LOVE by fanzhongqing


									            Web-service Based Information Integration for e-Mortgage Contract
               Matchmaking Decision Support: A Case Study in Hong Kong

        Gordon K.T. Lau1, Dickson K.W. Chiu2, Senior Member, IEEE, and Patrick C. K. Hung3
                Department of Computer Science, Hong Kong University of Science and Technology
                                         Dickson Computer Systems, Hong Kong
     Faculty of Business and Information Technology, University of Ontario Institute of Technology, Canada


                    Abstract                                   from heterogeneous sources. When significant contracts
                                                               are involved, gathering such information is particularly
Effective decision making often requires disparate infor-      slow and inefficient. This particularly puts the consumers
mation from heterogeneous sources. When significant con-       in an adverse position because large financial enterprises
tracts are involved, gathering such information is particu-    are much more equipped and informed with various deci-
larly slow and inefficient. As real estate business plays an   sion support and information systems..
important role in Hong Kong economy, there is a strong             Mortgage contracts are particularly sophisticated in or-
need for a handy and efficient service to assist the mort-     der to stratify the unique requirements of both the custom-
gage searching activities. Presently, a property buyer         ers and the financial institutions. To elaborate, each cus-
lacks of the mortgage information for the interested prop-     tomer has individual needs for the mortgage and payment
erty when making the purchase decision. Many buyers can        contract even for the same property because of the indi-
only search for the property mortgage after a down pay-        vidual’s financial health and personal preference. On the
ment is made because of keen competitions in purchasing.       other hand, each financial institution would offer different
However, sometimes the bank cannot offer the mortgage          mortgage contracts to different customers even for the
plan as the buyer expected because of the buyer’s finan-       same property based on its risk management, different
cial status, property valuation, and price trend. Therefore,   customer assessment methodologies, and different market-
we study on the requirements of an e-Mortgage Contract         ing strategies.
Matchmaking Service (EMS) for offering it in the form of a         Currently, the mortgage searching process is tedious
Web service via real estate agencies to the buyers. The        and time-consuming for customers. They are responsible
EMS can be incorporated into the agency’s property man-        for searching appropriate mortgage contracts that suit their
agement system so that the relevant mortgage information       needs. Sometimes, property agencies may provide some
(provided by the participant financial institutions) could     suggestions and assistances in the mortgage application
be revealed when the buyer is making the purchase deci-        process; but such help is often trivial in terms of the whole
sion. Moreover, the EMS offers mortgage contract evalua-       complicated process. To find a suitable mortgage, custom-
tion and ranking to assists the prospective buyer earlier in   ers need to collect the information of various mortgage
the decision-making stage.                                     contracts from different financial institutions, and compare
                                                               their complex terms either manually or with partially
1. Introduction                                                computerized help. In an ideal case, a customer may man-
                                                               age to get a suitable mortgage after putting significant
                                                               time and effort. However, customers often lack of the
    Nowadays, technology has been playing an important
                                                               proper financial knowledge or market information for
role in many businesses; the need of catching the pace of
                                                               making decision on their mortgage contract selection. Due
technology evolving is becoming very demanding. Effec-
                                                               to insufficient time or information, many customers tend
tive decision making often requires disparate information
to settle down on less beneficial or less appropriate mort-     2. Background and Methodology Overview
gage contracts. Thus, the current mortgage industry de-
mands a new platform making used of new technologies to             Hong Kong is one of the most crowded cities in the
facilitate the present mortgage matchmaking process and         world. The property market has been a vital economic
to assist the customer in decision making by ranking the        sector in the economy. From the early 90’s, the Hong
mortgage information collectively, timely, in a well com-       Kong property market raised rocket-high and continued
parable format according to the customer’s requirements.        this trends till the late 90’s. Although the property market
    As a strategy in the design, the application model          experienced a set back from early 2000, the market has
should consist of the following characteristics such as plat-   bounced back and developed into a more prosperous and
form independent, adjustable searching requirements, in-        mature business recently. This has bought about similar
ter-machine connectivity, and high extensibility. To ad-        positive effects to the mortgage business at the same time.
dress these issues, this paper proposes an e-Mortgage Con-      In aggregate terms, residential mortgage in Hong Kong in
tract Matchmaking Service (EMS) to automate the infor-          the 90’s accounted for over 20 per cent of all financial
mation integration and matchmaking of mortgage con-             institutions. By contrast, this share was only 10 per cent in
tracts for customers to make effective decisions and to         the early 80’s. On the other hand, outstanding residential
minimize the hassles in this process.                           mortgage as a percentage of GDP rose from 11.4 per cent
    Web services [17][18] provide loosely-coupled stand-        in 1984 to 29.2 percent in 1994 [3]. As the mortgage busi-
ard interfaces among autonomous systems within and              ness became an important component in almost all the
among organizations in the form of a set of well-defined        financial institutions, they are willing to offer more so-
functions for both programming and human user interfac-         phisticated terms to meet the customers’ needs. To attract
es. In particular, Web services support event-driven in-        potential customers, financial institutions tend to design
formation integration for timely service provision and in-      mortgage contracts with various options such as cash re-
teraction [4]. So, the EMS can help the users to search         bate, flexible monthly payment plan, and discounted
