A design methodology for distributed adaptive stream mining systems

Stephen Won, Inkeun Cho, Kishan Sudusinghe, Jie Xu, Yu Zhang, Mihaela Van Der Schaar, Shuvra S. Bhattacharyya

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

Abstract

Data-driven, adaptive computations are key to enabling the deployment of accurate and efficient stream mining systems, which invoke suitably configured queries in real-time on streams of input data. Due to the physical separation among data sources and computational resources, it is often necessary to deploy such stream mining systems in a distributed fashion, where local learners have access to disjoint subsets of the data that is to be mined, and forward their intermediate results to an ensemble learner that combines the results from the local learners. In this paper, we develop a design methodology for integrated design, simulation, and implementation of dynamic data-driven adaptive stream mining systems. By systematically integrating considerations associated with local embedded processing, classifier configuration, data-driven adaptation and networked communication, our approach allows for effective assessment, prototyping, and implementation of alternative distributed design methods for data-driven, adaptive stream mining systems. We demonstrate our results on a dynamic data-driven application involving patient health care monitoring.

Original languageEnglish (US)
Title of host publicationProcedia Computer Science
PublisherElsevier
Pages2482-2491
Number of pages10
Volume18
DOIs
StatePublished - 2013
Externally publishedYes
Event13th Annual International Conference on Computational Science, ICCS 2013 - Barcelona, Spain
Duration: Jun 5 2013Jun 7 2013

Other

Other13th Annual International Conference on Computational Science, ICCS 2013
CountrySpain
CityBarcelona
Period6/5/136/7/13

Fingerprint

Health care
Classifiers
Monitoring
Communication
Processing

Keywords

  • Adaptive stream mining
  • Dataflow graphs
  • Distributed signal processing

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Won, S., Cho, I., Sudusinghe, K., Xu, J., Zhang, Y., Van Der Schaar, M., & Bhattacharyya, S. S. (2013). A design methodology for distributed adaptive stream mining systems. In Procedia Computer Science (Vol. 18, pp. 2482-2491). Elsevier. https://doi.org/10.1016/j.procs.2013.05.425

A design methodology for distributed adaptive stream mining systems. / Won, Stephen; Cho, Inkeun; Sudusinghe, Kishan; Xu, Jie; Zhang, Yu; Van Der Schaar, Mihaela; Bhattacharyya, Shuvra S.

Procedia Computer Science. Vol. 18 Elsevier, 2013. p. 2482-2491.

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

Won, S, Cho, I, Sudusinghe, K, Xu, J, Zhang, Y, Van Der Schaar, M & Bhattacharyya, SS 2013, A design methodology for distributed adaptive stream mining systems. in Procedia Computer Science. vol. 18, Elsevier, pp. 2482-2491, 13th Annual International Conference on Computational Science, ICCS 2013, Barcelona, Spain, 6/5/13. https://doi.org/10.1016/j.procs.2013.05.425
Won S, Cho I, Sudusinghe K, Xu J, Zhang Y, Van Der Schaar M et al. A design methodology for distributed adaptive stream mining systems. In Procedia Computer Science. Vol. 18. Elsevier. 2013. p. 2482-2491 https://doi.org/10.1016/j.procs.2013.05.425
Won, Stephen ; Cho, Inkeun ; Sudusinghe, Kishan ; Xu, Jie ; Zhang, Yu ; Van Der Schaar, Mihaela ; Bhattacharyya, Shuvra S. / A design methodology for distributed adaptive stream mining systems. Procedia Computer Science. Vol. 18 Elsevier, 2013. pp. 2482-2491
@inproceedings{22157943e30b4ee090791baa322099ed,
title = "A design methodology for distributed adaptive stream mining systems",
abstract = "Data-driven, adaptive computations are key to enabling the deployment of accurate and efficient stream mining systems, which invoke suitably configured queries in real-time on streams of input data. Due to the physical separation among data sources and computational resources, it is often necessary to deploy such stream mining systems in a distributed fashion, where local learners have access to disjoint subsets of the data that is to be mined, and forward their intermediate results to an ensemble learner that combines the results from the local learners. In this paper, we develop a design methodology for integrated design, simulation, and implementation of dynamic data-driven adaptive stream mining systems. By systematically integrating considerations associated with local embedded processing, classifier configuration, data-driven adaptation and networked communication, our approach allows for effective assessment, prototyping, and implementation of alternative distributed design methods for data-driven, adaptive stream mining systems. We demonstrate our results on a dynamic data-driven application involving patient health care monitoring.",
keywords = "Adaptive stream mining, Dataflow graphs, Distributed signal processing",
author = "Stephen Won and Inkeun Cho and Kishan Sudusinghe and Jie Xu and Yu Zhang and {Van Der Schaar}, Mihaela and Bhattacharyya, {Shuvra S.}",
year = "2013",
doi = "10.1016/j.procs.2013.05.425",
language = "English (US)",
volume = "18",
pages = "2482--2491",
booktitle = "Procedia Computer Science",
publisher = "Elsevier",

}

TY - GEN

T1 - A design methodology for distributed adaptive stream mining systems

AU - Won, Stephen

AU - Cho, Inkeun

AU - Sudusinghe, Kishan

AU - Xu, Jie

AU - Zhang, Yu

AU - Van Der Schaar, Mihaela

AU - Bhattacharyya, Shuvra S.

PY - 2013

Y1 - 2013

N2 - Data-driven, adaptive computations are key to enabling the deployment of accurate and efficient stream mining systems, which invoke suitably configured queries in real-time on streams of input data. Due to the physical separation among data sources and computational resources, it is often necessary to deploy such stream mining systems in a distributed fashion, where local learners have access to disjoint subsets of the data that is to be mined, and forward their intermediate results to an ensemble learner that combines the results from the local learners. In this paper, we develop a design methodology for integrated design, simulation, and implementation of dynamic data-driven adaptive stream mining systems. By systematically integrating considerations associated with local embedded processing, classifier configuration, data-driven adaptation and networked communication, our approach allows for effective assessment, prototyping, and implementation of alternative distributed design methods for data-driven, adaptive stream mining systems. We demonstrate our results on a dynamic data-driven application involving patient health care monitoring.

AB - Data-driven, adaptive computations are key to enabling the deployment of accurate and efficient stream mining systems, which invoke suitably configured queries in real-time on streams of input data. Due to the physical separation among data sources and computational resources, it is often necessary to deploy such stream mining systems in a distributed fashion, where local learners have access to disjoint subsets of the data that is to be mined, and forward their intermediate results to an ensemble learner that combines the results from the local learners. In this paper, we develop a design methodology for integrated design, simulation, and implementation of dynamic data-driven adaptive stream mining systems. By systematically integrating considerations associated with local embedded processing, classifier configuration, data-driven adaptation and networked communication, our approach allows for effective assessment, prototyping, and implementation of alternative distributed design methods for data-driven, adaptive stream mining systems. We demonstrate our results on a dynamic data-driven application involving patient health care monitoring.

KW - Adaptive stream mining

KW - Dataflow graphs

KW - Distributed signal processing

UR - http://www.scopus.com/inward/record.url?scp=84897017590&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897017590&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2013.05.425

DO - 10.1016/j.procs.2013.05.425

M3 - Conference contribution

VL - 18

SP - 2482

EP - 2491

BT - Procedia Computer Science

PB - Elsevier

ER -