Machine learning is a topic which is leading the trend in the transformation of organisations from digitisation to data-driven. This course aims to expose the learner to the code underpinnings of constructing and assessing a subset of the machine learning tasks, namely the classification, time series and clustering/feature extraction tasks. The objective is to target commonly encountered tasks in the space of financial data analysis. Participants will be able to gain familiarity in such tasks and have a deeper understanding of tackling them.
2 days (16 hours)
$866.70/- (GST Inclusive)
Medium of Instruction: English
To enable learners to have an overall understanding of the landscape of Data Mining and Machine Learning.
- Introduction: Machine Learning and Data Mining
- Supervised and Unsupervised learning techniques
To give the participants the very first steps towards forming hypothesis and doing predictive analytics, learning some of the classification techniques in supervised learning.
- Machine Learning and Classification
- Classification: Logistic Regression, Decision Trees, Random Forest, XGBoost
It aims to equip participants with the basics of how supervised machine learning models work and how evaluation and optimization can be carried out.
- Evaluation and Metrics
- Understanding and comparing model results
- Feature importance
Day 2 will continue the journey for the learners in machine learning with focus on Neural Networks and Unsupervised learning techniques.
This module work with exploring time-based data, here participants will learn about how to assess and play with past data to explain future movements.
- ARIMA model – seasonality
To equip learners in understanding some of the Unsupervised learning techniques which is required for machine learning.
- Clustering and Segmentation
- Principal Component Analysis
- Feature Reduction
- AB Testing
To further equip learners with knowledge on neural networks, this is a flexible method for which new machine learning fields are being pioneer.
- Introduction to neural networks
- Working with a neural network
- Construct analytics models/results as part of solutions to address business problems
- Evaluate the performance of analytics models
- Analyse the results or outputs of analytics models
- Evaluate the importance of features which are used
- Understand how to reduce features
Mode of Assessment
Written and Practical
Windows OS or MAC OS installed with anaconda