Machine learning event

Who is the target audience of the workshop?

Individuals from all areas of infectious disease and infection science interested in learning more about data science and machine learning. No prior background in data science and machine learning is assumed and the workshop is suitable for beginners.

What are the learning objectives ?

1. Participants will develop an understanding of the “end-to-end” lifecycle of a machine learning solution.
2. Participants will develop a fundamental understanding of machine learning and data science.
3. Participants will develop their own machine learning workflow and models for tabular datasets with no-code tools.
4. Everyone will receive copies of all the workshop material, in addition to supplementary material including more machine learning examples so participants can further explore the field and develop their understanding beyond the workshop.

What is involved in the hands-on sections of the workshop?

The hands-on sections will involve participants using their own laptops, with the orange data mining software, which is free and open source and can be downloaded from orangedatamining.com. More details will be provided to participants 8 weeks, 4 weeks, and 2 weeks before the workshop. During the workshop, participants will be guided by the organisers, ensuring that all participants are successful in achieving the learning objectives.

Is there a more detailed curriculum for the workshop?

Yes, a more detailed curriculum is provided as follows:

• Participants will be Introduced to data science and machine learning. Examples of machine learning applied for infection science will be discussed and we will provide sufficient background to support the hands-on components of the workshop.
   
• We will then proceed to discuss the "end-to-end" machine learning lifecycle. We will explore the key principles of building machine learning solutions, the types of machine learning problems, and when to use machine learning and when not to use it. We will discuss preparing and managing data for machine learning, training machine learning models, evaluating machine learning models, and briefly discuss other areas such as deployment, monitoring, and continual learning.
   
• Hands on session 1 will introduce participants to the software that we will be using (Orange data mining software), which is open source and free to download and use. The practical project that we will be working on will be introduced, which focuses on blood culture outcome prediction using haematology data, replicating the results presented in our team's recent paper which can be found at bmcinfectdis.biomedcentral.com/articles/10.1186/s12879-023-08535-y. Additional examples will be provided as supplementary material.
   
• Hands on session 2 will be about cleaning and preparing the tabular dataset for machine learning, and we will also discuss feature engineering.
   
• This will flow into hands on session 3 where we will start training our different machine learning models and start implementing strategies for improving machine learning model performance.
   
• Hands on session 4 will focus on evaluation of our machine learning models, discussing how we can effectively evaluate machine learning performance, and the importance of correct machine learning model evaluation.
   
• Hands on session 5 will examine the important considerations for implementing machine learning in practice, and we will test our models with new data and retrain the models to try and improve performance.
   
• Towards the end of the workshop, we will provide an introduction to enhancing research and data analysis with large language models, before concluding with a discussion about the future of ML for infection science.
   
• There will be time for questions, open discussion, technical support, and networking.