| Authors: |
Siddhartha Bhattacharyya
|
| URL: |
http://portal.acm.org/ft_gateway.cfm?id=347186&type=pdf&coll=GUIDE&dl=GUIDE&CFID=43813975&CFTOKEN=68162530 |
| Tags: |
Algorithms,
Design,
Experimentation,
Factors,
Human
Management,
Measurement,
Pareto-optimal
Performance,
Theory,
algorithms,
computation,
data
database
evolutionary
genetic
marketing,
mining,
models,
multiple
objectives
programming,
|
| Abstract: |
Predictive models in direct marketing seek to identify
individuals most likely to respond to promotional
solicitations or other intervention programs. While
standard modelling approaches embody single objectives,
real-world decision problems often seek multiple
performance measures. Decision-makers here desire
solutions that simultaneously optimise on multiple
objectives, or obtain an acceptable tradeoff amongst
objectives. Multi-criteria problems often characterise
a range of solutions, none of which dominate the others
with respect to the multiple objectives - these specify
the Pareto-frontier of nondominated solutions, each
offering a different level of tradeoff. This paper
proposes the use of evolutionary computation based
procedures for obtaining a set of nondominated models
with respect to multiple stated objectives. The
targeting depth-of-file presents a crucial real-world
criterion in direct marketing, and models here are
tailored for specified file-depths. Decision-makers are
thus able to obtain a set of models along the
Pareto-frontier, for a specific file-depth. The choice
of a model to implement can be thus based on observed
tradeoffs in the different objectives, based on
possibly subjective and problem specific judgements.
Given distinct models tailored for different
file-depths, the implementation decision can also
consider performance tradeoffs at the different
depths-offile. Empirical results from a real-world
problem illustrate the benefits of the proposed
approach. Both linear and nonlinear models obtained by
genetic search are examined. |
@inproceedings{347186,
title = {Evolutionary algorithms in data mining:
multi-objective performance modeling for direct
marketing},
address = {Boston, Massachusetts, United States},
author = {Siddhartha Bhattacharyya},
booktitle = {KDD '00: Proceedings of the sixth ACM SIGKDD
international conference on Knowledge discovery and
data mining},
pages = {465--473},
publisher = {ACM Press},
url = {http://portal.acm.org/ft_gateway.cfm?id=347186&type=pdf&coll=GUIDE&dl=GUIDE&CFID=43813975&CFTOKEN=68162530},
year = {2000},
abstract = {Predictive models in direct marketing seek to identify
individuals most likely to respond to promotional
solicitations or other intervention programs. While
standard modelling approaches embody single objectives,
real-world decision problems often seek multiple
performance measures. Decision-makers here desire
solutions that simultaneously optimise on multiple
objectives, or obtain an acceptable tradeoff amongst
objectives. Multi-criteria problems often characterise
a range of solutions, none of which dominate the others
with respect to the multiple objectives - these specify
the Pareto-frontier of nondominated solutions, each
offering a different level of tradeoff. This paper
proposes the use of evolutionary computation based
procedures for obtaining a set of nondominated models
with respect to multiple stated objectives. The
targeting depth-of-file presents a crucial real-world
criterion in direct marketing, and models here are
tailored for specified file-depths. Decision-makers are
thus able to obtain a set of models along the
Pareto-frontier, for a specific file-depth. The choice
of a model to implement can be thus based on observed
tradeoffs in the different objectives, based on
possibly subjective and problem specific judgements.
Given distinct models tailored for different
file-depths, the implementation decision can also
consider performance tradeoffs at the different
depths-offile. Empirical results from a real-world
problem illustrate the benefits of the proposed
approach. Both linear and nonlinear models obtained by
genetic search are examined.},
organisation = {SIGKDD: ACM Special Interest Group on Knowledge
Discovery in Data AAAI : Am Assoc for Artifical
Intelligence SIGART: ACM Special Interest Group on
Artificial Intelligence SIGMOD: ACM Special Interest
Group on Management of Data}, publisher_address = {New York, NY, USA}, size = {9 pages}, isbn = {1-58113-233-6}, notes = {p470 {"}For the non-linear GP, results were found to
be similar to those observed for the linear GA.
{"}Elitism always provides improved performance{"}.}, doi = {doi:10.1145/347090.347186},
keywords = {Algorithms, Design, Experimentation, Factors, Human Management, Measurement, Pareto-optimal Performance, Theory, algorithms, computation, data database evolutionary genetic marketing, mining, models, multiple objectives programming, }
}