Bankruptcy prediction of construction businesses: Towards a big data analytics approach

Hafiz Alaka; Lukumon O. Oyedele; Muhammad Bilal; Olugbenga O. Akinade; Hakeen A. Owolabi; Saheed O. Ajayi

doi:10.1109/BigDataService.2015.30

Bankruptcy prediction of construction businesses: Towards a big data analytics approach

Hafiz Alaka, Lukumon O. Oyedele, Muhammad Bilal, Olugbenga O. Akinade, Hakeen A. Owolabi, Saheed O. Ajayi

School of Energy, Construction and Environment

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

26 Citations (Scopus)

233 Downloads (Pure)

Abstract

Bankruptcy prediction models (BPMs) are needed by financiers like banks in order to check the credit worthiness of companies. A very robust model needs a very large amount of data with periodic updates (i.e. appending new data). Such size of data cannot be processed directly by the tools used in building BPMs, however Big Data Analytics offers the opportunity to analyse such data. With data sources like DataStream, FAME, Company House, etc. that hold large financial data of existing and failed firms, it is possible to extract huge financial data into Hadoop database (e.g. HBase), whilst allowing periodic appending of data from the data sources, and carry out a Big Data analysis using a machine learning tool on Apache Mahout. Lifelong machine learning can also be employed in order to avoid repeated intensive training of the model using all the data in the Hadoop database. A framework is thus proposed for developing a Big Data Analytics based BPM.

Original language	English
Title of host publication	2015 IEEE First International Conference on Big Data Computing Service and Applications (BigDataService)
Publisher	IEEE
Pages	347-352
Number of pages	6
ISBN (Electronic)	978-1-4799-8128-1
DOIs	https://doi.org/10.1109/BigDataService.2015.30
Publication status	Published - 13 Aug 2015
Event	2015 IEEE First International Conference on Big Data Computing Service and Applications (BigDataService) - Redwood City, United States Duration: 30 Mar 2015 → 2 Apr 2015 Conference number: 1

Conference

Conference	2015 IEEE First International Conference on Big Data Computing Service and Applications (BigDataService)
Country/Territory	United States
City	Redwood City
Period	30/03/15 → 2/04/15

Bibliographical note

© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Keywords

Big data
Bankruptcy
Data models
Predictive models
Companies
Robustness
Support vector machines
learning (artificial intelligence)
Bid Data
construction industry
data analysis
financial data processing
lifelong machine learning
bankruptcy prediction model
construction business
BPM
Big Data analytics approach
credit worthiness
financial data
Hadoop database
Apache Mahout
Construction business failure
Big data analytics
Bankruptcy prediction models
Machine learning
Financial models

Access to Document

10.1109/BigDataService.2015.30

Post-PrintAccepted author manuscript, 611 KB

Cite this

Bankruptcy prediction of construction businesses: Towards a big data analytics approach. / Alaka, Hafiz; Oyedele, Lukumon O.; Bilal, Muhammad et al.
2015 IEEE First International Conference on Big Data Computing Service and Applications (BigDataService). IEEE, 2015. p. 347-352.

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

Alaka, H, Oyedele, LO, Bilal, M, Akinade, OO, Owolabi, HA & Ajayi, SO 2015, Bankruptcy prediction of construction businesses: Towards a big data analytics approach. in 2015 IEEE First International Conference on Big Data Computing Service and Applications (BigDataService). IEEE, pp. 347-352, 2015 IEEE First International Conference on Big Data Computing Service and Applications (BigDataService), Redwood City, California, United States, 30/03/15. https://doi.org/10.1109/BigDataService.2015.30

@inproceedings{2b4aee5c9f8a44c2b99310761807bb55,

title = "Bankruptcy prediction of construction businesses: Towards a big data analytics approach",

abstract = "Bankruptcy prediction models (BPMs) are needed by financiers like banks in order to check the credit worthiness of companies. A very robust model needs a very large amount of data with periodic updates (i.e. appending new data). Such size of data cannot be processed directly by the tools used in building BPMs, however Big Data Analytics offers the opportunity to analyse such data. With data sources like DataStream, FAME, Company House, etc. that hold large financial data of existing and failed firms, it is possible to extract huge financial data into Hadoop database (e.g. HBase), whilst allowing periodic appending of data from the data sources, and carry out a Big Data analysis using a machine learning tool on Apache Mahout. Lifelong machine learning can also be employed in order to avoid repeated intensive training of the model using all the data in the Hadoop database. A framework is thus proposed for developing a Big Data Analytics based BPM.",