suitable mortgage contracts offered by various financial        mortgage rate for a period of time. Moreover, different
institutions. The search can be tailor-made according to        financial institutions apply different property assessment
the customer’s needs, requirements, and preferences. On         schemes, which result in different property valuation on
the other hand, the financial institutions can propose dif-     the same property. With this growing complexity, the
ferent terms for customers as a response. In this sense, the    knowledge and effort required for customers to choose the
system can reduce the amount of time and effort of the          right mortgage contract has been enormously increased. In
customer in searching for appropriate mortgage contracts        Hong Kong, the duration of these mortgages usually lasted
as well as help the financial institutions meet potential       for 15 to 30 years. Therefore, choosing the right mortgage
customers’ need efficiently. Also, the EMS provides com-        becomes a very important but difficult problem.
parison and ranking function that uses the configurable
weights on different mortgage terms and calculates the
score of each mortgage offer. This also helps the user filter
out those unacceptable offers for more effective decisions.
    The contributions of this paper are as follows. (i) This
paper presents the requirements of the EMS with respect
to various stakeholders based on our consultation experi-
ences in Hong Kong; (ii) We introduce a meta-model of
EMS for the effective design and implementation of the
system; (iii) The EMS is equipped with a matchmaking
mechanism that facilitates the decision-making process;
and (iv) The system implementation architecture makes                    Figure 1. Typical flow of property acquisition
use of the contemporary Web service, which provides
                                                                    Based on our consultation experiences, Figure 1 shows
programmatic interactions as well as human user interac-
                                                                a typical property acquisition flow in Hong Kong. A buyer
tion. The rest of the paper is organized as follows. Section
                                                                typically searches for a property with the aids from vari-
2 discusses the mortgage business backgrounds and the
                                                                ous property agencies. After visiting and reviewing a
methodology overview. Section 3 discusses the related
                                                                number of properties repeatedly, the customer finally de-
work. Section 4 outlines the e-Mortgage model to address
                                                                termines his/her choice. Then, the buyer signs a temporary
the existing business requirements. Section 5 presents the
                                                                purchase contract for the selected property and pays the
implementation of the model in a Web service context and
                                                                down-payment, often without any detail mortgage infor-
its mechanisms to the searching and matchmaking process.
                                                                mation from the financial institutions yet. This is because
Section 6 concludes the paper with discussions and future
                                                                buying competition is often keen while the request for
work directions.
                                                                mortgage information is slow and time consuming. The
buyer is responsible for searching the mortgage within a        son. These channels have to be accurate, timely, conven-
certain period of time in order to fulfill the payment re-      ient, and secure. Moreover, they prefer to have decision
quirement of the purchase contract; otherwise he/she will       support for selecting among the complex mortgage con-
forfeit the down payment. Also, sometimes the financial         tracts.
institution may have a lower valuation on a property: this          Property Agencies: They are the representatives of the
means, the mortgage amount may be insufficient for the          property sellers. They provide advice and service to cus-
customer to pay the balance amount of the property. If that     tomers when selecting properties. As their incomes are
exceeds the customer’s financial scope, the customer will       mainly the commission generated from property trading,
also forfeit the down payment. On the other hand, if the        they are willing to offer add-on service to attract and assist
customer manages to obtain the mortgage after going             customers in property acquisition. Also, they want their
through a complicated and tedious process, a formal con-        services to be accurate and timely because good services
tract can be signed. Then, the customer will start paying       most probably help them to attract more customers.
the mortgage and the whole property acquisition process is          Credit Repository Providers: They are authorities or
completed.                                                      third party institutions responsible for providing the credit
    Figure 2 summarizes the stakeholders of the e-              history and credit rating to the financial institutions of a
Mortgage system (EMS). The system should cater for the          given mortgage customer. Their information should not be
need and interest of each stakeholder as follows.               associated with the interest of any financial institutions
                                                                and should be unbiased and transported through a secure
                                                                    In summary, the overall characteristics of all the stake-
                                                                holders in mortgage business are as follows. All stake-
                                                                holders want their systems to be interoperable but loosely
                                                                coupled for autonomous operations. They want their sys-
                                                                tems to be inter-connected to provide add-on services in
                                                                order to gain a competitive edge. At the same time, they
                                                                want to preserve the existing legacy systems, which are
                                                                heavily invested for the past decades and still plays im-
                                                                portant roles in their existing business processes. They
                                                                also required a new system, which can offer more auto-
                                                                mated processes and reduce much manual procedures. In
                                                                addition, they want to reduce the duration and efforts spent
                                                                on the mortgage searching and decision-making processes.

