DATA SCIENTIST / PHD IN COMPUTER SCIENCE

Machine learning for search, risk and language.

I build ML systems at Cdiscount: search relevance, fraud detection and credit scoring. Formerly NLP and information retrieval researcher at IRIT.

Portrait of Gia-Hung Nguyen
Bordeaux, FranceCdiscount

Experience

2021 / PRESENT

Cdiscount

Data Scientist, Feature Team Specialist

Two domains: Risk & Fintech (fraud detection, credit scoring with gradient boosting, drift monitoring) and Search & Discovery (product ranking relevance, production operations).

GBMSearchLLMMLOps
2019 / 2021

Myriad Data

Data Scientist

Neural models for image and text understanding, OCR and NLP pipelines, deployed on AWS.

NLPOCRAWS
2015 / 2018

IRIT, Toulouse

Research Assistant & PhD Candidate

Neural information retrieval enhanced with semantic resources (WordNet, DBpedia, UMLS). Taught Python, algorithms, C, SQL and web programming.

Deep LearningIRKnowledge GraphsTeaching
2015

IRIT, Toulouse

Research Intern

Twitter-based expert recommendation system: data collection, user profiling, clustering, crowdsourced validation.

Selected publications

PhD Thesis

Neural models for information retrieval: semantic source-driven approaches

University of Toulouse III, Paul Sabatier. 2018.

ICTIR 2017

DSRIM: a deep neural IR model enhanced by a knowledge resource driven representation of documents

With L. Tamine, L. Soulier, N. Souf.

ECIR 2017

Learning concept-driven document embeddings for medical information search

With L. Tamine, L. Soulier, N. Souf.

Full list on Google Scholar

Skills

MACHINE LEARNING

PythonTensorFlowKerasscikit-learnXGBoost

SEARCH & SEMANTICS

ElasticSearchLuceneSPARQLDBpediaRDF

DATA & INFRASTRUCTURE

SQLMongoDBAWSDockerGit

Education

University of Toulouse III, Paul Sabatier

PhD, Computer Science
2015 / 2018
Thesis: neural models for information retrieval, semantic source-driven approaches.

University of Toulouse III, Paul Sabatier

MSc, Computer Science
2014 / 2015
Information retrieval and databases.

Cantho University, Vietnam

Engineer, IT
2010 / 2014
Information systems and databases.

Photography

Landscape photography is how I spend time away from screens. A few favorites below, more on Flickr.