Challenges Faced by Industries and their Potential Solutions in Deploying Machine Learning Applications

Raj M Shukla; John P Cartlidge

doi:10.1109/CCWC54503.2022.9720900

Challenges Faced by Industries and their Potential Solutions in Deploying Machine Learning Applications

Raj M Shukla^*, John P Cartlidge

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

2 Citations (Scopus)

205 Downloads (Pure)

Abstract

Across all sectors, organizations attempt to make efficiency savings and performance improvements by incorporating machine learning (ML) into commercial application services. However, in comparison to traditional software applications, design, deployment, and maintenance of ML applications is more complicated. In particular, ML introduces new challenges of data availability, concept drift, scalability, and technical debt. In this paper, we introduce some of the practical challenges that arise when deploying ML applications, and describe potential solutions. Our analysis is based on experience designing and deploying a commercial spend classification service.

Original language	English
Title of host publication	2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022
Editors	Rajashree Paul
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	119-124
Number of pages	6
ISBN (Electronic)	978-1-6654-8303-2
ISBN (Print)	978-1-6654-8304-9
DOIs	https://doi.org/10.1109/CCWC54503.2022.9720900
Publication status	Published - 4 Mar 2022
Event	IEEE Annual Computing and Communication Workshop and Conference - Virtual, United States Duration: 26 Jan 2022 → 29 Jan 2022 Conference number: 12

Publication series

Name	2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022

Conference

Conference	IEEE Annual Computing and Communication Workshop and Conference
Abbreviated title	CCWC
Country/Territory	United States
Period	26/01/22 → 29/01/22

Bibliographical note

Funding Information:
This work was supported by Innovate UK Knowledge Transfer Partnership between University of Bristol and Claritum Limited (KTP 11952).

Publisher Copyright:
© 2022 IEEE.

Keywords

Industries
Conferences
Scalability
Machine learning
Organizations
Computer architecture
Maintenance engineering
Spend analysis
ML service
Application deployment

Access to Document

10.1109/CCWC54503.2022.9720900

Full-text PDF (accepted author manuscript)
This is the accepted author manuscript (AAM). The final published version (version of record) is available online via IEEE at 10.1109/CCWC54503.2022.9720900. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 960 KB

Handle.net

Persistent link

Claritum KTP - 11952
Cartlidge, J. (Principal Investigator)
22/08/19 → 31/01/23
Project: Research

Cite this

Shukla, R. M., & Cartlidge, J. P. (2022). Challenges Faced by Industries and their Potential Solutions in Deploying Machine Learning Applications. In R. Paul (Ed.), 2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022 (pp. 119-124). (2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/CCWC54503.2022.9720900

Shukla, Raj M ; Cartlidge, John P. / Challenges Faced by Industries and their Potential Solutions in Deploying Machine Learning Applications. 2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022. editor / Rajashree Paul. Institute of Electrical and Electronics Engineers (IEEE), 2022. pp. 119-124 (2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022).

@inproceedings{4e5928a7440f4d5ea1a2c5057d2e69f4,

title = "Challenges Faced by Industries and their Potential Solutions in Deploying Machine Learning Applications",

abstract = "Across all sectors, organizations attempt to make efficiency savings and performance improvements by incorporating machine learning (ML) into commercial application services. However, in comparison to traditional software applications, design, deployment, and maintenance of ML applications is more complicated. In particular, ML introduces new challenges of data availability, concept drift, scalability, and technical debt. In this paper, we introduce some of the practical challenges that arise when deploying ML applications, and describe potential solutions. Our analysis is based on experience designing and deploying a commercial spend classification service. ",

keywords = "Industries, Conferences, Scalability, Machine learning, Organizations, Computer architecture, Maintenance engineering, Spend analysis, ML service, Application deployment",

author = "Shukla, \{Raj M\} and Cartlidge, \{John P\}",

note = "Funding Information: This work was supported by Innovate UK Knowledge Transfer Partnership between University of Bristol and Claritum Limited (KTP 11952). Publisher Copyright: {\textcopyright} 2022 IEEE.; IEEE Annual Computing and Communication Workshop and Conference, CCWC ; Conference date: 26-01-2022 Through 29-01-2022",

year = "2022",

month = mar,

day = "4",

doi = "10.1109/CCWC54503.2022.9720900",

language = "English",

isbn = "978-1-6654-8304-9",

series = "2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

pages = "119--124",

editor = "Rajashree Paul",

booktitle = "2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022",

address = "United States",

}

Shukla, RM & Cartlidge, JP 2022, Challenges Faced by Industries and their Potential Solutions in Deploying Machine Learning Applications. in R Paul (ed.), 2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022. 2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022, Institute of Electrical and Electronics Engineers (IEEE), pp. 119-124, IEEE Annual Computing and Communication Workshop and Conference, United States, 26/01/22. https://doi.org/10.1109/CCWC54503.2022.9720900

Challenges Faced by Industries and their Potential Solutions in Deploying Machine Learning Applications. / Shukla, Raj M; Cartlidge, John P.
2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022. ed. / Rajashree Paul. Institute of Electrical and Electronics Engineers (IEEE), 2022. p. 119-124 (2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022).

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Challenges Faced by Industries and their Potential Solutions in Deploying Machine Learning Applications

AU - Shukla, Raj M

AU - Cartlidge, John P

N1 - Conference code: 12

PY - 2022/3/4

Y1 - 2022/3/4

N2 - Across all sectors, organizations attempt to make efficiency savings and performance improvements by incorporating machine learning (ML) into commercial application services. However, in comparison to traditional software applications, design, deployment, and maintenance of ML applications is more complicated. In particular, ML introduces new challenges of data availability, concept drift, scalability, and technical debt. In this paper, we introduce some of the practical challenges that arise when deploying ML applications, and describe potential solutions. Our analysis is based on experience designing and deploying a commercial spend classification service.

AB - Across all sectors, organizations attempt to make efficiency savings and performance improvements by incorporating machine learning (ML) into commercial application services. However, in comparison to traditional software applications, design, deployment, and maintenance of ML applications is more complicated. In particular, ML introduces new challenges of data availability, concept drift, scalability, and technical debt. In this paper, we introduce some of the practical challenges that arise when deploying ML applications, and describe potential solutions. Our analysis is based on experience designing and deploying a commercial spend classification service.

KW - Industries

KW - Conferences

KW - Scalability

KW - Machine learning

KW - Organizations

KW - Computer architecture

KW - Maintenance engineering

KW - Spend analysis

KW - ML service

KW - Application deployment

U2 - 10.1109/CCWC54503.2022.9720900

DO - 10.1109/CCWC54503.2022.9720900

M3 - Conference Contribution (Conference Proceeding)

SN - 978-1-6654-8304-9

T3 - 2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022

SP - 119

EP - 124

BT - 2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022

A2 - Paul, Rajashree

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - IEEE Annual Computing and Communication Workshop and Conference

Y2 - 26 January 2022 through 29 January 2022

ER -

Shukla RM, Cartlidge JP. Challenges Faced by Industries and their Potential Solutions in Deploying Machine Learning Applications. In Paul R, editor, 2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022. Institute of Electrical and Electronics Engineers (IEEE). 2022. p. 119-124. (2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022). doi: 10.1109/CCWC54503.2022.9720900

Challenges Faced by Industries and their Potential Solutions in Deploying Machine Learning Applications

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Handle.net

Fingerprint

Projects

Claritum KTP - 11952

Cite this