Statistical hypothesis testing for chemical detection in changing environments

Anna Ladi; Jon Timmis; Andy M. Tyrrell; Peter J. Hickey

doi:10.1109/CIDUE.2014.7007870

Statistical hypothesis testing for chemical detection in changing environments

Anna Ladi, Jon Timmis, Andy M. Tyrrell, Peter J. Hickey

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference Proceeding (Non-Journal item)

7 Citations (Scopus)

Abstract

This paper addresses the problem of adaptive chemical detection, using the Receptor Density Algorithm (RDA), an immune inspired anomaly detection algorithm. Our approach is to first detect when and if something has changed in the environment and then adapt the RDA to this change. Statistical hypothesis testing is used to determine whether there has been concept drift in consecutive time windows of the data. Five different statistical methods are tested on mass spectrometry data, enhanced with artificial events that signify a changing environment. The results show that, while no one method is universally best, statistical hypothesis testing performs reasonably well on the context of chemical sensing and it can differentiate between anomalies and concept drift.

Original language	English
Title of host publication	IEEE SSCI 2014
Subtitle of host publication	2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings
Publisher	IEEE Press
Pages	77-84
Number of pages	8
ISBN (Electronic)	9781479945160
DOIs	https://doi.org/10.1109/CIDUE.2014.7007870
Publication status	Published - 12 Jan 2014
Event	2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, CIDUE 2014 - Orlando, United States of America Duration: 09 Dec 2014 → 12 Dec 2014

Publication series

Name	IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings

Conference

Conference	2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, CIDUE 2014
Country/Territory	United States of America
City	Orlando
Period	09 Dec 2014 → 12 Dec 2014

Access to Document

10.1109/CIDUE.2014.7007870

Permanent link

Persistent link

Cite this

Ladi, A., Timmis, J., Tyrrell, A. M., & Hickey, P. J. (2014). Statistical hypothesis testing for chemical detection in changing environments. In IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings (pp. 77-84). Article 7007870 (IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings). IEEE Press. https://doi.org/10.1109/CIDUE.2014.7007870

Ladi, Anna ; Timmis, Jon ; Tyrrell, Andy M. et al. / Statistical hypothesis testing for chemical detection in changing environments. IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings. IEEE Press, 2014. pp. 77-84 (IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings).

@inproceedings{b876d2797e1c47dfb4ec4af95721a30e,

title = "Statistical hypothesis testing for chemical detection in changing environments",

abstract = "This paper addresses the problem of adaptive chemical detection, using the Receptor Density Algorithm (RDA), an immune inspired anomaly detection algorithm. Our approach is to first detect when and if something has changed in the environment and then adapt the RDA to this change. Statistical hypothesis testing is used to determine whether there has been concept drift in consecutive time windows of the data. Five different statistical methods are tested on mass spectrometry data, enhanced with artificial events that signify a changing environment. The results show that, while no one method is universally best, statistical hypothesis testing performs reasonably well on the context of chemical sensing and it can differentiate between anomalies and concept drift.",

author = "Anna Ladi and Jon Timmis and Tyrrell, {Andy M.} and Hickey, {Peter J.}",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, CIDUE 2014 ; Conference date: 09-12-2014 Through 12-12-2014",

year = "2014",

month = jan,

day = "12",

doi = "10.1109/CIDUE.2014.7007870",

language = "English",

series = "IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings",

publisher = "IEEE Press",

pages = "77--84",

booktitle = "IEEE SSCI 2014",

address = "United States of America",

}

Ladi, A, Timmis, J, Tyrrell, AM & Hickey, PJ 2014, Statistical hypothesis testing for chemical detection in changing environments. in IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings., 7007870, IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings, IEEE Press, pp. 77-84, 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, CIDUE 2014, Orlando, United States of America, 09 Dec 2014. https://doi.org/10.1109/CIDUE.2014.7007870

Statistical hypothesis testing for chemical detection in changing environments. / Ladi, Anna; Timmis, Jon; Tyrrell, Andy M. et al.
IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings. IEEE Press, 2014. p. 77-84 7007870 (IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference Proceeding (Non-Journal item)

TY - GEN

T1 - Statistical hypothesis testing for chemical detection in changing environments

AU - Ladi, Anna

AU - Timmis, Jon

AU - Tyrrell, Andy M.

AU - Hickey, Peter J.

PY - 2014/1/12

Y1 - 2014/1/12

N2 - This paper addresses the problem of adaptive chemical detection, using the Receptor Density Algorithm (RDA), an immune inspired anomaly detection algorithm. Our approach is to first detect when and if something has changed in the environment and then adapt the RDA to this change. Statistical hypothesis testing is used to determine whether there has been concept drift in consecutive time windows of the data. Five different statistical methods are tested on mass spectrometry data, enhanced with artificial events that signify a changing environment. The results show that, while no one method is universally best, statistical hypothesis testing performs reasonably well on the context of chemical sensing and it can differentiate between anomalies and concept drift.

AB - This paper addresses the problem of adaptive chemical detection, using the Receptor Density Algorithm (RDA), an immune inspired anomaly detection algorithm. Our approach is to first detect when and if something has changed in the environment and then adapt the RDA to this change. Statistical hypothesis testing is used to determine whether there has been concept drift in consecutive time windows of the data. Five different statistical methods are tested on mass spectrometry data, enhanced with artificial events that signify a changing environment. The results show that, while no one method is universally best, statistical hypothesis testing performs reasonably well on the context of chemical sensing and it can differentiate between anomalies and concept drift.

UR - http://www.scopus.com/inward/record.url?scp=84922949107&partnerID=8YFLogxK

U2 - 10.1109/CIDUE.2014.7007870

DO - 10.1109/CIDUE.2014.7007870

M3 - Conference Proceeding (Non-Journal item)

AN - SCOPUS:84922949107

T3 - IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings

SP - 77

EP - 84

BT - IEEE SSCI 2014

PB - IEEE Press

T2 - 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, CIDUE 2014

Y2 - 9 December 2014 through 12 December 2014

ER -

Ladi A, Timmis J, Tyrrell AM, Hickey PJ. Statistical hypothesis testing for chemical detection in changing environments. In IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings. IEEE Press. 2014. p. 77-84. 7007870. (IEEE SSCI 2014: 2014 IEEE Symposium Series on Computational Intelligence - CIDUE 2014: 2014 IEEE Symposium on Computational Intelligence in Dynamic and Uncertain Environments, Proceedings). doi: 10.1109/CIDUE.2014.7007870

Statistical hypothesis testing for chemical detection in changing environments

Abstract

Publication series

Conference

Access to Document

Permanent link

Other files and links

Fingerprint

Cite this