Who should attend
Any professional from domains (e.g. healthcare, finance, manufacturing) that need to manage and work with data.
Data engineers, researchers, healthcare professionals and more.
About the course
Machine Learning uses techniques to deal with data in the most intelligent way – by developing algorithms – to derive actionable insights. With machine learning, you can glean useful patterns from the deep, focused troves of data specific to your chosen domain. Analysing that data provide insights that can drive successive new waves of efficiency and automation, reducing operational costs and potentially pinpointing new sources of revenue.
This course is part of a series of courses around Machine Learning-driven Data Science (MLDDS), under the StackUp – Startup Tech Talent Development programme offered by NUS-ISS.
The courses in the MLDDS series are:
- NICF – Data and Feature Engineering for Machine Learning (SF)
- NICF – Supervised and Unsupervised Modeling with Machine Learning (SF) [this course]
- NICF – Feature Extraction and Supervised Modeling with Deep Learning (SF)
- NICF – Sequence Modeling with Deep Learning (SF)
Throughout all courses, you will experience the 3 building blocks in machine learning:
- Concepts and intuition: Participants learn and apply the concepts of machine learning using a methodology. You will learn to navigate smoothly through the data sciences and machine learning space by not only creating but also debugging your products with ease. You will then apply the same concepts covered in a real world scenario while in class.
- Architecture: Participants learn what is required to architect a data science platform / team and how to effectively design a machine learning driven data sciences product using a wide range of techniques that are taught and practised in class. This includes, but is not limited to, understanding different parts of a data science product. The main goal of this course is to provoke thinking, establish the context of learning with the objective of developing and enhancing your capabilities in establishing a machine learning product.
- Implementation: Finally, participants will have the opportunity to apply the concepts using industry standard libraries and tools to develop their own machine learning driven data science project.
Upon completion of the course, the participants will be able to:
- Explain the concept and applications of machine learning models for classification
- Apply the machine learning workflow on additional algorithms
- Apply unsupervised learning techniques for data exploration
- Combine multiple machine learning models to improve performance metrics
What Will Be Covered
- Introduction and applications of Classification
- Evaluation metrics
- Text Feature Extraction and Classification
- Introduction to Clustering and Unsupervised Learning
- Decision Trees
- Ensemble methods and Boosting
- Bridging Recap: Data Engineering, Feature Engineering and Modeling for Machine Learning
- Lecture: Classification: Logistic Regression, Support Vector Machines, metrics, applications
- Workshop: Recap, Classification
- Lecture: Text Processing: tokenization, bag-of-words, TF-IDF, embedding, applications
- Workshop: Text Processing
- Lecture: Clustering: K-means, Hierarchical clustering, metrics, applications
- Workshop: Clustering
Project Part 1: Train a machine learning model
- Lecture: Decision Trees: concept, ID3 algorithm, applications
- Workshop: Decision Trees
- Lecture: Ensemble Methods: bagging and boosting
- Workshop: Ensemble Methods
- Project Part 2: Tune and optimise machine learning model
- Project Presentations
Lisa is with the Software Engineering and Design Practice, StackUp program for National University of Singapore, Institute of Systems Science (NUS-ISS). Lisa has multiple years of extensive experience in software product research and development at Microsoft Corporation (USA). Her background in...
Because of COVID-19, many providers are cancelling or postponing in-person programs or providing online participation options.
We are happy to help you find a suitable online alternative.