Fraud Detection: Using Of Data Mining In

Accounting Information System

Abdallah Jabari

, and M.Suyunus

Faculty of Economics and Business, Universitas Airlangga, Surabaya, Indonesia

Abdullahacc2017@gmail.com, suyunusm@yahoo.com

Keywords: Accounting System Fraud Detection, Data Mining, Fraud Detection.

Abstract: The process of data mining is obtains the required data for analysis, study, take preventive measures,

administrative decisions and fraud detection. This study aims at detecting fraud in evading the payment of the

cheques and promissory notes payable by the customer in the Palestinian work environment through the

extraction of data from the accounting information system and applied them on “Sbitany Home” Company

that one of the largest Palestinian companies that use the system of sale by instalments through promissory

notes and cheques very large. A qualitative research methodology is adopted in undertaking the investigation

to understand the actual conduct of practices that aims to recommend improvements for fraud control,

detection and prevent it through data mining to determine solutions to the problem. The results indicate that

the company has a lot of cheques and promissory notes that are not paid very much, therefore the company

suffers from this scourge very much by many fraudulent customers, which affects its activities and profits.

The data was mining from the accounting information system "Priority" which is using in this company. The

research methodology used these data according to the Microsoft Excel and analysis and extract the results

related to the detection of this problem and prove the existence of this problem. As a result, the study provides

information on the causes of evading the customers the payment and solutions and suggestions that help to

reduce them very significantly and therefore the researcher sees the need to participate between all competent

authorities concerned with these matters to eliminate this scourge that threatens companies in particular and

the State and society in general.

1 INTRODUCTION

Fraud detection is the identification of fraudulent

behavior once it has occurred. Once detection has

occurred action can be taken to limit the fraudulent

activity. To accomplish this, fraud detection requires

constant monitoring as well as constant evolution.

Fraud detection methods are in a constant state of

change or flux due to the nature of fraud. This is

because as soon as detection of a type of fraudulent

behavior takes place, criminals create new plans and

schemes for fraud. Furthermore new criminals

entering the fraud business may carry out fraud using

existing methods causing some fraudulent activity to

be cyclical. Fraud detection therefore can be seen as

being part of an overall strategy which in many areas

has become a business critical issue. This is because

it being both common place and difficult to combat.

Fraud has been found difficult to combat due to

troublesome and complex nature of designing

measures to prevent it (Brennan, 2012).

Accounting fraud is intentional manipulation of

financial statements to create a facade of a company's

financial health. It involves an employee, account or

the organization itself and is misleading to investors

and shareholders. A company can falsify its financial

statements by overstating its revenue or assets, not

recording expenses and under-recording liabilities

(Gerety & Lehn, 1997).

Data mining is an interdisciplinary subfield of

computer science. It is the computational process of

discovering patterns in large data sets involving

methods at the intersection of artificial intelligence,

machine learning, statistics, and database systems.

The overall goal of the data mining process is to

extract information from a data set and transform it

into an understandable structure for further use. Aside

from the raw analysis step, it involves database and

data management aspects, data pre-processing, model

and inference considerations, interestingness metrics,

complexity considerations, post-processing of

discovered structures, visualization, and online

Jabari, A. and Suyunus, M.

Fraud Detection: Using Of Data Mining In Accounting Information System.

In Proceedings of the Journal of Contemporar y Accounting and Economics Symposium 2018 on Special Session for Indonesian Study (JCAE 2018) - Contemporary Accounting Studies in

Indonesia, pages 23-30

ISBN: 978-989-758-339-1

updating (Chakrabarti et al., 2006). Data mining tools

take data and construct a representation of reality in

the form of a Data mining activities fall into three

general categories (Desai & Deshmukh, 2013). Data

mining is the analysis step of the "knowledge

discovery in databases" process, or KDD (Fayyad,

Piatetsky-Shapiro, & Smyth, 1996).

One of the systems which collates and classifies

data collected by organizations is the Accounting

Information System (AIS). This system provides

financial information that can be used to plan,

evaluate and diagnose the impact of operating

activities and identify the financial position of the

organization. Given that these systems today collect

vast amounts of data, this data can be ‘intelligently’

analyzed by data mining technologies - sophisticated

and powerful cutting-edge technology that enables

the extraction of hidden predictive information from

a large database (Kurt, 2004).

An accounting information system (AIS) is a

system of collecting, storing and processing financial

and accounting data that are used by decision makers.

