Cargando…
Data Mining and Real Systems Modeling
<!--HTML-->Over the past fifty years, Database development has known a sudden and fast acceleration, especially those containing information about consumers on a given market. This has led to the explosion of the Relational Marketing, and later on to the rise of the CRM – the Customer Relation...
Autor principal: | |
---|---|
Lenguaje: | eng |
Publicado: |
2016
|
Materias: | |
Acceso en línea: | http://cds.cern.ch/record/2239040 |
_version_ | 1780952949995864064 |
---|---|
author | Chaskalovic, Joel |
author_facet | Chaskalovic, Joel |
author_sort | Chaskalovic, Joel |
collection | CERN |
description | <!--HTML-->Over the past fifty years, Database development has known a sudden and fast acceleration, especially those containing information about consumers on a given market. This has led to the explosion of the Relational Marketing, and later on to the rise of the CRM – the Customer Relationship Management – which one of its main applications is targeting optimization and consumer knowledge development. A large number of consequences related to Marketing and Media Strategies resulted from this approach on a given Market.
One fundamental in the process of developing knowledge on consumer's behavior is based on Data Mining; finding out from a suitable Datawarehouse the right information corresponding to a set of Marketing questions - recruitment, loyalty, sales. It could thus be helpful in exhibiting discriminant characteristics of prospects, ways to communicate with them or fixing rules for clients’ loyalty.
We propose to extend and generalize to other kind of “populations” this process that consists in collecting the right information, to mine and to analyze it, and therefore, to model particular and real behaviors for a specific group of the population.
Indeed, we suggest taking this methodological frame to apply the same process to understand real and physical systems.
The aim of this methodology is to discover by mining experimental Databases, the discriminant characteristics of a given physical system that will be taken into account in a relevant mathematical model.
It would be a complementary method that will show how laboratory experiments could participate very closely to mathematical modeling, particularly to decide which physical factors must be taken into account in a realistic model.
It means that one will have to build the right Database with potential explicative variables, in relation with the mechanisms to be tested. Therefore, the Data Mining process will lead to exhibit the relative importance and effectiveness of each mechanism supposed to be responsible for the life of a real system.
Several examples of this methodology will be proposed in Media & Marketing contexts on the one hand, and for scientific applications on the other hand.
Short Bio of the speaker:
Prof. Joel Chaskalovic is a member of D'Alembert Institute at University Pierre and Marie Curie in Paris. He has been qualified Full Professor of the French universities in Applied mathematics after he spent for many years in mathematical modelling applied to fluid and solid mechanics, media and marketing, geomarketing, medicine, and for complexity and randomness in physical systems. In 2014, he was invited as Chief Editor in the Comptes Rendus de Mécanique of the French Academy of Sciences for a special issue untitled “Theoretical and Numerical Approaches for Vlasov-Maxwell Equations”. Between 1993 to 2007, he also worked for Publicis Group as Director of Scientific Research, and therefore, as Director of Data Mining and CRM where he discovered in the 2000s the potential of these fields applied to any kind of Big Data that he implemented upon in several application contexts. |
id | cern-2239040 |
institution | Organización Europea para la Investigación Nuclear |
language | eng |
publishDate | 2016 |
record_format | invenio |
spelling | cern-22390402022-11-03T08:15:31Zhttp://cds.cern.ch/record/2239040engChaskalovic, JoelData Mining and Real Systems ModelingData Mining and Real Systems ModelingAcademic Training Lecture Regular Programme<!--HTML-->Over the past fifty years, Database development has known a sudden and fast acceleration, especially those containing information about consumers on a given market. This has led to the explosion of the Relational Marketing, and later on to the rise of the CRM – the Customer Relationship Management – which one of its main applications is targeting optimization and consumer knowledge development. A large number of consequences related to Marketing and Media Strategies resulted from this approach on a given Market. One fundamental in the process of developing knowledge on consumer's behavior is based on Data Mining; finding out from a suitable Datawarehouse the right information corresponding to a set of Marketing questions - recruitment, loyalty, sales. It could thus be helpful in exhibiting discriminant characteristics of prospects, ways to communicate with them or fixing rules for clients’ loyalty. We propose to extend and generalize to other kind of “populations” this process that consists in collecting the right information, to mine and to analyze it, and therefore, to model particular and real behaviors for a specific group of the population. Indeed, we suggest taking this methodological frame to apply the same process to understand real and physical systems. The aim of this methodology is to discover by mining experimental Databases, the discriminant characteristics of a given physical system that will be taken into account in a relevant mathematical model. It would be a complementary method that will show how laboratory experiments could participate very closely to mathematical modeling, particularly to decide which physical factors must be taken into account in a realistic model. It means that one will have to build the right Database with potential explicative variables, in relation with the mechanisms to be tested. Therefore, the Data Mining process will lead to exhibit the relative importance and effectiveness of each mechanism supposed to be responsible for the life of a real system. Several examples of this methodology will be proposed in Media & Marketing contexts on the one hand, and for scientific applications on the other hand. Short Bio of the speaker: Prof. Joel Chaskalovic is a member of D'Alembert Institute at University Pierre and Marie Curie in Paris. He has been qualified Full Professor of the French universities in Applied mathematics after he spent for many years in mathematical modelling applied to fluid and solid mechanics, media and marketing, geomarketing, medicine, and for complexity and randomness in physical systems. In 2014, he was invited as Chief Editor in the Comptes Rendus de Mécanique of the French Academy of Sciences for a special issue untitled “Theoretical and Numerical Approaches for Vlasov-Maxwell Equations”. Between 1993 to 2007, he also worked for Publicis Group as Director of Scientific Research, and therefore, as Director of Data Mining and CRM where he discovered in the 2000s the potential of these fields applied to any kind of Big Data that he implemented upon in several application contexts.oai:cds.cern.ch:22390402016 |
spellingShingle | Academic Training Lecture Regular Programme Chaskalovic, Joel Data Mining and Real Systems Modeling |
title | Data Mining and Real Systems Modeling |
title_full | Data Mining and Real Systems Modeling |
title_fullStr | Data Mining and Real Systems Modeling |
title_full_unstemmed | Data Mining and Real Systems Modeling |
title_short | Data Mining and Real Systems Modeling |
title_sort | data mining and real systems modeling |
topic | Academic Training Lecture Regular Programme |
url | http://cds.cern.ch/record/2239040 |
work_keys_str_mv | AT chaskalovicjoel dataminingandrealsystemsmodeling |