                Figure 2. Stakeholders of EMS

    Financial Institutions: They are the mortgage provid-
ers. They want to have secure channels to reach their po-
tential customers directly. Based on the highly competitive
mortgage business, they are willing to provide add-on ser-
vices to obtain any competitive edge. However, having
heavily invested on the existing legacy systems previous-
ly, they tend to rely on those legacy systems upon which
they build any new application systems. Moreover, as the
profit margin of each customer varies, they want to focus
on those can generate a high profit margin or have low
risks and avoid those less attractive ones who will con-
sume the financial institutions’ resources unnecessarily.
    Customers: They are property buyers and mortgage
service applicants. They are looking for mortgage plans
that suit their individual financial needs. They need ade-
quate and accurate mortgage information from financial
                                                                                Figure 3. Message Flow of EMS
institutions when making the property purchasing deci-
sion. This also enables them to prepare a better and con-           Web service technologies provide a language-neutral,
cise financial plan. Also, they prefer to have additional       environment-neutral programming model that accelerates
channels for searching mortgage services besides the tradi-     intra- and inter- enterprise application integration. Appli-
tional channel of visiting the financial institutions in per-   cation integration through Web services yields flexible
loosely coupled business systems. Because Web services                gage industry, namely, automated underwriting and
are easily applied as a wrapping technology around exist-             customer relationship management [8].
ing applications and information technology assets, new          Electronic signatures, or “e-signatures,” generally
solutions can be deployed quickly and recomposed to ad-               refer to any type of signature in electronic form at-
dress new opportunities [4]. There are formidable chal-               tached to or associated with an electronic document or
lenges to ensure rapid, reliable, relevant information ex-            record. It is an important tool to reduce the amount of
change in an IT environment encompassing both main-                   paper and streamline the process. E-signatures can al-
frames and next-generation, Internet-based system [1].                so help transform the verification process [8].
Many enterprises want to use Web services because it can
                                                                 Security technologies such as document integrity,
preserve the security, reliability, and speed of their under-
                                                                      authentication, and authorization are widely adopted
lying mainframe assets [2]. Web services, which use open
                                                                      in the mortgage process. There are several technolog-
standards to enable Web-based applications to interact
with one-another, is a promising integration solution in              ical implementations to ensure documents have not
this scenario [1].                                                    been tampered with: file creation and modification
    Figure 3 shows the message exchange flow of the                   dates, read-only files, password protection, checksum
EMS built on the Web service architecture (WSA). It aims              methods, watermarking technologies, and public key
to connect all participant institutions and presents the              cryptography. They can reduce the security risks in
mortgage information to the customers in a neat and well-             the application process, which are handled by people
formed way. As messages adhering to open standards are                [8].
employed to communicate between different institutions           Document imaging refers to the set of technologies
and property agencies, their existing systems such as                 for converting paper documents into electronic for-
mainframes can be retained and reused to provide funda-               mat. The process involves scanning or electronically
mental support to new valuable services. Also the EMS                 converting files into high-resolution images, which
can provide instant and accurate mortgage information via             are saved on a storage media. It helps to reduce the
the property agency to the customers, provided that it is             paper cost and errors caused by the human processing.
integrated with their property management systems. Fi-               In the organization connectivity context, some re-
nancial institutions are able to tailor-made its mortgage       searchers such as Hudson suggest the use of Enterprise
contract based on the received customer information and         Application Integration (EAI) architecture [6]. Hudson
financial strengths. They may even refuse to provide quote      continues to suggest that EAI architecture is effective
for those customers who are classified as unprofitable or       where one or more internal (stovepipe) systems or compo-
entitled with poor credit. This can reduce the effort and       nents must send data to or enact services on other hetero-
time needed to establish a mortgage offer for a property.       geneous internal systems or components. However, it is
With the term comparing function, the EMS provides              always more difficult to implement as the underlying ap-
mortgage ranking based on the configurable weights on           plications can be built on different technologies such as
various criteria to help customers make effective deci-         Visual Basic, C++, Java, etc. Also, systems must be totally
sions.                                                          compatible in the message structures they understand. It
                                                                usually requires spending weeks to months to develop
3. Related Work                                                 such systems.
                                                                     In response to the technological needs of the mortgage
    Bogdon [7] reports that technology has so far shown
                                                                business and the cumbersome nature of real estates trans-
the greatest impacts in mortgages servicing, leading to
                                                                action, Mittal [5] suggests that Web services technology
lower costs and higher labor productivity. Also, the use of
                                                                allows the buyer, seller, their agents, their mortgage and
technology has speeded up the process and improved the
                                                                title companies, and various other parties such as home
efficiency of the mortgage process [8]. In the past three
                                                                inspectors, property appraisers, notaries, home insurance
decades, technology has reduced the costs of storage,
                                                                agents to interact in order to achieve a better, faster,
