TRUSTED EMAIL

A Proposed Approach to Prevent Credit Card Fraud in Soft-Products E-Commerce

Nien T. Sui, Saleh I. Alfuraih, Dennis McLeod

Department of Computer Science, Integrated Media Systems Center, University of Southern California, Los Angeles, CA

90089-2561, USA

Keywords: E-Commerce, credit card fraud, Internet security, trusted computing

Abstract: Soft-products are intangible products that can be consumed without shipment, such as software, music and

calling cards (calling time). The demand for soft-products on the Internet has been increasing for the past

few years. At the same time, fraudulent credit card transactions have also increased. Compared to tangible

products, fraudulent credit card transactions on soft-products are easier to conduct while difficult to recover.

The fraudulent transaction is a major problem for e-commerce merchants, customers, and credit card

issuers. In this paper, we classify the type of products sold on the Internet, and the common fraud that

occurs for each type. We review some of the best existing credit card fraud prevention methods and

introduce the Trusted email mechanism as a new way to prevent fraudulent transactions on soft-product.

Trusted email is a custom email solution that can uniquely identify and authenticate the online customer,

prevent unauthorized credit card transactions, and effectively resolve e-commerce disputes.

1 INTRODUCTION

The demands to prevent fraudulent credit card

transactions on soft-products e-commerce are

increasing as never before. By soft-products, we

mean products that can be delivered via the Internet

without physical shipment, such as downloadable

movies, music, software, and prepaid phone cards.

Credit card fraud is costing merchants, and

sometimes customers, hundreds of millions of

dollars each year. Fraud occurs 25% of the times due

to stolen cards; 24% due to counterfeit cards, where

the criminal acquires the technology that allows him

to “skim” the data contained in the magnetic strip of

an authentic card and then manufactures a fake card;

21% due to non-physical presence transactions in

mail, telephone and internet orders (MTIO), which is

the fastest increasing fraud; 15% due to lost cards;

and the remaining 15% is distributed among the rest

of credit card fraud (Vantage Card Services, n.d).

We classify the merchandise products into two

main categories: hard-products and soft-products.

Then we further classify them based on the

traceability of the fraud (Alfuraih, Sui, McLeod

2002):

Hard-Products: This includes all tangible

products that require delivery to a physical address if

purchased, such as laptops or clothes.

Soft-Products: This includes all intangible

products that can be shipped electronically.

Has-Cost-Traceable: The loss in this one is not

very high. Often times the cost of tracing can be

higher than the cost of the product itself, due to the

high cost of tracing (time of the agents at the

financial institutes, merchants, and FBI).

Has-Cost-Non-Traceable: The loss is high, and

traceability is almost impossible. For example a

Calling card pin number is sent to a free email and

accessed from an Internet café, or from an

international location. In such case there is

absolutely no way to trace it, yet the loss can be

high.

Has-No-Cost-Traceable: The loss in this kind

of fraud is extremely low and there is no need to

waste more money in tracing the thief. One example

of this is downloading a piece of music or software.

This piece of music or software costs money to

generate, but the cost of tracing the fraud is much

higher since one has to find the IP address of the

thief who downloaded it and trace it.

Has-No-Cost-Non-Traceable: The loss in this

one is almost the same as the previous one, but it is

practically impossible to trace the thief. An example

106

T. Sui N., I. Alfuraih S. and McLeod D. (2004).

TRUSTED EMAIL - A Proposed Approach to Prevent Credit Card Fraud in Soft-Products E-Commerce.

In Proceedings of the Sixth International Conference on Enterprise Information Systems, pages 106-113

DOI: 10.5220/0002612101060113

 SciTePress

of this is delivering an ordered picture to email, not

by download (no exact IP). In this case, with so

many free email services and the ease of getting one,

it is almost non-traceable and therefore impractical

to spend time and investigate with the email

provider.

