About this Event
Lead trainer:
Pasan Kalansooriya (Melbourne Bioinformatics, The University of Melbourne)
Workshop Description:
This workshop is an introduction to machine learning for life scientists and clinicians who are new to predictive modelling. Working with a real clinical dataset of 569 breast tissue biopsies, we will build and evaluate a classifier that predicts whether a tumor is malignant or benign. We will cover exploring and scaling the data, using Principal Component Analysis to visualize 30 features at once, training a Random Forest, and interpreting the results using sensitivity, specificity, and feature importance. The focus is on building intuition rather than working through mathematics.
Learning objectives:
By the end of this training, participants will be able to:
- Map clinical research questions to the correct machine learning vocabulary (e.g., features, labels, classification vs. regression).
- Preprocess and scale clinical tabular data using pandas and scikit-learn.
- Apply Principal Component Analysis (PCA) to visualize high-dimensional patient data in a 2D plot.
- Train, evaluate, and interpret a Random Forest classifier.
- Extract "Feature Importance" to understand which biological markers drive the model's predictions.
Eligibility:
This free workshop is available to staff and students at The University of Melbourne and its affiliated institutes.
You must register for this event using an affiliated institutional email address or your registration may be cancelled.
Prerequisites and Requirements:
This workshop is designed for participants with a basic level of Python knowledge. You should be comfortable with variables, lists and dictionaries, and have seen pandas DataFrames before (loading a CSV and viewing it is enough). No prior machine learning experience is required. This is an in-person hands-on workshop and attendees must bring their own computers (laptop chargers also recommended). You will also need:
- A Google account, as all coding is done in Google Colab
- Wifi access to internet via UniWireless or Eduroam
- A web browser (Firefox or Chrome recommended)
No software installation is required.
Access
If you require any further information, or have any access requirements in order to participate in this workshop, please contact us as soon as possible to discuss your requirements:
We recommend following our Eventbrite page if you wish to be alerted when we release new workshops for registrations. If you require any further information, please contact Melbourne Bioinformatics at: [email protected]
Important Policy Update:
Starting 1 August 2025, participants who miss two or more workshops without proper notice (less than 48 hours or attending under 50%) within 12 months will be placed on a 6-month waitlist for full workshops. To avoid being marked as a no-show, cancellations must be made at least 24 hours in advance, with exceptions allowed for emergencies.Please read the full policy here: https://mbite.org/guides/policy/
Event venue & nearby stays
21 Bedford St, 21 Bedford Street, North Melbourne, Australia
Concerts, fests, parties, meetups — all the happenings, one place.