Range, across domains and methods.
Beyond the flagship case studies, a body of applied work (academic and independent) spanning forecasting, deep learning, biostatistics, NLP and interactive analytics. Several ship as live applications.
Heart-disease risk prediction with ML ensemble + CNN
A predictive study combining a machine-learning ensemble with a convolutional neural network to estimate heart-disease risk.
Cancer risk in type-2 diabetics over 10 years
An engineering study, in collaboration with an AP-HP hepatologist, modelling hepatocellular-carcinoma risk in type-2 diabetic patients from obesity, hypertension and clinical factors.
Land-cover classification from satellite time series
Classification of land-use parcels in Brittany from 2017 satellite imagery, learning from partially labelled data to predict land-cover class for new parcels.
Twitter's effect on Bitcoin price
An exploratory study scraping Bitcoin-related tweets and market data via APIs to investigate the relationship between social signal and price movement.
RFM customer segmentation for a lingerie brand
Behavioural analysis and RFM segmentation to help a French retail brand understand customers, differentiate communication and allocate marketing spend.
Global infant-mortality evolution since 2000
A data-visualization study of under-five mortality worldwide (2000-2021), relating changes to health expenditure and GDP to test where progress was universal and where it was not.
Additional projects and source code are available on GitHub and LinkedIn.
Have an AI system that needs to reach production?
Let's talk about the problem you're trying to solve and whether AI is the right tool for it. No buzzwords, just an honest technical read.