DATA1002: Informatics: Data and Computation

Learning Outcomes

• Python Automation: Wrote procedural Python programs to automate data workflows based on defined algorithms.
• Data Pipeline Skills: Hands-on experience in ingestion, cleaning, transformation, summarization modeling, and evaluation.
• Tool Proficiency: Learned to choose between spreadsheets and Python depending on stage of data science, understanding trade-offs.
• Visualization & Communication: Created charts with Python and spreadsheets, evaluated clarity of communication.
• Machine Learning Basics: Built predictive models (classification & regression), practiced model evaluation, understood overfitting/underfitting.
• Data Representation: Understood logical vs. physical dataset representation and data sharing considerations (metadata, integrity, portability).

Takeaways

This was my first systematic experience using Python to clean and analyze data. The course moved at a fast pace: we began with basic data exploration and quickly advanced to a wide range of machine learning methods. In the group project, each member was required to apply a different ML model—including at least one technique beyond the lectures—which really pushed us to self-learn and explore independently. This not only deepened my understanding of classification and regression but also gave me confidence in applying models I had never seen before. At the same time, I faced challenges in the Python tests. As a beginner, adapting to a new programming language while keeping up with the workload was difficult, and my early performance was not ideal. However, this became a turning point: I reflected on my weaknesses, put in extra effort to strengthen my coding fundamentals, and gradually improved my ability to use Python effectively. By the end of the course, I not only had stronger technical skills, but also a better mindset for approaching steep learning curves with resilience.