Programme Outline
Learning Objectives and Structure
- Perform the basic ML model component for the role of a junior data scientist
- Understand the types of data and databases in the business context
- Appreciate the use of data dictionary and harness the potential of metadata for data science
- Acquire organizational dataset from data lakes and other democratized data sources for data enrichment purposes
- Structure data into an appropriate form for data analysis
- Manipulate data structures to support data-wrangling phase
- Perform data wrangling on the acquired dataset
- Address data quality issues with appropriate data cleansing technique
- Iterate the data mining process progressively with the provision of data wrangling and exploratory analysis tools
- Understand healthcare case studies shared by SingHealth faculty members to gain insights into real-world scenarios.
- Utilise curated public healthcare datasets to perform hands-on activities and assignments, fostering practical experience and understanding of the subject matter.
Day 1
- Overview of Data Science Pipeline
- What is Data Wrangling and Data Preparation?
- Data Acquisition
- Understand how data scientist prepares the dataset for data modelling
- What is data discovery?
- Types of data
- Types of databases
- Data Dictionary and Metadata
- Data Models
Day 2
- Data Mining and CRISP-DM
- Common Computing Infrastructure
- Interactive Data Exploratory Analysis (IDEA)
- Basics of Descriptive Statistics
Day 3
- Breakdown of Data Preparation Phases
- Dataset Structuring: Data Frame Handling
- Data Cleaning
Day 4
- Data Enrichment and alternative sources
- Data Enrichment: Data Aggregation
- Data Enrichment: Data Standardisation
Day 5 – Consultation / Project presentation
Project Consultation
Each group of participants will present the progress of their projects and have the opportunity to ask questions and clarify any doubts pertaining to their projects.
Project Presentation
Each group of participants will showcase their work and respond to questions during a Q&A session.