Data mining algorithms for risk detection in bank loans

Alvaro Talavera; Luis Cano; David Paredes; Mario Chong

doi:10.1007/978-3-030-11680-4_16

Data mining algorithms for risk detection in bank loans

Alvaro Talavera, Luis Cano, David Paredes, Mario Chong

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

Abstract

This article proposes a new approach on detection of fraudulent credit operations applying computational intelligence techniques. We use a dataset of historical data of customers from a financial entity and we split it to train a classification and clustering algorithm. We train a radial basis function network to classify clients that commit or not credit fraud. Then, we build a Fuzzy c-means clustering to group data points to create customer profiles. This algorithm has the capacity of grouping the data inside clusters and assigning a degree of membership to the points outside the clusters. Subsequently, the trained classification algorithm is applied to the clusters to provide additional information about customer profiles. We demonstrate good performance for fraudulent credit operations and identification of customer profiles.

Original language	English
Title of host publication	Information Management and Big Data - 5th International Conference, SIMBig 2018, Proceedings
Editors	Hugo Alatrista-Salas, Denisse Muñante, Juan Antonio Lossio-Ventura
Pages	151-159
Number of pages	9
ISBN (Electronic)	9783030116798
DOIs	https://doi.org/10.1007/978-3-030-11680-4_16
State	Published - 1 Jan 2019
Event	Communications in Computer and Information Science - Duration: 1 Jan 2019 → …

Publication series

Name	Communications in Computer and Information Science
Volume	898
ISSN (Print)	1865-0929

Conference

Conference	Communications in Computer and Information Science
Period	1/01/19 → …

Bibliographical note

Publisher Copyright:
© 2019, Springer Nature Switzerland AG.

Keywords

Finance profiles
Fuzzy C-means
Radial basis function networks
Risk detection

Access to Document

10.1007/978-3-030-11680-4_16

Cite this

Talavera, A., Cano, L., Paredes, D., & Chong, M. (2019). Data mining algorithms for risk detection in bank loans. In H. Alatrista-Salas, D. Muñante, & J. A. Lossio-Ventura (Eds.), Information Management and Big Data - 5th International Conference, SIMBig 2018, Proceedings (pp. 151-159). (Communications in Computer and Information Science; Vol. 898). https://doi.org/10.1007/978-3-030-11680-4_16

@inproceedings{360840c78f374274adc0f6f29004012b,

title = "Data mining algorithms for risk detection in bank loans",

abstract = "This article proposes a new approach on detection of fraudulent credit operations applying computational intelligence techniques. We use a dataset of historical data of customers from a financial entity and we split it to train a classification and clustering algorithm. We train a radial basis function network to classify clients that commit or not credit fraud. Then, we build a Fuzzy c-means clustering to group data points to create customer profiles. This algorithm has the capacity of grouping the data inside clusters and assigning a degree of membership to the points outside the clusters. Subsequently, the trained classification algorithm is applied to the clusters to provide additional information about customer profiles. We demonstrate good performance for fraudulent credit operations and identification of customer profiles.",

keywords = "Finance profiles, Fuzzy C-means, Radial basis function networks, Risk detection, Finance profiles, Fuzzy C-means, Radial basis function networks, Risk detection",

author = "Alvaro Talavera and Luis Cano and David Paredes and Mario Chong",

note = "Publisher Copyright: {\textcopyright} 2019, Springer Nature Switzerland AG.; Communications in Computer and Information Science ; Conference date: 01-01-2019",

year = "2019",

month = jan,

day = "1",

doi = "10.1007/978-3-030-11680-4_16",

language = "English",

isbn = "9783030116798",

series = "Communications in Computer and Information Science",

pages = "151--159",

editor = "Hugo Alatrista-Salas and Denisse Mu{\~n}ante and Lossio-Ventura, {Juan Antonio}",

booktitle = "Information Management and Big Data - 5th International Conference, SIMBig 2018, Proceedings",

}

Talavera, A, Cano, L, Paredes, D & Chong, M 2019, Data mining algorithms for risk detection in bank loans. in H Alatrista-Salas, D Muñante & JA Lossio-Ventura (eds), Information Management and Big Data - 5th International Conference, SIMBig 2018, Proceedings. Communications in Computer and Information Science, vol. 898, pp. 151-159, Communications in Computer and Information Science, 1/01/19. https://doi.org/10.1007/978-3-030-11680-4_16

Data mining algorithms for risk detection in bank loans. / Talavera, Alvaro; Cano, Luis; Paredes, David et al.
Information Management and Big Data - 5th International Conference, SIMBig 2018, Proceedings. ed. / Hugo Alatrista-Salas; Denisse Muñante; Juan Antonio Lossio-Ventura. 2019. p. 151-159 (Communications in Computer and Information Science; Vol. 898).

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

TY - GEN

T1 - Data mining algorithms for risk detection in bank loans

AU - Talavera, Alvaro

AU - Cano, Luis

AU - Paredes, David

AU - Chong, Mario

PY - 2019/1/1

Y1 - 2019/1/1

N2 - This article proposes a new approach on detection of fraudulent credit operations applying computational intelligence techniques. We use a dataset of historical data of customers from a financial entity and we split it to train a classification and clustering algorithm. We train a radial basis function network to classify clients that commit or not credit fraud. Then, we build a Fuzzy c-means clustering to group data points to create customer profiles. This algorithm has the capacity of grouping the data inside clusters and assigning a degree of membership to the points outside the clusters. Subsequently, the trained classification algorithm is applied to the clusters to provide additional information about customer profiles. We demonstrate good performance for fraudulent credit operations and identification of customer profiles.

AB - This article proposes a new approach on detection of fraudulent credit operations applying computational intelligence techniques. We use a dataset of historical data of customers from a financial entity and we split it to train a classification and clustering algorithm. We train a radial basis function network to classify clients that commit or not credit fraud. Then, we build a Fuzzy c-means clustering to group data points to create customer profiles. This algorithm has the capacity of grouping the data inside clusters and assigning a degree of membership to the points outside the clusters. Subsequently, the trained classification algorithm is applied to the clusters to provide additional information about customer profiles. We demonstrate good performance for fraudulent credit operations and identification of customer profiles.

KW - Finance profiles

KW - Fuzzy C-means

KW - Radial basis function networks

KW - Risk detection

KW - Finance profiles

KW - Fuzzy C-means

KW - Radial basis function networks

KW - Risk detection

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85063510486&origin=inward

UR - http://www.scopus.com/inward/record.url?scp=85063510486&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/c8bc9a0c-e1ed-3d2f-afb2-6a253ed6ca17/

U2 - 10.1007/978-3-030-11680-4_16

DO - 10.1007/978-3-030-11680-4_16

M3 - Conference contribution

SN - 9783030116798

T3 - Communications in Computer and Information Science

SP - 151

EP - 159

BT - Information Management and Big Data - 5th International Conference, SIMBig 2018, Proceedings

A2 - Alatrista-Salas, Hugo

A2 - Muñante, Denisse

A2 - Lossio-Ventura, Juan Antonio

T2 - Communications in Computer and Information Science

Y2 - 1 January 2019

ER -

Data mining algorithms for risk detection in bank loans

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this