Biased logistic models applied to cervical cancer risk

José Antonio Cárdenas Garro; José Antonio Taquía Gutiérrez

Biased logistic models applied to cervical cancer risk

José Antonio Cárdenas Garro
, José Antonio Taquía Gutiérrez

Industrial Engineering School

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

76 Downloads (Pure)

Abstract

In the present work, the CRISP-DM methodology was proposed to develop a set of machine learning models applied
to evaluate cervical cancer risk suffer. For this research, a sample of 858 patients was taken, who were asked a series
of questions regarding this pathology. The database has an unbalanced dependent variable, since this is a health study,
the balancing technique will not be used to identify the variables that will enter the model, the Boruta library was used
for variable selection. For the development, five algorithms will be used: Support Vector Machine (SVM), decision
trees using the CHAID and CART algorithms, logistic regression and "asymmetric link" models. The models proposed
in this work were refined by means of the Auc, Gini, Log loss and KS (Kolmogorov-Smirnov) indicators, as a result
using the proposed models, AUC values of 98% were obtained.

Translated title of the contribution	Modelos de clasificación logística desbalanceada aplicada a la detección de cancer cervical.
Original language	English
Title of host publication	Biased logistic models applied to cervical cancer risk
Place of Publication	United States
Publisher	IEOM Society International
Chapter	1
Number of pages	7
ISBN (Electronic)	978-1-7923-9159-0
State	Published - 28 Nov 2022
Event	Proceedings of the 3rd South American International Industrial Engineering and Operations Management Conference, Asuncion, Paraguay, July 19-21, 2022 - Universidad Nacional de Asunción - Paraguay, Asunción, Paraguay Duration: 19 Jul 2022 → 21 Jul 2022 Conference number: 3 http://ieomsociety.org/paraguay2022/proceedings/

Conference

Conference	Proceedings of the 3rd South American International Industrial Engineering and Operations Management Conference, Asuncion, Paraguay, July 19-21, 2022
Abbreviated title	IEOM Paraguay 2022
Country/Territory	Paraguay
City	Asunción
Period	19/07/22 → 21/07/22
Internet address	http://ieomsociety.org/paraguay2022/proceedings/

UN SDGs

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

SDG 3 Good Health and Well-being

OECD Category

Otras ingenierías y tecnologías

Access to Document

Biased logistic models applied to cervical cancer risk Final published version, 393 KBLicense: CC BY

Cite this

@inproceedings{94c1b674eb5d4175befcc3900c04a5c4,

title = "Biased logistic models applied to cervical cancer risk",

abstract = "In the present work, the CRISP-DM methodology was proposed to develop a set of machine learning models applied to evaluate cervical cancer risk suffer. For this research, a sample of 858 patients was taken, who were asked a series of questions regarding this pathology. The database has an unbalanced dependent variable, since this is a health study, the balancing technique will not be used to identify the variables that will enter the model, the Boruta library was used for variable selection. For the development, five algorithms will be used: Support Vector Machine (SVM), decision trees using the CHAID and CART algorithms, logistic regression and {"}asymmetric link{"} models. The models proposed in this work were refined by means of the Auc, Gini, Log loss and KS (Kolmogorov-Smirnov) indicators, as a result using the proposed models, AUC values of 98\% were obtained. ",

keywords = "Machine learning, Asymmetric linkage models, Predictive models",

author = "\{C{\'a}rdenas Garro\}, \{Jos{\'e} Antonio\} and \{Taqu{\'i}a Guti{\'e}rrez\}, \{Jos{\'e} Antonio\}",

year = "2022",

month = nov,

day = "28",

language = "Ingl{\'e}s",

booktitle = "Biased logistic models applied to cervical cancer risk",

publisher = "IEOM Society International",

note = "Proceedings of the 3rd South American International Industrial Engineering and Operations Management<br/>Conference, Asuncion, Paraguay, July 19-21, 2022 ; Conference date: 19-07-2022 Through 21-07-2022",

url = "http://ieomsociety.org/paraguay2022/proceedings/",

}

Cárdenas Garro, JA & Taquía Gutiérrez, JA 2022, Biased logistic models applied to cervical cancer risk. in Biased logistic models applied to cervical cancer risk., 23, IEOM Society International, United States, Proceedings of the 3rd South American International Industrial Engineering and Operations Management
Conference, Asuncion, Paraguay, July 19-21, 2022, Asunción, Paraguay, 19/07/22.

TY - GEN

T1 - Biased logistic models applied to cervical cancer risk

AU - Cárdenas Garro, José Antonio

AU - Taquía Gutiérrez, José Antonio

N1 - Conference code: 3

PY - 2022/11/28

Y1 - 2022/11/28

N2 - In the present work, the CRISP-DM methodology was proposed to develop a set of machine learning models applied to evaluate cervical cancer risk suffer. For this research, a sample of 858 patients was taken, who were asked a series of questions regarding this pathology. The database has an unbalanced dependent variable, since this is a health study, the balancing technique will not be used to identify the variables that will enter the model, the Boruta library was used for variable selection. For the development, five algorithms will be used: Support Vector Machine (SVM), decision trees using the CHAID and CART algorithms, logistic regression and "asymmetric link" models. The models proposed in this work were refined by means of the Auc, Gini, Log loss and KS (Kolmogorov-Smirnov) indicators, as a result using the proposed models, AUC values of 98% were obtained.

AB - In the present work, the CRISP-DM methodology was proposed to develop a set of machine learning models applied to evaluate cervical cancer risk suffer. For this research, a sample of 858 patients was taken, who were asked a series of questions regarding this pathology. The database has an unbalanced dependent variable, since this is a health study, the balancing technique will not be used to identify the variables that will enter the model, the Boruta library was used for variable selection. For the development, five algorithms will be used: Support Vector Machine (SVM), decision trees using the CHAID and CART algorithms, logistic regression and "asymmetric link" models. The models proposed in this work were refined by means of the Auc, Gini, Log loss and KS (Kolmogorov-Smirnov) indicators, as a result using the proposed models, AUC values of 98% were obtained.

KW - Machine learning

KW - Asymmetric linkage models

KW - Predictive models

UR - https://ieomsociety.org/proceedings/2022paraguay/23.pdf

M3 - Articulo (Contribución a conferencia)

BT - Biased logistic models applied to cervical cancer risk

PB - IEOM Society International

CY - United States

T2 - Proceedings of the 3rd South American International Industrial Engineering and Operations Management<br/>Conference, Asuncion, Paraguay, July 19-21, 2022

Y2 - 19 July 2022 through 21 July 2022

ER -

Biased logistic models applied to cervical cancer risk

Abstract

Conference

UN SDGs

OECD Category

Access to Document

Other files and links

PeerJ Computer Science (Journal)

PeerJ Computer Science (Journal)

Cite this

Biased logistic models applied to cervical cancer risk

Abstract

Conference

UN SDGs

OECD Category

Access to Document

Other files and links

Activities

PeerJ Computer Science (Journal)

PeerJ Computer Science (Journal)

Cite this