An accounting information system is generally a

computer-based method for tracking accounting

activity in conjunction with information technology

resources. Accounting information systems are

designed to support all accounting functions and

activities including auditing, financial accounting and

reporting, managerial/ management accounting and

tax (Palshikar, 2002).

Justification for this research Due to the use of the

companies working in the private sector in Palestine

installment sales system by cheques and promissory

notes in a very large percentage which make these

companies face problems from the customer’s fraud

and evasion in the payment of the value of cheques

and promissory notes. Installment sales is by using

promissory notes and cheques as a guarantee for the

collection of amounts due on a monthly or semi-

annual or annual according to the agreement between

the company and customers, with the knowledge that

most of the payments are monthly.

Based on the above justification for the research

what happens when using the sales system in this way

we produce a problem. The problem is the

customer’s fraud and evasion in the payment of the

value of cheques and promissory notes at a specific

time after the purchase between the company and

customers, making promissory note unpaid and

cheques from the bank and will therefore affect the

activity of the company and its profitability and

increase accounts receivable.

There is a lack of knowledge of the status of the

implementation of the extraction of technological

data within the public information system in

Palestine, this technique “ Data Mining” is applied

and the best model for implementation and extraction

for data to look for trends and patterns in accounting

data that reveal fraud in Palestine. And therefore the

objective of this study is detect fraud through Data

Mining of Accounting Information System and Clear

to prove the role of accounting information system in

detect and prevent fraud, this could lead to prevent

fraud in the future.

The next sections contains the literature review

and the theory to adopt the principle of data mining

through the accounting information system to detect

fraud Furthermore section 3 documents the research

method . The results and discussions are presented in

section 4 and finally section 5 concludes the paper.

2 LITERATURE REVIEW,

THEORIES AND PREVIOUS

STUDIES

2.1 Data Mining and Fraud Detection

Data mining is a process that uses a variety of data

analysis tools to discover of data in all its forms and

types that may be used to make valid predictions

(Miller & Han, 2009). Data mining is capable of

answering questions about the past (what has

happened) the present (what is happening) and the

future (what might happen). Data mining permit

analysis and identification of ‘hidden’ relation in

large datasets. By permitting this, the uncovered

information previously is now covered and would

give more support in the process of decision making

(Nemati & Barko, 2002).

The elements of an effective and sound anti-fraud

strategy, prevention, detection, deterrence, response.

They are closely related and each playing its essential

role in combating fraud. Fraud detection can act as a

deterrent by spreading a message to potential

fraudsters that the company is fighting fraud and that

necessary procedures are placed to pick any unlawful

activity which could have happened. A potential

fraudster would desist from committing the crime if

there is a possibility of being caught. Suspected and

detected fraud incidents need a consistent and

comprehensive response, this will spread a message

across that fraud is a critical issue and that subsequent

action would be taken against the perpetrator. Any

fraud case which is detected and investigated should

strengthen deterrence and therefore, act as fraud

prevention technique or measure (Dzomira, 2015).

JCAE Symposium 2018 – Journal of Contemporary Accounting and Economics Symposium 2018 on Special Session for Indonesian Study

One of the techniques for detecting fraud by data

mining is Data Visualization that the researcher can

be used through following its steps of the work. Data

visualization refers to the techniques used to

communicate data or information by encoding it as

visual objects (e.g., points, lines or bars) contained in

graphics. It involves the creation and study of the

visual representation of data, meaning "information

that has been abstracted in some schematic form,

including attributes or variables for the units of

information"(Friendly & Denis, 2001).

2.2 Theories

There are several different theories that may explain

the demand for audit services. Some of them are well

known in research and some of them are more based

on perceptions such as Information theory,

visualization theory, policeman theory, Lending

credibility theory, Agency theory, theory of inspired

confidence.

2.2.1 Information Theory

Information theory studies the transmission,

processing, utilization, and extraction of information.

A principle is the information principle, focusing on

the provision of information to enable users to

making decisions. Investors value the audit as a

means of improving the quality of financial

information. (Eid, 2014). Since the information

theory is, processing, use and transfer of information

extraction studies, it is, according to this research can

be used through the information system of any data is

extracted, used and processed through this system

that helps and gives management accurate and right

information for decision-making such as giving

information about customers before agreeing to sell

them by cheques or promissory notes of exchange and

thus helps to reduce evasion problem in the case that

as a result of the information was positive or negative.

