SC4000 Machine Learning

Course Summary

The course offers an introduction to the variety of fundamental machine learning algorithms. The course focuses more on fundamental concepts and algorithms which proves to be useful for further studies in machine learning. The topics covered are:

1. Machine learning, its types and applications
2. Bayesian classifiers (e.g. naive Bayes, Bayesian belief networks)
3. Decision trees
4. Artificial neural networks
5. Support vector machines
6. Regression models
7. k-nearest neighbour classifiers
8. Ensemble learning
9. Clustering (e.g. k-means, hierarchical)
10. Density estimation
11. Dimensionality reduction

The topics covered may already be familiar to you if you have taken other courses like SC1015 and SC4020. New topics unique to this course may be Bayesian belief networks, density estimation and dimensionality reduction.

Workload and assessment

The module is not too heavy. The assessment is group project and final exam. The group project is a Kaggle competition (from a choice of competitions which you may choose from) or a research project. The project is likely to cover topics outside of what the course teaches so may expect self-learning to some extent. The final exam may have unexpected questions so do prepare yourself for it.

Conclusion

This course is a compulsory course for DSAI students and is definitely a useful module, especially if you have not seen the topics. Otherwise, you can treat this course as a refresher.