transmission, and processing of information by 25 to 35
                                                                cheaper, and more reliable way of conducting electronic
percent annually. This trend is expected to continue [9].
                                                                business. Mittal proposes a quick e-marketplace of ser-
    Given the huge benefits of adopting technology for
                                                                vices could be formed, where individuals can potentially
tackling the business processes in mortgage industry,
                                                                describe their requirements and allow the entire e-
many financial institutions are currently using the follow-
                                                                marketplace of businesses to bid on their contract.
ing technologies:
                                                                     As such, we propose an EMS as a more achievable ap-
 Intelligent systems are computer systems that can
                                                                proach to satisfying the current needs of the mortgage
     learn and perform functions such as decision making,       business. The EMS is built upon Service-Oriented Archi-
     image recognition, and planning. Intelligent systems       tecture (SOA), and it offers the following advantages over
     are currently being used for two purposes in the mort-     the current technologies employed in the mortgage indus-
try. The EMS connects many different systems which can          mortgage contracts to those customers with bad credit
run independently of one another on entirely different im-      status.
plementation platforms and run-time environments with a              The Web Service Provider class represents all the ser-
set of standard protocols and data formats such as Ex-          vice providers such as financial institutions and credit re-
tended Markup Language (XML), Web Service Descrip-              pository providers. This class will use the information
tion Language (WSDL), Simple Object Access Protocol             retrieved from the mortgage quote request and try to fill
(SOAP), Universal Description, Discovery and Integration        the requirement to meet the Web Service providers’ input
(UDDI). The EMS is encapsulated as a secure entity and          parameters. They class will query the database to find the
the only visible part is a Web service as the public inter-     relevant Web service provider and then send the formatted
face, which can meet the security concern of the industry.      request to them.
The EMS is a composition of distributed systems of sepa-             Once all the reply is received from the financial insti-
rate companies to provide strategic business workflow; in       tutions’ Web services, the EMS will start to process the
return it helps to form a paperless mortgage transaction.       returned offers. Sometimes, the system would not receive
                                                                all the replies from the providers due to network problems,
4. E-Mortgage Conceptual Model                                  service out of order, or other unknown reasons. The EMS
                                                                is equipped with a time-out mechanism so that it only wait
                                                                for a period of time to collect all the expected replies be-
                  Mortgage Quote Request                        fore continuing the process.
                                                                     The Mortgage Quote class represents the entire mort-
                                                                gage offers from the financial institutions. It contains the
                                                                mortgage information such as mortgage terms, special
                                                                offers such as cash rebate or flexible payment plan, prop-
                                                                erty valuation, and so on. When all the mortgage replies
                                                                are return or after a deadline, the EMS will use the Rank-
                                                                ing Rule class to rank the mortgage plan list.
                                                                     The Ranking Rule class consists of information for
                                                                comparing and ordering the mortgage terms. The ranking
            Figure 4. UML Class Diagram for EMS                 criteria can be configured with different weights; in other
                                                                words, some criteria will be treated as more important than
    Figure 4 shows the conceptual model of the EMS in a         others. Finally, the EMS mortgage will present the offers
Class Diagram of the Unified Modeling Language (UML)            in a ranked order and help customers choose a best mort-
[10], which depicts the relationship between important          gage contract based on their pre-defined criteria.
    The Mortgage Quote Request class represents the ob-         5. System Design and Implementation
jects, which will be created when a mortgage quote for a
particular property is presented. This class contains attrib-       Figure 5 summarizes the overall system architecture
utes such as property information, detail quote information,    for the EMS which contains the Web Service Agent, Ap-
and other customer information. It also contains the Pref-      plication Layer, Ranking Module, and Database Connec-
erence, Financial Status, and Credit classes. When a mort-      tion Module. We shall detail the important functions of
gage quote request object is created, the system has to         these components in the following subsections.
collect all the relevant information before sending the re-
quest to the relevant financial institutions.
    The Preference class consists of mainly customer pref-
erence on the mortgage request. Customers’ preferences
may include the length of the mortgage, monthly payment
amount, the mortgage provider, and so on. The Financial
Status class consists of all the relevant financial informa-
tion of a customer. It includes information about the cus-
tomer’s monthly income, personal liability, personal as-
sets, and so on. The Credit class represents a customer’s
credit history. The relevant credit history will be retrieved
from the Web service of a third party (such as a Credit
Repository). Financial institutions are often willing to of-
fer a more attractive mortgage plan to lure customers with
good credit status while they offer with a harsh or even no
                                                              sending to financial institutions since they have different
                                                              message requirements; and vice versa, the EMS needs to
                                                              transforms the returned mortgage quotes into EMS data
                                                              structures using the XSL.