2.2.2 Policeman Theory

The policeman theory posits that auditing is focused

on arithmetical accuracy and on prevention and

detection of fraud and the auditor is responsible for

search on fraud (Ittonen, 2010). The policeman theory

and Information theory are considered most relevant

for this study. Policeman theory is the most widely

held theory on auditing until the under this theory

(Salehi, 2011).

We can have advantage of the policeman's theory

in this research through AIS that is regarded as a cop

electronically rather checker person. Then the system

monitors and follows up sales operations through

cheques from customers during the audit information

and ensure the safety of all the sales process

measures. In addition to that, the detection of any case

of evasion and fraud, prevent and reduce them.

2.3 Previous Studies

Some previous studies have focused such as Farrell &

Franco, 1999 on the responsibility of the auditor in

detecting fraud and error and the extent of reduction

of misinformation in the financial reports and

addressed some special topics auditor legal

responsibility toward others and the difficulties faced

by the auditor and also discuss the development of

some models and techniques that can be used by

auditors in order to discover the fraud and the extent

of commitment Auditor international auditing

standards.

While some studies such as RamaKalyani &

UmaDevi, 2012; Stolfo, Fan, Lee, Prodromidis, &

Chan, 2000; Goode & Lacey, 2011; Soltaniziba &

Balafar, 2015; Sharma & Panigrahi, 2013;

Albashrawi, 2016 have focused on the types of fraud,

including mortgage fraud, financial reporting, bank

cheques in the banking sector, promissory notes,

credit card. And methods of fraud detection through

several ways including the method of extracting data

and methods of auditing and analysis.

Some studies such as Koornhof & Du Plessis,

2000 also touched on the originators of cheating like

executive managers, auditors, accountants, corporate

governance, whether in the private or public sector,

lenders and investors. All of these studies touched

each mentioned things and put a variety of solutions

to detect fraud and reduce it all analytical liquids or

applied through the use of technology and data

analysis.

3 RESEARCH METHODOLOGY

3.1 Type of Research

A qualitative research methodology is adopted in

undertaking the investigation to understand

fraudulent practices and aims to recommend

improvements for fraud control and detection it and

prevent it through data mining and analyze them and

making decision for solution these problems.

Thus, case study research involves the study of an

issue explored through one or more cases within a

bounded system. I choose to view it as a

methodology, a type of design in qualitative research,

Fraud Detection: Using Of Data Mining In Accounting Information System

or an object of study, as well as a product of the

inquiry, through detailed, in-depth data collection

involving multiple sources of information and reports

a case description and case-based themes. The case in

this study of grounded the company is suffering in

Palestine by its customers as i spoke earlier about the

problem of research and justification of research.

Approaching the research problem using an

interpretive worldview will be the most effective

approach at answering the research questions.

3.2 The Type of Data and Data

Collection Techniques

Through the telephone conversation, the researcher

gained data from a company operating in Palestine for

the last 5 years about sales through promissory note

and cheques, such as total sales through them, the

total commitment to pay the value of the promissory

note and cheques, the total evasion value of the

promissory note and cheques, the cause of fraud by

customers. And all the data needed by the researcher

for this study and an understanding concerning the

protocols, perceptions, beliefs, experience and action

activities from the respective heads of division at the

group level. The telephone conversation were

conducted with the highest level of division at the

group level.

3.3 Data Analysis Techniques

The researcher obtained all the data related to the

research, which includes the total sales of the

company in general each month separately for five

years, sales through cheques and promissory notes,

cheques cheques and promissory notes unpaid with

the knowledge of the reason of evasion by the

customer via e-mail. The researcher used the

Microsoft Excel to arrange the data that was random

and needed to be revised and arranged.

4 RESULTS

The Priority system is AIS that is used by the

company. Through this system the customer data is

entered, a large part of which is from the request form,

and a new account is created that this company used

it in order to get information around customers before

sell them . The sale is then recorded by installments

through the accounting information system used by

Sbitany Home in recording all its activities. The data

for the company's sales were extracted for the past

five years from 2012-2016. In addition, based on the

ability of the system to extract all the required data,

the sales data were extracted through cheques and

promissory notes as follows:

1. In the beginning, the received cheques are

entered from the customer in the system through

make of a check receipt and thus cheques are

entered into the cheques box in the system.