keywords = "Big data, Bankruptcy, Data models, Predictive models, Companies, Robustness, Support vector machines, learning (artificial intelligence), Bid Data, construction industry, data analysis, financial data processing, lifelong machine learning, bankruptcy prediction model, construction business, BPM, Big Data analytics approach, credit worthiness, financial data, Hadoop database, Apache Mahout, Construction business failure, Big data analytics, Bankruptcy prediction models, Machine learning, Financial models",

author = "Hafiz Alaka and Oyedele, {Lukumon O.} and Muhammad Bilal and Akinade, {Olugbenga O.} and Owolabi, {Hakeen A.} and Ajayi, {Saheed O.}",

note = "{\textcopyright} 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.; 2015 IEEE First International Conference on Big Data Computing Service and Applications (BigDataService) ; Conference date: 30-03-2015 Through 02-04-2015",

year = "2015",

month = aug,

day = "13",

doi = "10.1109/BigDataService.2015.30",

language = "English",

pages = "347--352",

booktitle = "2015 IEEE First International Conference on Big Data Computing Service and Applications (BigDataService)",

publisher = "IEEE",

address = "United States",

}

TY - GEN

T1 - Bankruptcy prediction of construction businesses

T2 - 2015 IEEE First International Conference on Big Data Computing Service and Applications (BigDataService)

AU - Alaka, Hafiz

AU - Oyedele, Lukumon O.

AU - Bilal, Muhammad

AU - Akinade, Olugbenga O.

AU - Owolabi, Hakeen A.

AU - Ajayi, Saheed O.

N1 - Conference code: 1

PY - 2015/8/13

Y1 - 2015/8/13

N2 - Bankruptcy prediction models (BPMs) are needed by financiers like banks in order to check the credit worthiness of companies. A very robust model needs a very large amount of data with periodic updates (i.e. appending new data). Such size of data cannot be processed directly by the tools used in building BPMs, however Big Data Analytics offers the opportunity to analyse such data. With data sources like DataStream, FAME, Company House, etc. that hold large financial data of existing and failed firms, it is possible to extract huge financial data into Hadoop database (e.g. HBase), whilst allowing periodic appending of data from the data sources, and carry out a Big Data analysis using a machine learning tool on Apache Mahout. Lifelong machine learning can also be employed in order to avoid repeated intensive training of the model using all the data in the Hadoop database. A framework is thus proposed for developing a Big Data Analytics based BPM.

AB - Bankruptcy prediction models (BPMs) are needed by financiers like banks in order to check the credit worthiness of companies. A very robust model needs a very large amount of data with periodic updates (i.e. appending new data). Such size of data cannot be processed directly by the tools used in building BPMs, however Big Data Analytics offers the opportunity to analyse such data. With data sources like DataStream, FAME, Company House, etc. that hold large financial data of existing and failed firms, it is possible to extract huge financial data into Hadoop database (e.g. HBase), whilst allowing periodic appending of data from the data sources, and carry out a Big Data analysis using a machine learning tool on Apache Mahout. Lifelong machine learning can also be employed in order to avoid repeated intensive training of the model using all the data in the Hadoop database. A framework is thus proposed for developing a Big Data Analytics based BPM.

KW - Big data

KW - Bankruptcy

KW - Data models

KW - Predictive models

KW - Companies

KW - Robustness

KW - Support vector machines

KW - learning (artificial intelligence)

KW - Bid Data

KW - construction industry

KW - data analysis

KW - financial data processing

KW - lifelong machine learning

KW - bankruptcy prediction model

KW - construction business

KW - BPM

KW - Big Data analytics approach

KW - credit worthiness

KW - financial data

KW - Hadoop database

KW - Apache Mahout

KW - Construction business failure

KW - Big data analytics

KW - Bankruptcy prediction models

KW - Machine learning

KW - Financial models

U2 - 10.1109/BigDataService.2015.30

DO - 10.1109/BigDataService.2015.30

M3 - Conference proceeding

SP - 347

EP - 352

BT - 2015 IEEE First International Conference on Big Data Computing Service and Applications (BigDataService)

PB - IEEE

Y2 - 30 March 2015 through 2 April 2015

ER -

Bankruptcy prediction of construction businesses: Towards a big data analytics approach

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this