Integrating Economic Knowledge in Data Mining Algorithms

H.A.M. Daniëls, A.J. Feelders

Research output: Working paperDiscussion paperOther research output

364 Downloads (Pure)

Abstract

The assessment of knowledge derived from databases depends on many factors. Decision makers often need to convince others about the correctness and effectiveness of knowledge induced from data.The current data mining techniques do not contribute much to this process of persuasion.Part of this limitation can be removed by integrating knowledge from experts in the field, encoded in some accessible way, with knowledge derived form patterns stored in the database.In this paper we will in particular discuss methods for implementing monotonicity constraints in economic decision problems.This prior knowledge is combined with data mining algorithms based on decision trees and neural networks.The method is illustrated in a hedonic price model.
Original languageEnglish
Place of PublicationTilburg
PublisherOperations research
Number of pages16
Volume2001-63
Publication statusPublished - 2001

Publication series

NameCentER Discussion Paper
Volume2001-63

Fingerprint

Data mining
Economics
Decision trees
Neural networks

Keywords

  • knowledge
  • neural network
  • data mining
  • decision trees

Cite this

Daniëls, H. A. M., & Feelders, A. J. (2001). Integrating Economic Knowledge in Data Mining Algorithms. (CentER Discussion Paper; Vol. 2001-63). Tilburg: Operations research.
Daniëls, H.A.M. ; Feelders, A.J. / Integrating Economic Knowledge in Data Mining Algorithms. Tilburg : Operations research, 2001. (CentER Discussion Paper).
@techreport{039b0321bd0b487ba542b1ebb2d30d86,
title = "Integrating Economic Knowledge in Data Mining Algorithms",
abstract = "The assessment of knowledge derived from databases depends on many factors. Decision makers often need to convince others about the correctness and effectiveness of knowledge induced from data.The current data mining techniques do not contribute much to this process of persuasion.Part of this limitation can be removed by integrating knowledge from experts in the field, encoded in some accessible way, with knowledge derived form patterns stored in the database.In this paper we will in particular discuss methods for implementing monotonicity constraints in economic decision problems.This prior knowledge is combined with data mining algorithms based on decision trees and neural networks.The method is illustrated in a hedonic price model.",
keywords = "knowledge, neural network, data mining, decision trees",
author = "H.A.M. Dani{\"e}ls and A.J. Feelders",
note = "Pagination: 16",
year = "2001",
language = "English",
volume = "2001-63",
series = "CentER Discussion Paper",
publisher = "Operations research",
type = "WorkingPaper",
institution = "Operations research",

}

Daniëls, HAM & Feelders, AJ 2001 'Integrating Economic Knowledge in Data Mining Algorithms' CentER Discussion Paper, vol. 2001-63, Operations research, Tilburg.

Integrating Economic Knowledge in Data Mining Algorithms. / Daniëls, H.A.M.; Feelders, A.J.

Tilburg : Operations research, 2001. (CentER Discussion Paper; Vol. 2001-63).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Integrating Economic Knowledge in Data Mining Algorithms

AU - Daniëls, H.A.M.

AU - Feelders, A.J.

N1 - Pagination: 16

PY - 2001

Y1 - 2001

N2 - The assessment of knowledge derived from databases depends on many factors. Decision makers often need to convince others about the correctness and effectiveness of knowledge induced from data.The current data mining techniques do not contribute much to this process of persuasion.Part of this limitation can be removed by integrating knowledge from experts in the field, encoded in some accessible way, with knowledge derived form patterns stored in the database.In this paper we will in particular discuss methods for implementing monotonicity constraints in economic decision problems.This prior knowledge is combined with data mining algorithms based on decision trees and neural networks.The method is illustrated in a hedonic price model.

AB - The assessment of knowledge derived from databases depends on many factors. Decision makers often need to convince others about the correctness and effectiveness of knowledge induced from data.The current data mining techniques do not contribute much to this process of persuasion.Part of this limitation can be removed by integrating knowledge from experts in the field, encoded in some accessible way, with knowledge derived form patterns stored in the database.In this paper we will in particular discuss methods for implementing monotonicity constraints in economic decision problems.This prior knowledge is combined with data mining algorithms based on decision trees and neural networks.The method is illustrated in a hedonic price model.

KW - knowledge

KW - neural network

KW - data mining

KW - decision trees

M3 - Discussion paper

VL - 2001-63

T3 - CentER Discussion Paper

BT - Integrating Economic Knowledge in Data Mining Algorithms

PB - Operations research

CY - Tilburg

ER -

Daniëls HAM, Feelders AJ. Integrating Economic Knowledge in Data Mining Algorithms. Tilburg: Operations research. 2001. (CentER Discussion Paper).