This paper will concentrate on the credit card

frauds that can occur with the first type of soft-

products which is Has-Cost-Traceable. The

information needed to charge a credit card is only

the credit card number and the expiration date, but

merchants usually request more information to

prevent fraud. For example, the information required

for some online systems to approve an order is the

following: name, address, phone number, email,

credit card number, expiration date, the customer

service number of the credit card issuer, and special

codes on the physical card. The merchant then

verifies with the credit card issuer for the correctness

of the supplied information. So far, the industry

state-of-the-practice verification system is the online

address verification service (AVS), which can only

check whether the supplied address and the zip code

match or mismatch with the stored information in

the credit card issuer system (CPPS, n.d.) (VeriSign,

n.d). AVS is practical for hard-products since the

billing address can be the same as the shipment

address, or at least can be traced to a physical

address. However, AVS is almost useless for soft-

products. Any individual that obtains credit card

information and the billing address of that card can

easily place a fraudulent order for a soft-product.

Neither AVS nor the conventional tips for fraud

prevention, (PayPal, n.d.), (AntiFraud.com, n.d.), is

designed for soft-product e-commerce.

2 PREVIOUS SOLUTIONS

Credit card fraud has been a problem since the

beginning, and solutions for fraud detection have

been continuously proposed:

The most widely used billing address

verification system (AVS). (CCPS, n.d.), (VeriSign,

n.d.)

A distributed data mining approach for credit

card fraud detection (Chan, Fan, Prodromidis, et. al.

Stolfo 1999)

Various types of neural network solutions were

proposed for credit card fraud detection (Aleskerov,

Freisleben, 1997), (Brause, Langsdorf, Hepp, 1999)

(Ghosh, and Reilly, 1994), (Syeda, Zhang, Pan

2002)

A density-based clustering and radial basis

function modeling to generate credit card fraud

scores (Hanagandi, Dhar, Buescher 1996)

However, the credit card fraud in soft-products is

an emerging problem. The old solutions that were

used for hard-products may not be sufficient for the

problem at hand. Various new ways of detecting and

preventing credit card frauds have been proposed as

well. Most of these solutions are vendor specific

solutions and/or require major modifications on

existing e-commerce sites.

As an example of solutions from IT vendors,

Microsoft and other vendors, Orbiscom Inc. and

Cyota, have come up with a credit card surrogate

number that can be used for only one online

transaction (Bruno n.d). American Express has a

similar “proxy/filter” service that shields users’

credit card information from being exposed on the

net. Visa has introduced “Verified by Visa” solution,

and similarly, Master Cards and Discover have their

own fraud prevention methods (Heun 2002).

The notion of trust in e-commerce is not new.

Tradecraft, (Duvall 2001), has obtained a patent for

its Internet Payments solution. Their proposed

solution is a system of payment where a cable

company or ISP acts as a trusted authority for billing

subscribers when they purchase from online third

parties. This consortium of ISPs and/or vendors will

act as the intermediary to facilitate Internet payment,

somewhat similar to Visa International. Hsiung et al

pointed out the importance of trust in e-Business

growth (Hsiung, Sheurich, Ferrante 2001). Manchala

(2000) developed trust models and metrics for e-

commerce. Arif (2002) proposed establishing a trust

service provider (TSP) that is somehow similar to

the role of conventional banks that manage overseas

trading via letter of credits. The TSP will act as an

Internet-based intermediary and is responsible for

verifying the commodity of the online merchant and

the payment of the online customer.

Often, these solutions are not universal, meaning

they are specific to certain credit cards (Visa, or

American Express, but not both) or specific to

certain vendors (e.g. Tradecraft). These solutions

require customers to download some software (as in

Master Card and Discover solutions) or require the

bank and merchants to modify their system or

checkout procedure (as in Verified by Visa). All

these solutions cost money for merchants or banks to

be compliant with, and require mass acceptance

before they can be declared successful.

