Identifying Relevant Formal Concepts through the Collapse Index

D Joseph; N Mehandjiev; Babis Theodoulidis; J Davies; I Thurlow

doi:10.1109/BigDataCongress.2015.37

Identifying Relevant Formal Concepts through the Collapse Index

D Joseph, N Mehandjiev, Babis Theodoulidis, J Davies, I Thurlow

Management Sciences

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

Abstract

In this paper we introduce the Collapse Index, a new measure of the relevance of individual formal concepts in a concept lattice, the application of which improves the performance of concept pruning and reduces the bias against "outlier" concepts. The measure determines the relevance of a formal concept in the lattice by calculating the minimum number of objects which need to be removed from the domain before the formal concept collapses. We demonstrate the effectiveness of the Collapse Index as a measure of pattern selection by comparing the collapse indices found in two datasets. We cover the case where the two datasets are disjoint and the case where one dataset is a subset of the other. Results are contrasted to those of the Stability Index measure.

Original language	English
Title of host publication	2015 4th International Congress on Big Data (BigData Congress)
Pages	207-214
Number of pages	8
DOIs	https://doi.org/10.1109/BigDataCongress.2015.37
Publication status	Published - Jun 2015
Event	2015 4th International Congress on Big Data - New York, United States Duration: 27 Jun 2015 → 2 Jul 2016

Conference

Conference	2015 4th International Congress on Big Data
Country/Territory	United States
City	New York
Period	27/06/15 → 2/07/16

Keywords

formal concept analysis
lattice theory
statistical analysis
collapse index
concept lattice
concept pruning
formal concept
outlier concept
pattern selection
stability index measure
Context
Generators
Indexes
Lattices
Motion pictures
Stability criteria
fca
lattice
pruning

Access to Document

10.1109/BigDataCongress.2015.37

http://www.ieeebigdata.org/

Cite this

@inproceedings{c734890a82c24640b627af622d2fd3c0,

title = "Identifying Relevant Formal Concepts through the Collapse Index",

abstract = "In this paper we introduce the Collapse Index, a new measure of the relevance of individual formal concepts in a concept lattice, the application of which improves the performance of concept pruning and reduces the bias against {"}outlier{"} concepts. The measure determines the relevance of a formal concept in the lattice by calculating the minimum number of objects which need to be removed from the domain before the formal concept collapses. We demonstrate the effectiveness of the Collapse Index as a measure of pattern selection by comparing the collapse indices found in two datasets. We cover the case where the two datasets are disjoint and the case where one dataset is a subset of the other. Results are contrasted to those of the Stability Index measure.",

keywords = "formal concept analysis, lattice theory, statistical analysis, collapse index, concept lattice, concept pruning, formal concept, outlier concept, pattern selection, stability index measure, Context, Generators, Indexes, Lattices, Motion pictures, Stability criteria, fca, lattice, pruning",

author = "D Joseph and N Mehandjiev and Babis Theodoulidis and J Davies and I Thurlow",

year = "2015",

month = jun,

doi = "10.1109/BigDataCongress.2015.37",

language = "English",

pages = "207--214",

booktitle = "2015 4th International Congress on Big Data (BigData Congress)",

note = "2015 4th International Congress on Big Data ; Conference date: 27-06-2015 Through 02-07-2016",

}

TY - GEN

T1 - Identifying Relevant Formal Concepts through the Collapse Index

AU - Joseph, D

AU - Mehandjiev, N

AU - Theodoulidis, Babis

AU - Davies, J

AU - Thurlow, I

PY - 2015/6

Y1 - 2015/6

N2 - In this paper we introduce the Collapse Index, a new measure of the relevance of individual formal concepts in a concept lattice, the application of which improves the performance of concept pruning and reduces the bias against "outlier" concepts. The measure determines the relevance of a formal concept in the lattice by calculating the minimum number of objects which need to be removed from the domain before the formal concept collapses. We demonstrate the effectiveness of the Collapse Index as a measure of pattern selection by comparing the collapse indices found in two datasets. We cover the case where the two datasets are disjoint and the case where one dataset is a subset of the other. Results are contrasted to those of the Stability Index measure.

AB - In this paper we introduce the Collapse Index, a new measure of the relevance of individual formal concepts in a concept lattice, the application of which improves the performance of concept pruning and reduces the bias against "outlier" concepts. The measure determines the relevance of a formal concept in the lattice by calculating the minimum number of objects which need to be removed from the domain before the formal concept collapses. We demonstrate the effectiveness of the Collapse Index as a measure of pattern selection by comparing the collapse indices found in two datasets. We cover the case where the two datasets are disjoint and the case where one dataset is a subset of the other. Results are contrasted to those of the Stability Index measure.

KW - formal concept analysis

KW - lattice theory

KW - statistical analysis

KW - collapse index

KW - concept lattice

KW - concept pruning

KW - formal concept

KW - outlier concept

KW - pattern selection

KW - stability index measure

KW - Context

KW - Generators

KW - Indexes

KW - Lattices

KW - Motion pictures

KW - Stability criteria

KW - fca

KW - lattice

KW - pruning

U2 - 10.1109/BigDataCongress.2015.37

DO - 10.1109/BigDataCongress.2015.37

M3 - Conference contribution

SP - 207

EP - 214

BT - 2015 4th International Congress on Big Data (BigData Congress)

T2 - 2015 4th International Congress on Big Data

Y2 - 27 June 2015 through 2 July 2016

ER -

Identifying Relevant Formal Concepts through the Collapse Index

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this