Applying big data analytics and machine learning in Fintech
The FinTech revolution refers to the application of many new concepts and ideas that are reshaping the financial industry such as technology-driven transformation (e.g. banking value networks), increased automation (e.g. algorithmic trading, robotic advice), emerging digital and virtual financial markets (e.g. virtual currencies, crowdfunding) and big data analytics (e.g. novel analytics approaches to risk modelling). In this project, we are specifically focusing on how AI and machine learning techniques are creating opportunities to rethink the way enterprises are making decisions based on processing large amounts of data using AI and machine learning techniques.
In particular, existing decision-making and information management practices tend to be qualitative and have shown many limitations when confronted to some new challenges such as those arising from the adoption of ESG (Environment, Social, and Governance) regulations and guidelines worldwide., leading to high costs and inaccuracies. This project will explore many opportunities to exploit the power of AI and big data analytics to address these challenges and create new ways to support new and emerging FinTech applications. In particular, the area of ESG is forcing many companies to investigate adequate information processing infrastructures that can assist them in supporting their ESG goals and meet their compliance requirements.
This project will bring innovation by integrating many established analytics techniques e.g.:
Text processing and information mining from enterprise and public repositories
Event analytics and event processing technologies
Real-time machine learning processing that includes IoT data
As a case study, this research project will aim at supporting organisations in meeting their ESG goals from a quantitative rather than a qualitative perspective.
1. Investigate novel FinTech applications that use machine learning techniques particularly in the area of ESG.
2. Develop new data models for creating machine learning data sets that integrate both structured and unstructured data.
3. Combining machine learning and event-based analytical models to address the target application requirements.
Fethi Rabhi, Mingqi Yu, Alan Hsiao