Anomalies in multidimensional contexts

Neil Dunstan; Ioan Despi; Charles R Watson

Anomalies in multidimensional contexts

Neil Dunstan
, Ioan Despi
, Charles R Watson

Computer Science

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

Abstract

This paper investigates the problem of presenting anomalies in a multidimensional data set. In such a data set, some dimensions may be merely descriptive, while others represent measures and attribute values used to determine whether the data is anomalous. A data cube of the descriptive dimensions is used as a data structure to partition the data set into sub-groups at each note, or context. It is shown that it is possible for a datum to be anomalous in more than one context. Previous work has dealt with this problem by embedding exception indicators in the data cube. Since the data cube is potentially large and anomalies are rare, searching for anomalies is inconvenient. Instead, it is proposed to construct a report for each anomaly that shows its status in each possible context. This results in a direct presentation of anomalous data.

Original language	English
Title of host publication	Data Mining X: Data Mining, Detection and other Security Technologies - Proceedings of Data Mining 2009: the 10th International Conference on Data Mining, Detection, Protection and Security
Editors	A Zanasi, N F F Ebecken, C A Brebbia
Place of Publication	Southampton, United Kingdom
Publisher	WIT Press
Pages	173-182
ISBN (Print)	9781845641849
Publication status	Published - 2009
Event	Data Mining 2009: 10th International Conference on Data Mining, Detection, Protection and Security - Royal Mare Village, Crete, Greece Duration: 27 May 2009 → 29 May 2009

Conference

Conference	Data Mining 2009: 10th International Conference on Data Mining, Detection, Protection and Security
City	Crete, Greece
Period	27/05/09 → 29/05/09

Keywords

Pattern Recognition and Data Mining

Access to Document

https://hdl.handle.net/1959.11/5484

Handle.net

Persistent link

Cite this

Dunstan, N., Despi, I., & Watson, C. R. (2009). Anomalies in multidimensional contexts. In A. Zanasi, N. F F Ebecken, & C. A Brebbia (Eds.), Data Mining X: Data Mining, Detection and other Security Technologies - Proceedings of Data Mining 2009: the 10th International Conference on Data Mining, Detection, Protection and Security (pp. 173-182). WIT Press. https://hdl.handle.net/1959.11/5484

Dunstan, Neil ; Despi, Ioan ; Watson, Charles R. / Anomalies in multidimensional contexts. Data Mining X: Data Mining, Detection and other Security Technologies - Proceedings of Data Mining 2009: the 10th International Conference on Data Mining, Detection, Protection and Security. editor / A Zanasi ; N F F Ebecken ; C A Brebbia. Southampton, United Kingdom : WIT Press, 2009. pp. 173-182

@inproceedings{ba80864206f44a1cb6d37f9de6c3a2af,

title = "Anomalies in multidimensional contexts",

abstract = "This paper investigates the problem of presenting anomalies in a multidimensional data set. In such a data set, some dimensions may be merely descriptive, while others represent measures and attribute values used to determine whether the data is anomalous. A data cube of the descriptive dimensions is used as a data structure to partition the data set into sub-groups at each note, or context. It is shown that it is possible for a datum to be anomalous in more than one context. Previous work has dealt with this problem by embedding exception indicators in the data cube. Since the data cube is potentially large and anomalies are rare, searching for anomalies is inconvenient. Instead, it is proposed to construct a report for each anomaly that shows its status in each possible context. This results in a direct presentation of anomalous data.",

keywords = "Pattern Recognition and Data Mining",

author = "Neil Dunstan and Ioan Despi and Watson, \{Charles R\}",

year = "2009",

language = "English",

isbn = "9781845641849",

pages = "173--182",

editor = "A Zanasi and \{F F Ebecken\}, N and \{A Brebbia\}, C",

booktitle = "Data Mining X: Data Mining, Detection and other Security Technologies - Proceedings of Data Mining 2009: the 10th International Conference on Data Mining, Detection, Protection and Security",

publisher = "WIT Press",

note = "Data Mining 2009: 10th International Conference on Data Mining, Detection, Protection and Security ; Conference date: 27-05-2009 Through 29-05-2009",

}

Dunstan, N, Despi, I & Watson, CR 2009, Anomalies in multidimensional contexts. in A Zanasi, N F F Ebecken & C A Brebbia (eds), Data Mining X: Data Mining, Detection and other Security Technologies - Proceedings of Data Mining 2009: the 10th International Conference on Data Mining, Detection, Protection and Security. WIT Press, Southampton, United Kingdom, pp. 173-182, Data Mining 2009: 10th International Conference on Data Mining, Detection, Protection and Security, Crete, Greece, 27/05/09. <https://hdl.handle.net/1959.11/5484>

Anomalies in multidimensional contexts. / Dunstan, Neil; Despi, Ioan; Watson, Charles R.
Data Mining X: Data Mining, Detection and other Security Technologies - Proceedings of Data Mining 2009: the 10th International Conference on Data Mining, Detection, Protection and Security. ed. / A Zanasi; N F F Ebecken; C A Brebbia. Southampton, United Kingdom: WIT Press, 2009. p. 173-182.

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

TY - GEN

T1 - Anomalies in multidimensional contexts

AU - Dunstan, Neil

AU - Despi, Ioan

AU - Watson, Charles R

PY - 2009

Y1 - 2009

N2 - This paper investigates the problem of presenting anomalies in a multidimensional data set. In such a data set, some dimensions may be merely descriptive, while others represent measures and attribute values used to determine whether the data is anomalous. A data cube of the descriptive dimensions is used as a data structure to partition the data set into sub-groups at each note, or context. It is shown that it is possible for a datum to be anomalous in more than one context. Previous work has dealt with this problem by embedding exception indicators in the data cube. Since the data cube is potentially large and anomalies are rare, searching for anomalies is inconvenient. Instead, it is proposed to construct a report for each anomaly that shows its status in each possible context. This results in a direct presentation of anomalous data.

AB - This paper investigates the problem of presenting anomalies in a multidimensional data set. In such a data set, some dimensions may be merely descriptive, while others represent measures and attribute values used to determine whether the data is anomalous. A data cube of the descriptive dimensions is used as a data structure to partition the data set into sub-groups at each note, or context. It is shown that it is possible for a datum to be anomalous in more than one context. Previous work has dealt with this problem by embedding exception indicators in the data cube. Since the data cube is potentially large and anomalies are rare, searching for anomalies is inconvenient. Instead, it is proposed to construct a report for each anomaly that shows its status in each possible context. This results in a direct presentation of anomalous data.

KW - Pattern Recognition and Data Mining

UR - http://www.wessex.ac.uk/09-conferences/data-mining-2009.html

UR - http://trove.nla.gov.au/work/28703709

M3 - Conference contribution

SN - 9781845641849

SP - 173

EP - 182

BT - Data Mining X: Data Mining, Detection and other Security Technologies - Proceedings of Data Mining 2009: the 10th International Conference on Data Mining, Detection, Protection and Security

A2 - Zanasi, A

A2 - F F Ebecken, N

A2 - A Brebbia, C

PB - WIT Press

CY - Southampton, United Kingdom

T2 - Data Mining 2009: 10th International Conference on Data Mining, Detection, Protection and Security

Y2 - 27 May 2009 through 29 May 2009

ER -

Anomalies in multidimensional contexts

Abstract

Conference

Keywords

Access to Document

Handle.net

Fingerprint

Cite this