            Figure 5. System architecture of EMS                              Figure 7. Sample WSDL of EMS

5.1 Web Service Agent Module                                      Web Service Description Language (WSDL) [14] is an
                                                              XML language that contains information about the inter-
                                                              face, semantics, and administration of a call to a Web ser-
                                                              vice. It specifies the location of the service and the opera-
                                                              tions (or methods) the service exposes. The EMS provides
                                                              a WSDL to all the participant partners to communicate
                                                              with the system. Figure 7 shows a partial WSDL of the
                                                                  SOAP (Simple Object Access Protocol) [13], is an ex-
                                                              tensible, text-based framework for enabling communica-
                                                              tion between diverse parties. The EMS acts as a SOAP
                                                              client and SOAP server at the same time. As a SOAP cli-
                                                              ent, EMS creates an XML document containing the infor-
                                                              mation needed to invoke the method from the remote fi-
                                                              nancial institutions. While the EMS acts as a SOAP serv-
                                                              er, it listens for the mortgage quote request (SOAP mes-
                                                              sage) and interpreters the message for later processing.
              Figure 6. Web Service Operations                SOAP message are built on HTTP and XML standards. As
                                                              a result, SOAP documents are able to traverse almost any
    Web Services Architecture (WSA) requires three fun-       firewall, enabling the exchange of information across di-
damental operations (shown in Figure 6): publish, find,       vergent platforms. Figure 8 shows some sample SOAP
and bind. Service providers publish services to a service     messages of the EMS.
broker. Service requesters find required services using a
service broker and bind to them [11]. The Web Service
Agent Module is designed to communicate with other
third parties’ Web service interfaces such as financial in-
stitutions and credit repository provider. This module
makes use of the following technologies like XML,
WSDL, and SOAP to be executed within the EMS. XML
[12], as the core syntax within EMS, helps to format the
data to be transported within different parties by adopting
an agreed XML schema and helps exchange data from
various message formats by using eXtended Markup Lan-
guage Stylesheet Language (XSL). For example, the Web
Service Agent Module needs to transform the same mort-
gage quote request message into different formats before
              Figure 8. Sample SOAP messages