TRUSTED EMAIL: A PROPOSED APPROACH TO PREVENT CREDIT CARD FRAUD IN SOFT-PRODUCTS

E-COMMERCE

107

3 TRUSTED EMAIL SOLUTION

All MTIO merchants of soft-products that require an

email delivery are taking a risk by emailing the

customer the purchased products (if the transaction

is fraudulent). As explained before, this risk is even

worse than delivering to a non-billing address for a

hard-product, since email tracking is much more

difficult compared to physical address tracking.

We propose a trusted email (TE) solution, which

is a custom email server that can prevent online

frauds for soft-products and investigate disputes

efficiently, see Figure 1. The concept of this solution

is to treat the email address as the shipment/billing

address, just as the physical address is used in the

case of hard-products. The trusted email solution

will enable a more comprehensive address

verification system, called full address verification

system (FAVS), which will verify the email address

of the customer on top of traditional AVS. The email

address needs to be stored in the credit card issuer

system in addition to the other information. Email

accounts are unique which makes it easy to be

verified. In case the email does not match, the only

thing the merchant has to do is to email the customer

to send his “trusted” email address or cancel the

order.

The solution of simply registering the email in

the credit card issuer system will work perfectly if

an outsider commits the fraud. However, if the credit

card owner is the one who commits fraud, it is

difficult to convince the credit card issuer that the

dispute is invalid, because the credit card company

will not investigate with the free mail provider (e.g.

Hotmail or Yahoo) to verify if the merchant did send

an email to the customer containing the ordered soft-

product. Even if the email provider agrees to

investigate, there is no automatic investigation

technology, and there may be many privacy and

legal issues involved. The situation can be more

challenging if the customer who denied his own

transaction owns the email server.

Therefore, we propose to have a trusted email

server (TES) to be used with soft-products e-

commerce. The parities involved in the transaction

(the stakeholders: shopping-customers, merchants,

and credit card issuers) are requested to have an

account on TES and to use it for their transactions.

By that, when a fraud occurs, the credit card issuer

can query that trusted email server if the merchant M

emailed customer C to his email E on a certain date

and whether the email content included the ordered

soft-product or not. After verifying that the

transaction has occurred and the product has indeed

been delivered, the credit card issuer can then

reverse the charges to the merchant and reject

customer’s dispute, as in Figure 2. This process can

save the time of all the stakeholders and can protect

the merchants from fraudulent transactions or unjust

disputes.

Figure 1: Proposed New AVS

ICEIS 2004 - SOFTWARE AGENTS AND INTERNET COMPUTING

108

4 HOW TRUSTED EMAIL WORKS

Trusted email (TE) is the name of the technology

itself, and it consists of the following elements:

credit card issuers, merchants, customers, and the

trusted email server (TES). The details are explained

below.

Credit Card Issuer: Each issuer will be granted

access to the TES that enables them to use these

services:

Checking a customer credit history before

approving the credit application (since eventually,

TES will contain payment and dispute history of its

users).

Informing the customer of credit card approval

and managing card activations through TES.

On going dispute management for online

purchases. When there is a dispute regarding any

card the issuer will access TES to investigate the

dispute. Every issuer can only view the details of its

customer accounts.

Setting the dispute period duration after the

charge is billed in the statement, which is usually 3

months but some increase or decrease it.

Merchants: Each merchant will have to register

for a merchant TE account to deliver any soft-

product items to their customers. The TES has

mainly two types of accounts: normal accounts for

shopping customers, and special accounts for

merchants and credit card issuers. One of the

differences between the merchant account and

normal accounts is that the merchant account has the

capability to send to an unlimited number of

customers, while the normal account can only

receive emails and reply to the sender. TES allows

merchants to send emails to TE normal accounts

only. Through the TES, merchants will be receiving

dispute emails from customers or credit card issuers.

