OxfordRUG September 2023 | ML to Prevent Human Trafficking

For our next meetup on 25th September, we are very excited to have Chantelle Whelan giving a talk on "Using Machine Learning to Prevent Human Trafficking"

Agenda:

6:00 pm: Talk "Using Machine Learning to Prevent Human Trafficking" by Chantelle Whelan

7:00 pm: Pizza!

Abstract:

Merkle have a pro-bono partnership with the charity STOP THE TRAFFIK (STT), a pioneer in human trafficking prevention through an innovative intelligence-led approach. One piece of work from this partnership has been to create a data-driven Human Trafficking Incident Tool, making use of Machine Learning (ML) techniques, such as Decision Tree, Random Forest and XGBoost, to create a picture of the most likely trafficking scenarios.

The models were created in R using ~60k global trafficking incidents recorded by STT and run in Power BI to create interactive dashboards that STT could access in a user-friendly format.
What’s game-changing about this tool is that ML algorithms can identify patterns and trends of trafficking incidents with limited representation in the data. For example, where there are few recorded incidents for certain combinations of a victim’s nationality, gender and age group. The models will consider the patterns for those characteristics individually to then build a picture of the most likely scenarios, something that we wouldn’t be able to achieve by simple visualisations or exploratory data analysis.

This tool allows STT to optimise targeting of their campaigns, increasing the impact and reach of their work.

Speaker bio:

Chantelle Whelan is a Lead Analytical Consultant at Merkle, where she has been working since 2019. Her role involves helping her clients leverage their customer data to inform their marketing campaigns. Chantelle’s primary focus is to deliver powerful solutions that meet her client’s business needs. She achieves this through a combination of machine learning models, descriptive data analysis and data visualisation. Prior to her work as a data analyst, Chantelle worked in academia where she conducted research in Psychology and Animal Behaviour.

Chantelle has a keen interest in DataOps and the development of ethical machine learning. In her role at Merkle she has helped clients integrate best practices in their data operations, such as the use of CI/CD. She is currently developing solutions to track and reduce carbon emissions when building machine learning models, as well as highlighting biases in the training data.

Thanks to our sponsor:

Oxford RUG is greateful for funding from the R consortium.