Efficiently mining approximate models of associations in evolving databases

Adriano Veloso, Bruno Gusmão, Wagner Meira, Marcio Carvalho, Srini Parthasarathy, Mohammed Zaki

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Much of the existing work in machine learning and data mining has relied on devising efficient techniques to build accurate models from the data. Research on how the accuracyof a model changes as a function of dynamic updates to the databases is very limited. In this work we show that extracting this information: knowing which aspects of the model are changing; and how theyare changing as a function of data updates; can be verye ffective for interactive data mining purposes (where response time is often more important than model qualityas long as model qualityi s not too far off the best (exact) model. In this paper we consider the problem of generating approximate models within the context of association mining, a keyda ta mining task. We propose a new approach to incrementallyg enerate approximate models of associations in evolving databases. Our approach is able to detect how patterns evolve over time (an interesting result in its own right), and uses this information in generating approximate models with high accuracy at a fraction of the cost (of generating the exact model). Extensive experimental evaluation on real databases demonstrates the effectiveness and advantages of the proposed approach.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages435-448
Number of pages14
Volume2431 LNAI
Publication statusPublished - 1 Dec 2002
Externally publishedYes
Event6th European Conference on Principles and Practice of Knowledge Discovery in Databases, PKDD 2002 - Helsinki, Finland
Duration: 19 Aug 200223 Aug 2002

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2431 LNAI
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other6th European Conference on Principles and Practice of Knowledge Discovery in Databases, PKDD 2002
CountryFinland
CityHelsinki
Period19/8/0223/8/02

Fingerprint

Approximate Model
Mining
Model
Data Mining
Update
Data mining
Experimental Evaluation
Response Time
Machine Learning
High Accuracy
Information use
Learning systems
Costs
Demonstrate

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Veloso, A., Gusmão, B., Meira, W., Carvalho, M., Parthasarathy, S., & Zaki, M. (2002). Efficiently mining approximate models of associations in evolving databases. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2431 LNAI, pp. 435-448). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2431 LNAI).

Efficiently mining approximate models of associations in evolving databases. / Veloso, Adriano; Gusmão, Bruno; Meira, Wagner; Carvalho, Marcio; Parthasarathy, Srini; Zaki, Mohammed.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2431 LNAI 2002. p. 435-448 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2431 LNAI).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Veloso, A, Gusmão, B, Meira, W, Carvalho, M, Parthasarathy, S & Zaki, M 2002, Efficiently mining approximate models of associations in evolving databases. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 2431 LNAI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2431 LNAI, pp. 435-448, 6th European Conference on Principles and Practice of Knowledge Discovery in Databases, PKDD 2002, Helsinki, Finland, 19/8/02.
Veloso A, Gusmão B, Meira W, Carvalho M, Parthasarathy S, Zaki M. Efficiently mining approximate models of associations in evolving databases. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2431 LNAI. 2002. p. 435-448. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Veloso, Adriano ; Gusmão, Bruno ; Meira, Wagner ; Carvalho, Marcio ; Parthasarathy, Srini ; Zaki, Mohammed. / Efficiently mining approximate models of associations in evolving databases. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2431 LNAI 2002. pp. 435-448 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{78f5c1697b934efa930ae380fe987318,
title = "Efficiently mining approximate models of associations in evolving databases",
abstract = "Much of the existing work in machine learning and data mining has relied on devising efficient techniques to build accurate models from the data. Research on how the accuracyof a model changes as a function of dynamic updates to the databases is very limited. In this work we show that extracting this information: knowing which aspects of the model are changing; and how theyare changing as a function of data updates; can be verye ffective for interactive data mining purposes (where response time is often more important than model qualityas long as model qualityi s not too far off the best (exact) model. In this paper we consider the problem of generating approximate models within the context of association mining, a keyda ta mining task. We propose a new approach to incrementallyg enerate approximate models of associations in evolving databases. Our approach is able to detect how patterns evolve over time (an interesting result in its own right), and uses this information in generating approximate models with high accuracy at a fraction of the cost (of generating the exact model). Extensive experimental evaluation on real databases demonstrates the effectiveness and advantages of the proposed approach.",
author = "Adriano Veloso and Bruno Gusm{\~a}o and Wagner Meira and Marcio Carvalho and Srini Parthasarathy and Mohammed Zaki",
year = "2002",
month = "12",
day = "1",
language = "English",
isbn = "3540440372",
volume = "2431 LNAI",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "435--448",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Efficiently mining approximate models of associations in evolving databases

AU - Veloso, Adriano

AU - Gusmão, Bruno

AU - Meira, Wagner

AU - Carvalho, Marcio

AU - Parthasarathy, Srini

AU - Zaki, Mohammed

PY - 2002/12/1

Y1 - 2002/12/1

N2 - Much of the existing work in machine learning and data mining has relied on devising efficient techniques to build accurate models from the data. Research on how the accuracyof a model changes as a function of dynamic updates to the databases is very limited. In this work we show that extracting this information: knowing which aspects of the model are changing; and how theyare changing as a function of data updates; can be verye ffective for interactive data mining purposes (where response time is often more important than model qualityas long as model qualityi s not too far off the best (exact) model. In this paper we consider the problem of generating approximate models within the context of association mining, a keyda ta mining task. We propose a new approach to incrementallyg enerate approximate models of associations in evolving databases. Our approach is able to detect how patterns evolve over time (an interesting result in its own right), and uses this information in generating approximate models with high accuracy at a fraction of the cost (of generating the exact model). Extensive experimental evaluation on real databases demonstrates the effectiveness and advantages of the proposed approach.

AB - Much of the existing work in machine learning and data mining has relied on devising efficient techniques to build accurate models from the data. Research on how the accuracyof a model changes as a function of dynamic updates to the databases is very limited. In this work we show that extracting this information: knowing which aspects of the model are changing; and how theyare changing as a function of data updates; can be verye ffective for interactive data mining purposes (where response time is often more important than model qualityas long as model qualityi s not too far off the best (exact) model. In this paper we consider the problem of generating approximate models within the context of association mining, a keyda ta mining task. We propose a new approach to incrementallyg enerate approximate models of associations in evolving databases. Our approach is able to detect how patterns evolve over time (an interesting result in its own right), and uses this information in generating approximate models with high accuracy at a fraction of the cost (of generating the exact model). Extensive experimental evaluation on real databases demonstrates the effectiveness and advantages of the proposed approach.

UR - http://www.scopus.com/inward/record.url?scp=84864840575&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864840575&partnerID=8YFLogxK

M3 - Conference contribution

SN - 3540440372

SN - 9783540440376

VL - 2431 LNAI

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 435

EP - 448

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -