CV

Education

Master Degree in Mathematics and Applications Specialty Mathematical Engineering Financial Engineering and Random Models, Sorbonne University, 2014-2016
Certification In Big Data, Sorbonne University, 2015-2016
Master Degree in Actuarial Science, ISUP, 2017-2019 (expected)
Ph.D in Data Science/Insurance, Sorbonne University, 2017-2020 (1st July 2022)

Work experience

May 2022 - Today: Applied Scientist, Jellysmack
- Creating and deploying a video recommendation model for Facebook based on trends
- Working on a topic extraction on Youtube videos
- Creating data quality analysis function for Time Series
- Creating preprocessing function for Time Series
August 2020 - April 2022: Machine Learning Engineer, PacketAI
- Creating and deploying an anomaly detection model in log sequence streams (on Tensorflow)
- Working on a scalable method for log parsing based on an Embedding approach
- Working on a Python bot for log parsing
- Creating preprocessing function for Time Series and Log files
March 2017 - March 2020: PhD, LPSM (Sorbonne University) and Pacifica (CAA)
- Using text to predict the severity of a claim. Predicted cost 40% better than traditional figures (on Tensorflow, Pyspark)
- Programming on GPU (on AWS and local)
- Combining Deep Learning models (LSTM and CNN) on a temporal dataset containing structured and unstructured data
- Creating a Graphic tool to interpret Grid Search and Model results (on Plotly)
September 2016 - February 2017: P&C Actuarial Science, Pacifica (CAA)
- Extreme Value Theory.
- Impact of bodily injury claims.
March 2016 - August 2016: P&C Actuarial Science (Internship), Pacifica (CAA)
- Studying the geographical component on a bodily claim on Life Accident Guarantee
- Challenging GLM, Random Forest and Gradient Boosting for modeling Average Cost and Frequency. Improvement of the loss by two (on Scikit-Learn)

Side Projects

Automatic invoice processing with Tensorflow for Digipharmacie.
Discord bots for Elrond NFT projects.
Learning Rust to deploy Smart contrat on Elrond blockchain.
Crypto portfolio tracker in Python.

Skills

Programming
- Python
- R
- C++
- Julia
- C
- Cuda
AWS
Docker
Python packages
- Tensorflow
- PyTorch
- Scikit-Learn
- Plotly
- Matplotlib

Publications

Automatic analysis of insurance reports through deep neural networks to identify severe claims.

Cohen Sabban, I., Lopez, O., & Mercuzot, Y. (2021). Automatic analysis of insurance reports through deep neural networks to identify severe claims. Annals of Actuarial Science, 1-26. doi:10.1017/S174849952100004X.

Method for automatically detecting anomalies in log files.

Statistical analysis of the evolution of extreme claims for bodily risk cover.

Isaac Cohen Sabban. Analyse statistique de l'évolution des sinistres graves pour une garantie risque corporel. Mathématiques. Sorbonne Université, 2022. Français.