Vincent Lemaire

Topics: machine learning, data mining

Application domains 2021 : fraud detection, XAI, time series, weakly supervised learning

Research

My research results are detailed in my scientific [publications...], [invited talks...] or [patents...].

Collaborations

Students: Up to the end of 2021 I have been the director of 24 students including 5 PhDs (see the page [students...]).
Participation and organization of research in and outside the Orange Group: I (co)organize workshop, challenges, etc (see the page [committee, review...]).
Collaborations: My main collaborations are with the members of my team at Orange Labs. I have also currently collaborations with external partners; see the page [collaborations...].

Development:

Software: Since 2012 my activities do not allow me to have sufficient time to code new software by myself. But I manage software engineers (I give very detailed specifications) who code for me new software or improve existing software (as Khiops Ennéade or Khiops interpretation for example (see the page [software...])). I remain only coding in python at the moment, before 2012 I was coding in C.
Methods: I develop new generic methods (or methodologies) and I conduct data analysis; for the Orange Business Units.

Expertise:

to help and advice the Business units of Orange (if you read “French” you may find details on this role in my [HDR...] dissertation (section A.4.1 pages 94-96).
or to participate to the evaluation of others research teams or students, in and outside Orange (see the page [jury...]).

Influence: I give regularly [courses...] to data miners, data scientist of the Orange Group (in France and others countries). The courses are also the opportunity to share good practices and advices.

Research project manager: I also manage a research project on “data analytics” (in the research domain “Data and Knowledge” (DAK) of Orange Labs Research. The budget of this project corresponds to the management of research team of 12 people (including 4-5 PhD students). The decision-making supports and data analytics are crucial for the operators of telecoms. The project recovers a set of techniques allowing: to classify, to discover or to learn from the data, and thus to help in the decision; to analyze the behavior of a network or a customer; to exploit more and more important data...

Advanced Analytics, Machine Learning, Fraud Detection, Time Serie. Vincent Lemaire is a data scientist and a research project manager in the area "Data Analytics and Knowledge" in the research domain "Data and Knowledge" at Orange Labs, France. He obtained his undergraduate degree from the University of Paris 12 in signal processing and was in the same period an Electronic Teacher. He obtained a PhD in Computer Science from the University of Paris 6 in 1999. He thereafter joined the R&D Division of Orange (France Telecom), where he became a senior expert in data-mining. He obtained his Research Accreditation (HDR) in Computer Science from the University of Paris-Sud 11 (Orsay) in 2008. His research interests are the application of machine learning in various areas for telecommunication companies with an actual main application in data mining for business intelligence, fraud detection, churn prediction. He has organized several machine learning workshops and competitions including the AISTATS 2010 challenge on Active Learning and the ECML PKDD 2012 challenge on Active Learning in Real-World Application, the KDD Cup 2009, ...

The main topic is machine lerning which is a part of Artificial Intelligence (AI). AI is a vast and multidisciplinary field with the ultimate ambition of making machines, especially computer systems, reproduce the cognitive skills that are normally the prerogative of the human being. These cognitive skills include (but are not limited to) environment perception, knowledge representation, learning, reasoning and problem solving. The main goal of AI research is to produce the technology to create such systems, using many tools like mathematical optimization, neural networks, logic, statistics and machine learning (ML).