2. The next step will be to send to the bank for

collection and thus get them out of the system

from the cheques box to the bank's box in

accounting system.

3. The bank will in turn collect the cheques from the

customer. If the customer fails to pay the cheques

will be returned to the company and then the

accountant will then re-enter the system in the

system.

4. Choose financials that contain from the main list.

5. Go to the cash managements that contain

multiple options related to the cashiers.

6. Choose the cash management report in order to

get the report.

7. Choose Cheques box and then the system will

mining all the data related to the cheques

according to the period specified by the user, as

it contains the Cheques box to determine the time

period to obtain the data.

As for promissory notes, the output of data on

them is in the same manner as for cheques, but the

bond fund should be selected instead of the cheques

fund. There are several different things in dealing

with the promissory notes in the system:

1. When the customer is sold through promissory

notes they are entered into the promissory note

fund in the system.

2. Promissory notes are not sent anywhere to

collect, but they remain in the company and are

collected on the date specified directly by the

customer in the company.

3. In the case of a customer's obligation to pay the

value, it is removed from the system and

therefore is not considered an unpaid

promissory note. In case of non-payment, the

promissory note remains in the system and thus

the fraud process is obtained.

4. In the process of extracting data relating to

unpaid promissory notes, their fund is selected

in the system and a specific date such as the

date of the day is established. Therefore, any

promissory note appearing in the report prior to

this date is an unpaid promissory note and that

he is still in the company and thus the existence

of the process of fraud.

JCAE Symposium 2018 – Journal of Contemporary Accounting and Economics Symposium 2018 on Special Session for Indonesian Study

After mining all the required data from the system

and following the steps of identification and use of

the accounting information systems in extracting the

required data, then the data was obtained as complete

at the end of the research in the tables that contain the

data after processing, and to remove data that is not

important or required for research.

The results of the percentage of sales by cheques

and promissory notes of sales in general for each year

from 2012-2016. The percentage of promissory notes

sold was 13%, 8%, 9%, 9% and 11%, respectively.

The percentage of sales by cheques was as follows:

45%, 46%, 36%, 27% and 11%, respectively. These

percentages and figures show that the company uses

the system of sales by cheques and unpaid promissory

notes, and therefore, after knowing the existence and

use of this method by the company has been working

on the extraction of data on the escape by customers.

According to the ratio of sales of cheques and

promissory of exchange, the results of the table 1

which contains the ratio of unpaid promissory notes

of exchange and cheques from the total promissory

notes and cheques sold for five years each year from

2012-2016 were as follows: 16%, 9%, 14%, 6%,

respectively, and the percentage of cheques returned

16%, 7%, 9%, 20% and 42%, respectively. This table

shows that there is a large amount of cheques and

promissory notes, the existence of the problem that

lies in the evasion of customers in the payment of the

value of both promissory notes of exchange and

cheques on time and which have been talked about

the first unit of the search.

Table 1: Percentage of cheques and promissory notes

due from cheques and promissory notes sold.

Year

Total

P.N sold

Total

Cheques

sold

%Total

Unpaid

P.N of

sold

%Total

Cheques

Returned

of sold

2012

1.549.884

5.324.481

10%

16%

2013

721.945

4.211.325

16%

2014

983.266

4.127.559

2015

1.152.234

3.268.824

14%

20%

2016

1.383.473

1.412.683

42%

Table 2: Reasons for evading payment of cheques.

The

reason

2012

2013

2014

2015

2016

Insufficient

balance

75%

56%

92%

90%

85%

The

signature

does not

match

21%

44%

A mistake

in writing

the check

The

account is

closed

The owner

of the

check died

This data shows the existence of the problem and

the customers evade the obligation to pay the dues in

a way that affects the activities of the company and

its purpose of profit, which led us to continue the

research objectives, which lies in the discovery of this

problem through the extraction of data and make

recommendations and guidance necessary for that.

Based on the results and as attached to the table 2,

which explains the reasons for customers evading the

obligation to pay cheques. If we look at this table, we

find that the highest percentage is evasion because of

insufficient balance, which gives us the result that the

customer has the intention not to pay the obligation

and fraud and therefore not to put the value of the

check in the bank to spend chase the fraudsters and

record all the observations that help solve this

problem and participate in decision-making with

administrators and decision-makers.

These reasons are known after returning to the

company from the bank, which in turn writes the

reason for reference against the back of the cheques.