The TES dispute management system will integrate

the necessary information for the dispute for all

parties in order to resolve it efficiently. The TES

will handle any mail delivery failure and notify

senders of any problems. Merchants can have

multiple TE email accounts if they have multiple

sales departments. It is to the benefit of the

merchants to request their customers to register for a

TE account before purchasing using credit cards,

hence preventing any possible credit card fraudulent

transactions. The registration steps of TE accounts

can be easily integrated with the merchant website

and the process is similar to setting up any free

email account (but with less marketing options not

like most of the free emails). If a merchant allows a

non-TE customer to purchase the FAVS will return

N in the 4

field of the verification result.

Figure 2: Disputes cycle in the new AVS

TRUSTED EMAIL: A PROPOSED APPROACH TO PREVENT CREDIT CARD FRAUD IN SOFT-PRODUCTS

E-COMMERCE

109

Customers: The customers will have to register

with the TE and provide correct information, such as

name, address, and telephone. If any of information

provided is incorrect or outdated, the customer will

lose the right to dispute any fraudulent transactions

through the TES. After obtaining an account on

TES, the customer has to register this TE account

with all his credit cards issuers, in order to use the

FAVS while shopping online. Each customer can

have an unlimited number of accounts (since he

might share them with family members), but the

customer is informed that sharing access to the TE

account is similar to sharing the credit card.

Customers’ TE accounts can only receive or reply to

emails. They cannot send new emails, which

prevents them from becoming a burden on the TES.

Trusted Email Server (TES): The server will

be handling all the logistics to help optimize the

efficiency of this solution. The TES is a mail server

with added capabilities. The new capabilities are

listed below with the benefit of each one:

1- Tracking and record keeping of all purchase

transactions: The emails in the TES will remain in

the system until the period of the dispute ends. The

emails are then archived to offline storage. The

dispute period is usually from the time of the

purchase (in this case the time when the email is

received) to the maximum time set by the credit

cards issuers. The dispute period is typically from 30

days up to a maximum of 180 days from the date of

receiving the statement (as in the case of American

Express).

2- Allowing only TE user mail to be processed:

TES will allow only the emails that are originated

from a subscribed merchant or credit card issuer

system (to prevent spam-mails and to save

bandwidth and storage) and will block emails from

non-TE users. Merchants are not allowed to send

any marketing emails, such as new offers or deals;

for this purpose they need to ask their customer for

another email to which they can send such

marketing emails. If a merchant violates these

regulations, a fine will be charged. Repeated

violations will result in disabling the account, which

means that the merchant cannot send new emails,

but it can still view old emails and investigate

disputes. There are three types of emails that the

merchant can send: order confirmations, ordered

materials, and dispute follow-ups and chargebacks.

Figure 3: How Trusted Email Works

YES

Visit any

E-Site

Have a TE

Go to TES.com

Apply for TE

Call all your credit card

issuers and register your TE

with them

When you apply for a

new credit card use the TE

in the application form

Approved

If not the

first time

(To avoid infinite loop)

{Deny Order}

Else

Notify customer to add or update his

credit card with his TE

FAVS

Send Order

Make Order

ICEIS 2004 - SOFTWARE AGENTS AND INTERNET COMPUTING

110

For every email, the customers are facilitated with

the functionality to report any merchant violations to

the regulations.

3- Encrypting the connection with a special

email client: Trusted email clients will connect to

the TES using encrypted communication. The client

software differs based on the type of the users,

normal-shoppers, merchants, and credit card issuers.

4- Internationalization and localization: The

TES will allow business rule settings and dispute

policy depending on the country of the credit card

issuer. The language of the user interface is

customizable as well to meet the local needs.

Figure 3 shows the steps for approval or denial

when an order is placed. The customer will visit the

e-commerce site (e-site). If he does not have a

Trusted Email account, the customer needs to

email. If he has a Trusted Email (TE) he can proceed

and place the order, which will be subject to full

address verification (FAVS) including Trusted

Email verification, in addition to the traditional

AVS. If the verification is unsuccessful the customer

will be notified and his order will be verified one

more time only. Repeated denied transactions will

be black listed, to avoid infinite looping for FAVS

step.