5.2 Application Layer
                                                                        Figure 10. Mortgage quote reply process flow
    The Application Layer consists of business logic of the                        in UML activity diagram
EMS. This layer controls and monitors the flow and com-
munications between different modules. Figure 9 demon-          5.3 Web Service Security
strates the typical flow of mortgage quote request process.
The EMS performs validation on some of the incoming                 WS-Security, as the main Web Service security speci-
message before passing it to the Web Service Agent mod-         fication, describes enhancements to SOAP messaging to
ule, which forms the SOAP messages. Also, it communi-           provide quality of protection through message integrity,
cates with the Database Connection Module to obtain the         message confidentiality, and single message authentica-
existing mortgage service providers’ information. On the        tion. These mechanisms can be used to accommodate a
other hand, Figure 10 shows a typical flow of the mort-         wide variety of security models and encryption technolo-
gage quote reply process. The Application Layer also con-       gies [16]. It provides an important foundation layer for
trols the Ranking & Decision module to process the re-          developers to build more secure and broadly interoperable
turned mortgage quote.                                          Web Services. It is a solid, open-standards-based security
                                                                model [15].

                                                                            Figure 11. End-to end security context

                                                                    The EMS uses Secure Sockets Layer (SSL), a point-to-
                                                                point technology to enforce the network level security. For
                                                                end-to-end security, the EMS adopts the WS-Security
                                                                standard to ensure the security of its service (See Figure
                                                                12). Firstly, the system uses an athentication mechanism to
                                                                ensure the identities of the service requester and provider.
                                                                This can be done through passwords or security
                                                                certificates such as Public Key Infrastructure (PKI).
                                                                Secondly, in order to control access to resources, some
                                                                authorization mechanisms are applied by self-defined
                                                                access rights, which are controlled by the system’s
                                                                application layer. Finally, the EMS makes use of data
Figure 9. Mortgage quote request flow in UML activity diagram   encryption and digital signature techniques to enforce data
integrity and data confidentiality. Figure 12 shows a
sample XML with security control applied.

                                                                             Figure 13. Sample weight XML

                                                                  Firstly, the EMS will find out the divergence between
                                                              the customer’s expected and the mortgage plan in terms of
                                                              the mortgage components. For example, customer Bill
                                                              receives 2 mortgage offers and these offers just consist of
                                                              3 components for the sake of simplicity.

                                                              Mortgage              Offer A     Offer B      Expected
             Figure 12. XML with security control
                                                              Interest (%)          5           6            5.5
5.4 Ranking & Decision Making Module                          Monthly Payment       7000        6000         6500
    The ranking and decision-making module is responsi-
ble for manipulating the mortgage offers returned from        Duration (Months)     120         300          180
different financial institutions to generate a ranking in a
logical order. A mortgage plan includes many components          Then, the EMS subtracts the offered value from the
such as the mortgage amount, mortgage duration, repay-        expected value for each component. The following table
ment payment, repayment schedule, interest rate, cash         shows the result after subtraction.
rebate (if any), early mortgage full payment, and so on. A
weighting system is designed to appropriately assign im-      Mortgage              Offer A               Offer B
portance to each of the components based on the custom-       Components
er’s ranking towards them. Figure 13 shows a sample
XML to illustrate the weighting of some key components.       Interest (%)          0.5 (5.5 – 5)         -0.5 (5.5 -6)
Once the ranking of each mortgage component is defined,       Monthly Payment       -500 (6500 –          500 (6500-
EMS will try to calculate the total score for each mortgage   Amount                7000)                 6000)
plan based on an algorithm with user-supplied weights.        Duration (Months)     60 (180-120)          -120 (180-300)

                                                                 Say, customer Bill has assigned weights (from 0 to 10,
                                                              weakest to strongest) to each mortgage component as fol-