As for the promissory notes, the company does not

know the reason until follow the customer directly to

find out why. According to the company's response to

the question of the reasons for the return of the

promissory notes. This means that the reason can only

be determined when the customer is followed up and

Fraud Detection: Using Of Data Mining In Accounting Information System

ask him/her for the amount, and therefore the reason

is added to the report that the employee makes for the

mangers and not to the accounting information

system and its emergence when extracting data on the

promissory notes. And therefore not to get a specific

reason or clear except that he does not want to pay or

want to postpone longer. Thus, the employee or

accountant in charge of follow-up customers prepares

full reports and sends them to senior management for

decision making. Therefore, it is difficult to give a

specific percentage on the reasons for evasion in the

payment of promissory notes because there is a large

amount of them.

According to information theory and the

relationship of this theory research and its essence in

the use of data and analysis to discover the problem

and make decisions based on these data has already

been extracted and analyzed and extracted results to

discover the problem and take decisions as shown in

the tables above. This data is capable of answering

questions about the past (what has happened) by

extracting and analyzing old data in order to benefit

from them and take lessons from the previous results

in decision-making. According to the data extracted

from previous years help the company to learn from

the previous mistakes that were the cause of this

problem in the sale process and avoid falling again

after study and analysis and know the reasons.

The present (what is happening) by extracting

new data currently available and comparing it with

old data and analyzing it, which further reduces errors

and updates new decisions that increase the positive

decision-making in order to minimize the problem as

much as possible. In other words, additional measures

follow the procedures taken from the analysis of past

data.

And the future (what might happen) such as

predicting a future event as a result of the existence

of data that is constantly repeated by analyzing these

available data. The use of the data available in this

research helps to establish preventive measures and

laws that help to reduce this problem significantly. It

may not eliminate the problem permanently but limit

it and help the company to continue its activities.

These data are all in accordance with the theory of

the policeman, which was addressed in the second

unit of the search, this system of accounting

information used in the company will be the

accountant of the company to include the full

information on the sale process from the beginning

until the problem and write all these observations and

the necessary reports to make the necessary and

important decisions By the administration to resolve

this problem and to develop measures to reduce it. In

addition to this, first follow the clients in order to

collect the obligations owed to them. That is, he does

his work like a policeman who collects and writes all

the information and chase the fraudsters and record

all the observations that help solve this problem and

participate in decision-making with administrators

and decision-makers.

5 CONCLUSIONS AND

RECOMENDATIONS

5.1 Conclusion

Our goal is to detect the evasion of promissory notes

and cheques by customers, which is known as

customer fraud in the Palestinian environment

through the extraction of data through accounting

information systems. Found that this company is

facing this problem very significantly as a result of

the data mining by a company that is work in

Palestine, which is used by cheques and promissory

notes very large and the negative thing is that this

problem did not stop or decrease significantly but has

been going on since Years so far. I find that there was

a lot of indulgence by the company in the pre-sale

process until after the evasion process in the

procedures followed in the process of tracking

fraudsters, but the positive thing is that the company

began to develop and take measures to reduce the

problem and therefore found that this research will It

helps them develop in reducing them by helping to

develop these procedures.

5.2 Recommendations and Suggestions

The recommendations from this study for future

research include:

1. Based on the application form to open a new

account used by the company to collect data

about the customer before the sale, I find that this

model lacks some things that would reduce the

chances of falling into this problem, so I

recommend the company need to work another

model follows this model includes his bank

statement To fully know that the customer is

committed to pay cheques in general, which

gives the company very important information

enables them to agree to sell it or not. In addition,

one or more persons will be brought as the

guarantor of the customer, especially when

buying through promissory notes, and the

JCAE Symposium 2018 – Journal of Contemporary Accounting and Economics Symposium 2018 on Special Session for Indonesian Study

guarantor shall be liable in case of non-

compliance by the customer.

2. I recommend banks to close the account of any

person who does not commit to pay up to three

cheques and to conduct a complete and detailed

study of customers who want to open a bank

account, which will help in the inability of

scammers to have a cheques book and thus the

inability to buy through cheques. And thus helps

companies not to fall into the trap of these

fraudsters and prevent them from falling into this

problem.

3. The government must impose strict laws against

those who fail to commit themselves to paying

the amounts due, forcing any person to fear the

consequences before thinking about fraud from

the bank and forcing him to commit to pay the

value of cheques and promissory notes on time.

Such as long prison terms or pay fines.

