Intelligent business systems and modeling

We live in a world where things are digitalizing, providing an unprecedented amount of information to support better decision-making, but at the same time posing challenges for effective utilization as traditional models overlooked such unstructured information. We develop models and methods that help gain business advantages by being more intelligent!

Case: digitalization of the modern portfolio theory (MPT)

• Financial sentiment analysis for monitoring public moods and discussions.
• Personalized risk profiling using social media data.

Knowledge engineering and design science

Information and knowledge are proprietary, hence its transformation creates value and is a cornerstone of digital economy. We create more than AI models and knowledge bases --- this part largely overlaps with computer science except in a more purposely motivated and controlled way --- but also frameworks, empirical rules, management tools, and regulatory implications.

AI artifacts: NEEQ business taxonomy, STFGAN, Dual-Graph NN, FinSenticNet, ESGSenticNet etc.
Framework: HAD (how to management algorithms), RAMAS etc.
Empirical rule: mk-NN (how to build neuro-symbolic systems) etc.

Understanding human-AI interaction

As we use more AI in our daily life, two research opportunities emerge: one (human->AI) is the pressing need to shape AI according to our social norms and values; another (AI->human) is using text analytics and beyond to see and study human behavior from new angles

Cases:

• Human-like stock market prediction.
• Understanding human happiness factors.
• Social effects of AI-generated-content.