Big data and open-source computation ...
Document type :
Autre communication scientifique (congrès sans actes - poster - séminaire...): Communication dans un congrès avec actes
Title :
Big data and open-source computation solutions, opportunities and challenges for marketing scientists. Applications to customer base predictive modeling using RFM variables
Author(s) :
Moulins, Jean-Louis [Auteur]
Centre de Recherche sur le Transport et la Logistique [CRET-LOG]
Francis, Salerno [Auteur]
Lille économie management - UMR 9221 [LEM]
Calciu, Michel [Auteur]
Centre de Recherche sur le Transport et la Logistique [CRET-LOG]
Francis, Salerno [Auteur]
Lille économie management - UMR 9221 [LEM]
Calciu, Michel [Auteur]
Conference title :
Marketing Trends
City :
Venise
Country :
Italie
Start date of the conference :
2016-01
English keyword(s) :
Big data
Map Reduce
HPC
predictive modeling
RFM
Map Reduce
HPC
predictive modeling
RFM
English abstract : [en]
Marketing researchers and analysts, confined in classical transactional marketing paradigms where customer knowledge was limited to sample based surveys and panels, have somehow been overwhelmed by the avalanche of behavioral ...
Show more >Marketing researchers and analysts, confined in classical transactional marketing paradigms where customer knowledge was limited to sample based surveys and panels, have somehow been overwhelmed by the avalanche of behavioral data coming from new digital and relationship marketing techniques. This occurred up to a point where a part of what belonged to their core competencies has been overtaken by computer scientists. The relatively recent open-source solutions that form an ecosystem around the most elegant statistical system, R and the distributed computation system Hadoop (some kind of Linux for computer clusters) democratize BigData calculations and offer excellent opportunities to marketing analysts to operate huge computing factories. An illustration of the implementation of the evoked solutions will be presented, applications to customer data base predictive modeling using RFM variables will be developed and performance gains will be tested.Show less >
Show more >Marketing researchers and analysts, confined in classical transactional marketing paradigms where customer knowledge was limited to sample based surveys and panels, have somehow been overwhelmed by the avalanche of behavioral data coming from new digital and relationship marketing techniques. This occurred up to a point where a part of what belonged to their core competencies has been overtaken by computer scientists. The relatively recent open-source solutions that form an ecosystem around the most elegant statistical system, R and the distributed computation system Hadoop (some kind of Linux for computer clusters) democratize BigData calculations and offer excellent opportunities to marketing analysts to operate huge computing factories. An illustration of the implementation of the evoked solutions will be presented, applications to customer data base predictive modeling using RFM variables will be developed and performance gains will be tested.Show less >
Language :
Anglais
Peer reviewed article :
Oui
Audience :
Internationale
Popular science :
Non
Collections :
Source :