5 DISCUSSION

We believe that our solution of using TE is a

universal solution that precisely targeting the core

problem of fraud purchases in soft-product e-

commerce. Email has been already widespread and

adopted by e-commerce stakeholders (customers,

merchants, and credit card issuers). Therefore, using

email as a starting point for our solution comes very

natural. However, the success of this proposed

solution depends on its technical soundness and its

ability to meet the needs of the stakeholders that are

affected by the new solution.

5.1 High-level Assessment

We will perform a high level assessment for the TE

solution in light of the following aspects:

technological, economic, social, and regulatory

aspects (Lee; Yu; Ku 2001). Although the analysis

performed by Lee et. al. 2001) was on payment

systems and TE is not a payment system,

nevertheless these aspects are applicable and are

vital to the success of such solutions.

TE solution covers the following key

requirements:

Authentication: TES provides user login

validation.

Privacy: The credit card issuer and merchants

can only view the transactions that are under dispute.

As for credit history, the credit card issuer can only

know the credit rating of the customer not the type

of merchants that were involved in the purchases.

Integrity: TES maintains integrity by ensuring

recording of the transactions and handling email

failures.

Non-repudiation: All transactions are recorded

and logged. If the merchant has delivered the

product to the customer, yet the customer denied

payment, all these details can be viewed in the

system.

Low Cost: The cost of such system can be

economic compared to monopoly solutions from

single vendors, such as Visa or American Express

alone.

Ease of Use: The concept of email is known to

all the stakeholders of e-commerce. The TES special

secure client-software can be easily integrated into

the existing email client software (a plug in).

There are other quality issues (level of service)

related to the design of the TE solution. We will

point out the main ones here:

Scalability: The system shall be able to scale as

the number of users grows. This requirement is

achievable as we see in the current email systems.

However, there are many trade-off issues in how to

scale and whether or not we should divide the users

according to their countries.

Security: Encrypted email is not a new

technology. However, using it in a large scale by the

public, and integrating with other non-secure

existing software requires a more careful study.

Availability: TES shall be used by users 24*7 as

in current verification and financial systems.

Therefore, single failure may affect a large number

of e-commerce stakeholders.

Compatibility: The interface to TES shall be

simple and shall adhere to current existing e-

TRUSTED EMAIL: A PROPOSED APPROACH TO PREVENT CREDIT CARD FRAUD IN SOFT-PRODUCTS

E-COMMERCE

111

commerce technologies. As compatibility is a major

factor that affects the acceptance and adoption of the

users. It is necessary for the merchant systems to use

the new FAVS codes, and to redirect shopping-

customers to register for TE accounts. The credit

card issuers systems need to store the new TE

information of their card holders, and to generate the

proper FAVS codes.

5.2 Making Winners of the

Stakeholders

For a technology to succeed, the acceptance and

adoption of the critical stakeholders are vital (Moore

1999). In TE solution, we want to make winners of

the technology key stakeholders (Boehm and Ross

1989), (Boehm, Bose, Horowitz, Lee 1994). For TE

solution, we analyze the needs and the win-

conditions of all key stakeholders, namely credit

card issuers, merchants, and shopping customers.

Each type of stakeholders has a different level of

needs for using the TE solution. Some can realize

more benefits than others. From analyzing one type

of soft-product e-commerce sites, namely calling

card sites, we reached to the following results:

Merchants: Calling card site merchants are the

most eager to adopt TE solution, due to the hassle

and inefficiency that they encounter using the

current manual verification system, and due to the

high loss if the transaction is fraudulent. With TE,

calling card merchants can avoid the delay and effort

of using manual verifications, and can eliminate the

risk of approving fraudulent transactions that will

later cause chargeback penalty.

Credit Issuers: It is to the advantage of the

credit card issuers to use the TE system. They can

eliminate the time spent to respond and investigate

customer dispute claims. The dispute process often

involves taking the claim from the customer,

contacting the merchant’s payment gateway provider

for chargeback penalty, waiting for the gateway

provider to get the merchant’s explanation, replying

back to the customer, and refunding the customer or

declining the dispute.

