The Power of Sampling and Stacking for the PaKDD-2007 Cross-Selling Problem

Paulo J.L. Adeodato; Germano C. Vasconcelos; Adrian L. Arnaud; Rodrigo C.L.V. Cunha; Domingos S.M.P. Monteiro; Rosalvo F.O. Neto

doi:10.4018/jdwm.2008040104

To Support Customers in Easily and Affordably Obtaining Titles in Electronic Format IGI Global is Now Offering a 50% Discount on ALL E-Books and E-Journals Ordered Directly Through IGI Global’s Online Bookstore
Additionally, Enjoy a 20% Discount on all Other Products and FormatsBrowse Titles

Special Offers
- As Part of Our Efforts to Assist Customers with More Easily and Affordably
  Obtaining Titles in Electronic Format, IGI Global is Now Offering a 50% Discount on
  All E-Books and E-Journals Ordered Through IGI Global’s Online Bookstore*
  To support customers with accessing online resources, IGI Global is offering a 50% discount on all e-book and e-journals. This opportunity is ideal for librarian customers convert previously acquired print holdings to electronic format at a 50% discount.
  *The 50% discount is offered for all e-books and e-journals purchased on IGI Global’s Online Bookstore. E-books and e-journals are hosted on IGI Global’s InfoSci® platform and available for PDF and/or ePUB download on a perpetual or subscription basis. This discount cannot be combined with any other discount or promotional offer. Offer expires June 30, 2020.
  Browse Titles
  Convert Your IGI Global Print Holdings to Electronic Format through IGI Global’s
  InfoSci® Platform or ProQuest’s Ebook Central, or EBSCOhost
  To assist you during the COVID-19 pandemic, IGI Global will convert libraries previously acquired print holdings to electronic formats directly through our InfoSci® platform, ProQuest’s E-Book Central, or EBSCOhost at a 50% discount. Send us a list of IGI Global publications you would like to convert, and we’ll promptly facilitate the set-up and access.
  Learn More
- View Over 900+ IGI Global Titles Related to COVID-19 Pandemic
  IGI Global offers a rich volume of content related to treatment, mitigation, and emergency and disaster preparedness surrounding epidemics and pandemics such as COVID-19. All of these titles are available in electronic format at a 50% discount making them ideal resources for online learning environments.
  View Titles
- Create a Free IGI Global Library Account to Receive an Additional 5% Discount on All Purchases
  Exclusive benefits include one-click shopping, flexible payment options, free COUNTER 5 reports and MARC records, and a 5% discount on single all titles, as well as the award-winning InfoSci^®-Databases.
  Sign Up Now!
Books
Journals
InfoSci^®-Databases
Articles/Chapters
Publish with Us
Resources
- - Instructors
  - Course Adoption
  - Teaching Cases
  - K-12 Online Learning Collection
  - Authors and Editors
  - eEditorial Discovery^® System
  - Peer Review Process
  - Ethics and Malpractice
  - COPE Membership
  - Fair Use Policy
  - Open Access Publishing
  - Editorial Services
  - FAQ
Catalogs
About Us
Newsroom

The Power of Sampling and Stacking for the PaKDD-2007 Cross-Selling Problem

Paulo J.L. Adeodato (NeuroTech Ltd. and Federal University of Pernambuco, Brazil), Germano C. Vasconcelos (NeuroTech Ltd. and Federal University of Pernambuco, Brazil), Adrian L. Arnaud (NeuroTech Ltd. and Federal University of Pernambuco, Brazil), Rodrigo C.L.V. Cunha (NeuroTech Ltd. and Federal University of Pernambuco, Brazil), Domingos S.M.P. Monteiro (NeuroTech Ltd. and Federal University of Pernambuco, Brazil) and Rosalvo F.O. Neto (NeuroTech Ltd. and Federal University of Pernambuco, Brazil)

Source Title: International Journal of Data Warehousing and Mining (IJDWM) 4(2)

DOI: 10.4018/jdwm.2008040104

OnDemand PDF Download:

$37.50

Abstract

This article presents an efficient solution for the PAKDD-2007 Competition cross-selling problem. The solution is based on a thorough approach which involves the creation of new input variables, efficient data preparation and transformation, adequate data sampling strategy and a combination of two of the most robust modeling techniques. Due to the complexity imposed by the very small amount of examples in the target class, the approach for model robustness was to produce the median score of the 11 models developed with an adapted version of the 11-fold cross-validation process and the use of a combination of two robust techniques via stacking, the MLP neural network and the n-tuple classifier. Despite the problem complexity, the performance on the prediction data set (unlabeled samples), measured through KS2 and ROC curves was shown to be very effective and finished as the first runner-up solution of the competition.

Complete Article List

Search this Journal:

Reset

Open Access Articles

Volume 17: 4 Issues (2021): Forthcoming, Available for Pre-Order

Volume 16: 4 Issues (2020): 3 Released, 1 Forthcoming

Volume 15: 4 Issues (2019)

Volume 14: 4 Issues (2018)

Volume 13: 4 Issues (2017)

Volume 12: 4 Issues (2016)

Volume 11: 4 Issues (2015)

Volume 10: 4 Issues (2014)

Volume 9: 4 Issues (2013)

Volume 8: 4 Issues (2012)

Volume 7: 4 Issues (2011)

Volume 6: 4 Issues (2010)

Volume 5: 4 Issues (2009)

Volume 4: 4 Issues (2008)

Volume 3: 4 Issues (2007)

Volume 2: 4 Issues (2006)

Volume 1: 4 Issues (2005)

View Complete Journal Contents Listing

MLA

APA

Chicago

Export Reference

The Power of Sampling and Stacking for the PaKDD-2007 Cross-Selling Problem

Abstract

Complete Article List