Machine Learning – Jessica Cebrian

Data Scientist

Showing 2 result(s)

Distilling knowledge from complex models to increase interpretability of predictions

on 01/10/202101/10/2021

The top high performance machine learning algorithms of today are expensive to store, hard to evaluate and interpret by humans. Imagine being able to have the great accuracy of a Neural Network or a Random Forest but with the ability to be interpretable. This would be a super helpful implementation in industries where transparency is …

List of projects

on 14/09/202127/09/2021

A ground-breaking matter price prediction tool for the legal sector. Retail: Store-product clusterization algorithms. Distilling Knowledge from complex models to explainable simple models: using Random Forest accuracy along with DT interpretability to provide interpretable predictions. Using brain signal data and Machine Learning to provide insights on maternal feeding behaviour.