The main advantage for credit card issuers in

adopting TE solution is preventing credit card fraud

in one of the most difficult to trace fraud types in e-

commerce (soft-product, has-cost-non-traceable).

This has a far impact, as it will boost the confidence

of the online customers, and consequently promote

the use of credit cards in e-commerce transactions.

Shopping Customers: It is of great convenience

for the shopping customers to be able to purchase

simple products such as calling cards online

whenever they are needed without going to a local

gas station or grocery store. However, currently one

difficulty for the customers is that online calling

card sites require manual approval that can take 24

hours or more to verify customer information (which

the AVS cannot verify, such as phone number and

email). Thus, it is to the advantage of the customers

to use TE, which facilitates FAVS, in order to

receive immediate approval for their purchase.

More important, with TE, unauthorized

purchases of soft-products can be traced (if not

prevented) which in turn relieves the customers from

the problem of fraudulent transactions and dispute

hassles.

6 CONCLUSION

The best solution to prevent credit card fraud

transactions is the one that can be implemented with

minimum cost, requires minimum changes for all

stakeholders (customers, merchants, and credit card

issuers), and has strong incentive for all stakeholders

to adopt.

The proposed TE solution is very promising and

precisely serves the need of preventing credit card

fraud in soft-product e-commerce. However, several

technical challenges are there, such as scalability,

security, and integration with existing systems; in

addition to several non-technical challenges as well,

such as adoption and mass acceptance.

We believe that the proposed solution is far

superior to any current system that uses billing

address or other codes for verification. This solution

is easier to implement than the credit card secret pin

or surrogate credit card numbers. The solution has

none of the common drawbacks of other proposed

solutions, such as altering current e-commerce sites’

checkout procedures (as in verified by Visa) or

changing the systems of existing credit card numbers

(as in surrogate numbers). The proposed solution

also carries high incentives for current soft-product

e-commerce stakeholders to adopt.

7 FUTURE WORK

This is an on going research and the details of the

TES architecture and implementation are still in the

design stage. Other future plans include:

Investigating the use of Web Services

technology in the TE solution (particularly in

implementing the FAVS)

ICEIS 2004 - SOFTWARE AGENTS AND INTERNET COMPUTING

112

Developing the notion of “Trusted Receiver” in

the email domain, as so far the industry is mainly

focused on “Trusted Sender” (Postiva.com n.d.),

(TRUSTe and ePrivacy Group, n.d.)

REFERENCES

Aleskerov, E., Freisleben, B., Rao, B. (1997)

CARDWATCH: a neural network based database

mining system for credit card fraud detection.

Computational Intelligence for Financial Engineering

(CIFEr), 1997., Proceedings of the IEEE/IAFE 1997 ,

1997 Page(s): 220 –226

Alfuraih, Saleh, Sui, Nien T. Sui, McLeod, Dennis (2002).

