Mining knowledge at multiple concept levels

Research output: Contribution to conferencePaper

Abstract

Most studies on data mining have been focused at mining rules at single concept levels, i.e., either at the primitive level or at a rather high concept level. However, it is often desirable to discover knowledge at multiple concept levels. Mining knowledge at multiple levels may help database users find some interesting rules which are difficult to be discovered otherwise and view database contents at different abstraction levels and from different angles. Methods for mining knowledge at multiple concept levels can often be developed by extension of existing data mining techniques. Moreover, for efficient processing and interactive mining of multiple-level rules, it is often necessary to adopt techniques such as step-by-step generalization/specialization or progressive deepening of a knowledge mining process. Other issues, such as visual representation of knowledge at multiple levels, and 'redundant' rule filtering, should also be studied in depth.

Original languageEnglish (US)
Pages19-24
Number of pages6
StatePublished - Dec 1 1995
Externally publishedYes
EventProceedings of the 1995 ACM CIKM 4th International Conference on Information and Knowledge Management - Baltimore, MD, USA
Duration: Nov 28 1995Dec 2 1995

Other

OtherProceedings of the 1995 ACM CIKM 4th International Conference on Information and Knowledge Management
CityBaltimore, MD, USA
Period11/28/9512/2/95

Fingerprint

Data mining
Data base
Process mining

ASJC Scopus subject areas

  • Decision Sciences(all)
  • Business, Management and Accounting(all)

Cite this

Han, J. (1995). Mining knowledge at multiple concept levels. 19-24. Paper presented at Proceedings of the 1995 ACM CIKM 4th International Conference on Information and Knowledge Management, Baltimore, MD, USA, .

Mining knowledge at multiple concept levels. / Han, Jiawei.

1995. 19-24 Paper presented at Proceedings of the 1995 ACM CIKM 4th International Conference on Information and Knowledge Management, Baltimore, MD, USA, .

Research output: Contribution to conferencePaper

Han, J 1995, 'Mining knowledge at multiple concept levels' Paper presented at Proceedings of the 1995 ACM CIKM 4th International Conference on Information and Knowledge Management, Baltimore, MD, USA, 11/28/95 - 12/2/95, pp. 19-24.
Han J. Mining knowledge at multiple concept levels. 1995. Paper presented at Proceedings of the 1995 ACM CIKM 4th International Conference on Information and Knowledge Management, Baltimore, MD, USA, .
Han, Jiawei. / Mining knowledge at multiple concept levels. Paper presented at Proceedings of the 1995 ACM CIKM 4th International Conference on Information and Knowledge Management, Baltimore, MD, USA, .6 p.
@conference{ad424c4b8ea9471c89dac12f805182bc,
title = "Mining knowledge at multiple concept levels",
abstract = "Most studies on data mining have been focused at mining rules at single concept levels, i.e., either at the primitive level or at a rather high concept level. However, it is often desirable to discover knowledge at multiple concept levels. Mining knowledge at multiple levels may help database users find some interesting rules which are difficult to be discovered otherwise and view database contents at different abstraction levels and from different angles. Methods for mining knowledge at multiple concept levels can often be developed by extension of existing data mining techniques. Moreover, for efficient processing and interactive mining of multiple-level rules, it is often necessary to adopt techniques such as step-by-step generalization/specialization or progressive deepening of a knowledge mining process. Other issues, such as visual representation of knowledge at multiple levels, and 'redundant' rule filtering, should also be studied in depth.",
author = "Jiawei Han",
year = "1995",
month = "12",
day = "1",
language = "English (US)",
pages = "19--24",
note = "Proceedings of the 1995 ACM CIKM 4th International Conference on Information and Knowledge Management ; Conference date: 28-11-1995 Through 02-12-1995",

}

TY - CONF

T1 - Mining knowledge at multiple concept levels

AU - Han, Jiawei

PY - 1995/12/1

Y1 - 1995/12/1

N2 - Most studies on data mining have been focused at mining rules at single concept levels, i.e., either at the primitive level or at a rather high concept level. However, it is often desirable to discover knowledge at multiple concept levels. Mining knowledge at multiple levels may help database users find some interesting rules which are difficult to be discovered otherwise and view database contents at different abstraction levels and from different angles. Methods for mining knowledge at multiple concept levels can often be developed by extension of existing data mining techniques. Moreover, for efficient processing and interactive mining of multiple-level rules, it is often necessary to adopt techniques such as step-by-step generalization/specialization or progressive deepening of a knowledge mining process. Other issues, such as visual representation of knowledge at multiple levels, and 'redundant' rule filtering, should also be studied in depth.

AB - Most studies on data mining have been focused at mining rules at single concept levels, i.e., either at the primitive level or at a rather high concept level. However, it is often desirable to discover knowledge at multiple concept levels. Mining knowledge at multiple levels may help database users find some interesting rules which are difficult to be discovered otherwise and view database contents at different abstraction levels and from different angles. Methods for mining knowledge at multiple concept levels can often be developed by extension of existing data mining techniques. Moreover, for efficient processing and interactive mining of multiple-level rules, it is often necessary to adopt techniques such as step-by-step generalization/specialization or progressive deepening of a knowledge mining process. Other issues, such as visual representation of knowledge at multiple levels, and 'redundant' rule filtering, should also be studied in depth.

UR - http://www.scopus.com/inward/record.url?scp=0029457111&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029457111&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0029457111

SP - 19

EP - 24

ER -