Summary

As the amount of financial assets and information about them in the market increases, it becomes more challenging for investors and financial advisors to select relevant assets to add to financial portfolios. Financial asset recommendations alleviate this information overload by leveraging AI methods to identify a reduced set of assets of interest to the investor. In this event, researchers from the University of Glasgow will discuss how these technologies work, their current challenges and display and demonstrate recent advances on financial technologies.

This event is part of the Scottish Fintech Festival 2023.

Organizers

Dr. Javier Sanz-Cruzado Dr. Richard McCreadie
Postdoctoral researcher Senior lecturer
Information Retrieval group Information Retrieval group
School of Computing Science School of Computing Science
University of Glasgow University of Glasgow

Workshop Program

9:30am Introduction to financial asset recommendations
Dr. Richard McCreadie (University of Glasgow)
In this talk, Dr. Richard McCreadie will introduce the Financial Informatics theme at the University of Glasgow. Afterwards, he shall introduce financial asset recommendations and the motivation behind those methods.
Link to the slides
9:50am How to provide effective financial asset recommendations
Dr. Javier Sanz-Cruzado (University of Glasgow)
In this talk, Dr. Javier Sanz-Cruzado will explore the ways to train financial asset recommendations and get suggestions on investments through five different challenges to address: data collection, method selection, evaluation, time management and investment preference management.
Link to the slides
10:45am Coffee & Tea Break
11:00am Hands-on session
Dr. Javier Sanz-Cruzado (University of Glasgow)
In this session, you shall learn the insights of financial asset recommendation methods through two Python notebooks: one showcasing a profitability prediction method, and the other, a collaborative filtering method (user-based kNN) based on transactions.
Data: Link
Profitability prediction (Google Colab): Link
Collaborative filtering (Google Colab): Link
11:30am Demonstrations, poster session and networking
Dr. Javier Sanz-Cruzado (University of Glasgow), Dr. Richard McCreadie (University of Glasgow), Edward Richards (University of Glasgow), Ivan Tryskyba (University of Glasgow)
In this session, we shall make a live demonstration of a financial asset recommendation system created at the University of Glasgow as a result of several research projects (EU H2020 Infinitech, EPSRC IAA PPC-FI and EPSRC IAA FAR-AI). Afterwards, we will hold a poster session where recent advances on Fintech at the University of Glasgow shall be highlighted.
Poster list:
  • Richard McCreadie. Fintech at Glasgow. Link
  • Richard McCreadie. The Infinitech project. Link
  • Javier Sanz-Cruzado, Richard McCreadie, Craig Macdonald & Iadh Ounis. On Transaction-Based Metrics as a Proxy for Profitability of Financial Asset Recommendations. Link
  • Lubingzhi Guo, Javier Sanz-Cruzado, Richard McCreadie, Craig Macdonald & Iadh Ounis. Exploiting Knowledge Graph Embeddings for Profitability Prediction on Financial Asset Recommendations. Link
  • Edward Richards, Richard McCreadie, Javier Sanz-Cruzado, Craig Macdonald & Iadh Ounis. Comparing Time-Series Prediction Strategies for Automated Trading on Commodity Markets. Link
  • Ivan Tryskyba. Predicting Bitcoin Prices Using ML and NNs. Link
  • Javier Sanz-Cruzado, Pablo Castells. RELISON: A Framework for Link Recommendation in Social Networks. Link