4. I proposed the establishment of a huge

information system between the banks operating

in the country and the Association of

Accountants and a committee of the government,

where the idea to collect all data in this system

includes all persons with bank accounts and

therefore the company and before the approval of

the sale back to this system through the

Association of Accountants And the approval of

the government committee to obtain this

information, which makes the names of

fraudsters in this system on the black list. Not

only cheques, but all kinds of installment as the

companies have many names of referrals who

dealt with them through promissory notes and

then can add their names to the black list in the

system, which benefits all other companies that

follow the system of sale installments. This

significantly reduces the sale of fraudsters

because the company has advance information

about this person through this common system.

Not only that, but companies can add information

about all the customers who have bought from

them whether they were fraudulent or committed

to pay their dues.

REFERENCES

Albashrawi, M. (2016). Detecting financial fraud using data

mining techniques: A Decade Review from 2004 to

2015. Journal of Data Science, 14(3), 553-569.

Brennan, P. (2012). A comprehensive survey of methods for

overcoming the class imbalance problem in fraud

detection. Institute of technology Blanchardstown

Dublin, Ireland.

Chakrabarti et al. (2006). Data mining curriculum: A

proposal (Version 1.0). Intensive Working Group of

ACM SIGKDD Curriculum Committee, 140.

Desai, B. and Deshmukh, R. (2013). Data mining

techniques for fraud detection. International Journal of

Computer Science and Information Technologies

(IJCSIT), 4(1), 1-4.

Dzomira, S. (2015). Fraud prevention and detection

research. Journal of finance and Accounting, 6.

Eid, A. (2014). The impact of accounting information

systems (ais) on performance measures with value

relevance of auditors’communications. Internal

auditing and risk management, 2(34), 143-158.

Farrell, R. and Franco, R. (1999). The role of the auditor in

the prevention and detection of business fraud: SAS No.

82. Western Criminology Review, 2(1), 1-12.

Fayyad et al. (1996). From data mining to knowledge

discovery in databases. AI magazine, 17(3), 37.

Friendly, M. and Denis, J. (2001). Milestones in the history

of thematic cartography, statistical graphics, and data

visualization. URL http://www. datavis. ca/milestones,

32.

Gerety, M. and Lehn, K. (1997). The causes and

consequences of accounting fraud. Managerial and

Decision Economics, 587-599.

GOODE, S. and LACEY, D. (2011). Detecting complex

account fraud in the enterprise: The role of technical

and non-technical controls. Decision Support Systems,

50(4), 702-714.

Ittonen, K. (2010). A theoretical examination of the role of

auditing and the relevance of audit reports. University

of Vaasa, Finland.

Koornhof, C., & Du Plessis, D. (2000). Red flagging as an

indicator of financial statement fraud: the perspective

of investors and lenders. Research Journal of the

School of Accounting Sciences, 8(1), 69-93.

Kurt, T. (2004). An Introduction to Data Mining.

doi:http://www.thearling.com/text/dmwhite/dmwhite.h

Miller, J. and Han, J. (2009). Geographic data mining and

knowledge discovery. CRC Press.

Nemati, R. and Barko, D. (2002). Enhancing enterprise

decisions through organizational data mining. Journal

of computer information systems, 42(4), 21-28.

doi:10.1080/08874417.2002.11647049

Palshikar, K. (2002). A critical application for business

intelligence.

RamaKalyani, K. and UmaDevi, D. (2012). Fraud detection

of credit card payment system by genetic algorithm.

International Journal of Scientific & Engineering

Research, 3(7), 1-6.

Salehi, M. (2011). Audit expectation gap: Concept, nature

and trace. African Journal of Business Management,

5(21), 8376-8392.

Sharma, A. and Panigrahi, K. (2013). A review of financial

accounting fraud detection based on data mining

techniques.

Fraud Detection: Using Of Data Mining In Accounting Information System

Soltaniziba, S. and Balafar, A. (2015). The Study of Fraud

Detection in Financial and Credit Institutions with Real

Data. Global Journal of Computer Science and

Technology, 15(6).

Stolfo et al. (2000). Cost-based modeling for fraud and

intrusion detection: Results from the JAM project.

Paper presented at the DARPA information

survivability conference and exposition, 2000.

discex'00. proceedings.

JCAE Symposium 2018 – Journal of Contemporary Accounting and Economics Symposium 2018 on Special Session for Indonesian Study