Using trusted Email to Prevent Credit Card Frauds in

Multimedia Products, World Wide Web 5(2): 245-262

AntiFraud.com, Online Fraud Prevention Tips, (n.d.),

Retrieved March 10, 2002, from

http://www.antifraud.com/tips.htm

Atif, Y. (2002), Building trust in e-commerce, Internet

Computing, IEEE , Volume: 6 Issue: 1 , Jan/Feb 2002

Page(s): 18 –24

Boehm, B.W.; Ross, R. (1989), Theory-W software

project management principles and examples, IEEE

Transactions on Software Engineering, Volume: 15

Issue: 7, Jul 1989 Page(s): 902 –916

Boehm, B., Bose, P., Horowitz, E., Ming-June Lee (1994),

Software requirements as negotiated win conditions,

Proceedings of Requirements Engineering, 1994., 18-

22 Apr 1994 Page(s): 74 –83

Brause, R., Langsdorf, T., Hepp, M. (1999) Neural data

mining for credit card fraud detection. Proceedings of

11th IEEE International Conference on Tools with

Artificial Intelligence, 1999 Page(s): 103 –106

Bruno, Maria, (n.d) Microsoft Gives Boost to Surrogate

Card Numbers. Bank Technology News, Retrieved

March 10, 2002, from

http://www.banktechnews.com/btn/articles/btnoct01-

1.shtml

CCPS, Address Verification Service (n.d.), Retrieved

March 10, 2002, from http://mcvisa.com/avs.html

Chan, P.K., Fan, W., Prodromidis, A.L., Stolfo, S.J

(1999), Distributed data mining in credit card fraud

detection. IEEE Intelligent Systems Volume: 14 Issue:

6 , Nov.-Dec. 1999 Page(s): 67 –74

Duvall, Mel, (2001) Consortium to Facilitate Internet

Payments. Interactive Week September 24, 2001 Issue

page 26

Ghosh, S., Reilly, D.L.(1994) Credit card fraud detection

with a neural-network. System Sciences, Proceedings

of the Twenty-Seventh Hawaii International

Conference on Information Systems: Decision Support

and Knowledge-Based Systems, 1994 Vol.III Page(s):

621 –630

Hanagandi, V.; Dhar, A.; Buescher, K. (1996) Density-

based clustering and radial basis function modeling to

generate credit card fraud scores. Computational

Intelligence for Financial Engineering, 1996.,

Proceedings of the IEEE/IAFE 1996 Conference on ,

1996 Page(s): 247 -251

Heun, Christopher T. (2002), Fear of Fraud

InformationWeek.com (March 4, 2002), Retrieved

March 10, 2002, from

http://www.informationweek.com/story/IWK20020301

S0002

Hsiung, H. Sheurich, S. and Ferrante, F. (2001), Bridging

E-Business and Added Trust: Keys to E-Business

Growth, IT Professional, vol. 3, no. 2, Mar. 2001, pp.

41-45.

Lee, Zon-Yau, Yu, Hsiao-Cheng, Ku, Pei-Jen (2001), An

analysis and comparison of different types of electronic

payment systems, Management of Engineering and

Technology, 2001. PICMET '01, Volume: Supplement,

2001, Page(s): 38 -45 vol.2

Manchala, D.W. (2000), E-Commerce Trust Metrics and

Models, IEEE Internet Computing, vol. 4, no. 2, 2000,

pp. 36-44.

Moore, Geoffrey A. (1999), Crossing the Chasm:

Marketing and Selling High-Tech Products to

Mainstream Customers, HarperCollins Publishers,

1999

PayPal, Fraud Prevention Tips For Sellers, (n.d.),

Retrieved March 10, 2002, from

http://www.paypal.com/cgi-

bin/webscr?cmd=p/gen/fraud-tips-sellers-outside

Postiva.com Trusted Sender Program, (n.d.), Retrieved

June 10, 2003, from

http://www.postiva.com/article/articlestatic/17/1/5/

Syeda, M., Yan-Qing Zhang, Yi Pan, (2002) Parallel

granular neural networks for fast credit card fraud

detection, Fuzzy Systems, 2002. Proceedings of FUZZ-

IEEE'02, Volume: 1, 2002 Page(s): 572 –577

TRUSTe and ePrivacy Group, Trusted Sender program,

(n.d.), Retrieved June 10, 2003, from

http://www.truste.org/programs/pub_trustedsender.htm

Vantage Card Services, Inc, Prevent Chargebacks, (n.d)

Retrieved Oct 1, 2003, from

http://www.vantagecard.com/html/preventchargebacks.

html

VeriSign, Address Verification Service (AVS) Security

Feature, (n.d.), Retrieved March 10, 2002, from

http://www.verisign.com/support/payflow/genResource

s/avs.html

TRUSTED EMAIL: A PROPOSED APPROACH TO PREVENT CREDIT CARD FRAUD IN SOFT-PRODUCTS

E-COMMERCE

113