Singapore University of Social Sciences

Machine Learning and AI for FinTech

Machine Learning and AI for FinTech (FIN313)

Applications Open: 01 October 2024

Applications Close: 15 November 2024

Next Available Intake: January 2025

Course Types: Modular Undergraduate Course

Language: English

Duration: 6 months

Fees: $1391.78 View More Details on Fees

Area of Interest: Finance

Schemes: Alumni Continuing Education (ACE)

Funding: To be confirmed

School/Department: School of Business


Synopsis

FIN313 Machine Learning and AI for FinTech builds on the foundation from FIN312 to give an introduction to machine learning techniques in the handling of large datasets – the basis of AI. Students will learn fundamental concepts from the field, including supervised/unsupervised learning, bias-variance tradeoff, principal component analysis and neural networks. The course is peppered with examples of learning of datasets from finance. Students will be equipped with the understanding of how AI is applied in finance and the skill to implement machine learning algorithms to extract key features from financial datasets.

Level: 3
Credit Units: 5
Presentation Pattern: EVERY JAN

Topics

  • Supervised and unsupervised learning
  • Structured and unstructured data handling
  • Prediction accuracy – bias-variance tradeoff
  • Model interpretability
  • Regression and classification
  • ML models
  • Deep learning
  • Model testing using cross validation and bootstrapping
  • Dimension reduction – ridge and lasso regression
  • Principal component analysis
  • Neural networks
  • Python packages – numpy, scipy, pandas, scikit-learn and statsmodels

Learning Outcome

  • Distinguish between supervised machine learning (ML), unsupervised ML, deep learning and artificial intelligence.
  • Design and implement supervised ML algorithms to apply to financial datasets.
  • Examine and interpret ML models’ outputs and translate outputs into appropriate business decisions in financial settings.
  • Operate with high-dimensional financial datasets.
  • Formulate business requirements in Python code for automation.
  • Use suitable Python packages to build ML models for prediction or classification tasks.
Back to top
Back to top