This course is about identifying underlying structure in data. We will talk about clustering, dimensionality reduction, word embeddings, and recommendation systems.
Course Learning Outcomes
By the end of the course, students are expected to be able to
- Explain the unsupervised paradigm.
- Explain the intuition behind clustering and use appropriate clustering algorithms for applications such as customer segmentation and document clustering.
- Interpret the results obtained after applying clustering.
- Explain the intuition behind dimensionality reduction and use such algorithms for applications such as anomaly detection.
- Explain the intuition of word2vec model to create word embeddings.
- Train your own word embeddings and use pre-trained word embeddings.
- Build word2vec style recommender system.
- Explain and build recommender systems, specifically using collaborative filtering approaches.