-exchange activity 

• Recognize irregular transactions 

• Identify patterns in share-price data.