Machine learning using synthetic and real data: Similarity of evaluation metrics for different healthcare datasets and for different algorithms

Rachael Heyburn; RR Bond; Michaela Black; Maurice Mulvenna; J. G. Wallace; Debbie Rankin; Brian Cleland

doi:DOI: 10.1142/9789813273238_0160

Machine learning using synthetic and real data: Similarity of evaluation metrics for different healthcare datasets and for different algorithms

Rachael Heyburn, RR Bond, Michaela Black, Maurice Mulvenna, J. G. Wallace, Debbie Rankin, Brian Cleland

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

1956 Downloads (Pure)

Abstract

Sharing data is often a risk in terms of security and privacy especially if the data is sensitive. Algorithms can be used to generate synthetic data from an original raw dataset in order to share data that are considered more ‘privacy preserving’, and that increase the level of anonymity. In this paper, we carry out an experiment to study the validity of conducting machine learning on synthetic data. We compare the evaluation metrics produced from machine learning models that were trained using synthetic data with metrics yielded from machine learning models that were trained using the corresponding real data.

Original language	English
Title of host publication	Data Science and Knowledge Engineering for Sensing Decision Support
Publisher	World Scientific Publishing
Pages	1281-1291
Number of pages	11
Volume	11
ISBN (Electronic)	978-981-3273-24-5
ISBN (Print)	978-981-3273-22-1
DOIs	https://doi.org/DOI: 10.1142/9789813273238_0160
Publication status	Published (in print/issue) - 24 Aug 2018

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

DOI: 10.1142/9789813273238_0160

Paper ID 72Author Accepted Manuscript, 1.02 MB

Michaela Black
- mm.blackulster.acuk
- School of Computing, Eng & Intel. Sys - Professor of Artificial Intelligence
- Faculty Of Computing, Eng. & Built Env. - Full Professor
Person: Academic
Debbie Rankin
- d.rankin1ulster.acuk
- School of Computing, Eng & Intel. Sys - Senior Lecturer in Computer Science
- Faculty Of Computing, Eng. & Built Env. - Senior Lecturer
Person: Academic

Cite this

Heyburn, R., Bond, RR., Black, M., Mulvenna, M., Wallace, J. G., Rankin, D., & Cleland, B. (2018). Machine learning using synthetic and real data: Similarity of evaluation metrics for different healthcare datasets and for different algorithms. In Data Science and Knowledge Engineering for Sensing Decision Support (Vol. 11, pp. 1281-1291). World Scientific Publishing. https://doi.org/DOI: 10.1142/9789813273238_0160

@inproceedings{d39c146e61a74dbc85fe942dd8c56532,

title = "Machine learning using synthetic and real data: Similarity of evaluation metrics for different healthcare datasets and for different algorithms",

abstract = "Sharing data is often a risk in terms of security and privacy especially if the data is sensitive. Algorithms can be used to generate synthetic data from an original raw dataset in order to share data that are considered more {\textquoteleft}privacy preserving{\textquoteright}, and that increase the level of anonymity. In this paper, we carry out an experiment to study the validity of conducting machine learning on synthetic data. We compare the evaluation metrics produced from machine learning models that were trained using synthetic data with metrics yielded from machine learning models that were trained using the corresponding real data.",

author = "Rachael Heyburn and RR Bond and Michaela Black and Maurice Mulvenna and Wallace, {J. G.} and Debbie Rankin and Brian Cleland",

year = "2018",

month = aug,

day = "24",

doi = "DOI: 10.1142/9789813273238_0160",

language = "English",

isbn = "978-981-3273-22-1 ",

volume = "11",

pages = "1281--1291",

booktitle = "Data Science and Knowledge Engineering for Sensing Decision Support",

publisher = "World Scientific Publishing",

address = "Singapore",

}

Heyburn, R, Bond, RR , Black, M , Mulvenna, M , Wallace, JG , Rankin, D & Cleland, B 2018, Machine learning using synthetic and real data: Similarity of evaluation metrics for different healthcare datasets and for different algorithms. in Data Science and Knowledge Engineering for Sensing Decision Support. vol. 11, World Scientific Publishing, pp. 1281-1291. https://doi.org/DOI: 10.1142/9789813273238_0160

Machine learning using synthetic and real data: Similarity of evaluation metrics for different healthcare datasets and for different algorithms. / Heyburn, Rachael; Bond, RR ; Black, Michaela et al.
Data Science and Knowledge Engineering for Sensing Decision Support. Vol. 11 World Scientific Publishing, 2018. p. 1281-1291.

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

TY - GEN

T1 - Machine learning using synthetic and real data: Similarity of evaluation metrics for different healthcare datasets and for different algorithms

AU - Heyburn, Rachael

AU - Bond, RR

AU - Black, Michaela

AU - Mulvenna, Maurice

AU - Wallace, J. G.

AU - Rankin, Debbie

AU - Cleland, Brian

PY - 2018/8/24

Y1 - 2018/8/24

N2 - Sharing data is often a risk in terms of security and privacy especially if the data is sensitive. Algorithms can be used to generate synthetic data from an original raw dataset in order to share data that are considered more ‘privacy preserving’, and that increase the level of anonymity. In this paper, we carry out an experiment to study the validity of conducting machine learning on synthetic data. We compare the evaluation metrics produced from machine learning models that were trained using synthetic data with metrics yielded from machine learning models that were trained using the corresponding real data.

AB - Sharing data is often a risk in terms of security and privacy especially if the data is sensitive. Algorithms can be used to generate synthetic data from an original raw dataset in order to share data that are considered more ‘privacy preserving’, and that increase the level of anonymity. In this paper, we carry out an experiment to study the validity of conducting machine learning on synthetic data. We compare the evaluation metrics produced from machine learning models that were trained using synthetic data with metrics yielded from machine learning models that were trained using the corresponding real data.

U2 - DOI: 10.1142/9789813273238_0160

DO - DOI: 10.1142/9789813273238_0160

M3 - Conference contribution

SN - 978-981-3273-22-1

VL - 11

SP - 1281

EP - 1291

BT - Data Science and Knowledge Engineering for Sensing Decision Support

PB - World Scientific Publishing

ER -

Heyburn R, Bond RR , Black M , Mulvenna M , Wallace JG , Rankin D et al. Machine learning using synthetic and real data: Similarity of evaluation metrics for different healthcare datasets and for different algorithms. In Data Science and Knowledge Engineering for Sensing Decision Support. Vol. 11. World Scientific Publishing. 2018. p. 1281-1291 doi: DOI: 10.1142/9789813273238_0160

Machine learning using synthetic and real data: Similarity of evaluation metrics for different healthcare datasets and for different algorithms

Abstract

UN SDGs

Access to Document

Fingerprint

Profiles

Michaela Black

Debbie Rankin

Cite this