                                                              Mortgage Components             Weight
                                                              Interest (%)                    5
                                                              Monthly Payment Amount          3
                                                              Duration (Months)               10
    Then, the EMS will calculate the score for each mort-         The major stackholders of EMS are mortgage
gage offer by an appropriate formula. A simple one for        applicants, property agencies, financial institutions, and
illustration purpose is as follows:                           credit repository providers. As discussed, our approach
                   N                                          can filful their major requirements. The implementation
         Score =   W
                         j   * Vj                             based on Web Service Architecture (WSA) provides a
                                                              more efficient channel to communicate among them in the
     This formula can be expressed in plain English as the    business processes, and provides an effective and efficient
EMS gets the product of weight and variations for each        way to exchange information. WS-Security further ensures
mortgage component and then sum up all the products and       the security of the Web service. Moreover, Web service
find the score for a particular offer. The system repeats     technology is a good practice to extend the value of
this process for each mortgage offer. The following table
                                                              current legacy systems. In summary, the EMS achieves
is the score for this example.
                                                              these goals with less paper exchange as well as more
                                                              automated and accelerated processes.
Mortgage        Offer A (var-       Offer B (vari-   Weight
                                                                  Since the EMS currently focuses on the mortgage
Component       iation)             ation)                    searching and ranking process, a good extension is to work
Interest (%)    2.5                 -2.5             5        on the further enhancement and automotion of the
                (0.5 * 5)           (-0.5*5)                  mortgage application workflow. We target the future
Monthly         -1500               1500             3        system to support the entire mortgage application process
                                                              in order to reach a state of paperless and all-in-one stop for
Payment         (-500 * 3)          (500 * 3)                 mortgage applications. To enhance the current built-in
Amount                                                        raking function, we are experimenting with various
Duration        600                 -1200            10       ranking formulae. We also plan to extend it into an
(Months)        (60 * 10)            (-120 * 10)              intelligent system with the aids of pattern recognition and
                                                              data mining techniques.
Total (Score)   -397.5              297.5                         Given the advantages offered by Web Services, we
                                                              propose the EMS to be delivered in two channels, namely,
    In this example, EMS will rank offer B higher then of-    Web-service interface and traditional Web site. The EMS
fer A based on their scores.                                  can be deploy as value-added services of large property
                                                              agencies, as well as an independent service that imposes
6. Discussion and Summary                                     service charges or commission.
                                                                  Further, we are working on abstracting the experience
    We have conducted a case study on an e-Morgtage           gathered from our case study to a higher level
Contract Matchmaking Service (EMS) in Hong Kong               methodology and meta-model, which employs Web
based on our consultation experiences. The current            services as a basic infrastructure to facilitate the
problem of mortgage acquistions can be described as           integration of disparate information required for efficient
follows. Mortgage applicants need to go through a long        and effective decision support. We aim at the
and complicated searching process for mortgage offeres,       improvement of various manual decision processes
during which a lot of time and effort on many repeated        significantly upon such automations in the consumers’
steps is consumed. Besides, complicated mortgage terms        perspective. In this way, decision support technologies
are always a barrier for many applicants to compare and       will no longer be monopolized priviledge only for large
select an optimal one. We propose an EMS to solve this        enterprises. This not only helps individual consumers but
particular problem. The system acts as a Web service          also could create value-added services or even new
intermediator, which accepts mortgage requests from the       business opportunities in automated service consultations
clients and formats the requests with a set of pre-defined    and brokering.
data structure based on the WSDL of other service
providers such as credit repository provider and financial
institutions. With the mortgage quotes from various           References
financial institutions, the EMS ranks and presents them to    [1] J. Ganem, "Mainfrane Web Services Development the Easy
the client, according to their weight of perferences. As a        Way", Business Integration Journal, April 2005, pp. 16-17.
result, the EMS greatly reduces the time and effort
required and provides pragmatic assistance to the             [2] E. Newcomer, "Are You Getting the Most From Your Host?
customer in decision making. This is particularly essential       Using Web Services Can Increase Mainframe System
for the Hong Kong property market (as well as loan                Value", Business Integration Journal, March 2005, pp. 43-
services) because of the keen competitions among various          45.
                                                                   Order Received

                      Check                                Req
            Enquiry               Prepare       Send                  Send          Prepare       Deliver &   Payment
  Begin    Received
                      System                               Extra                                                          End
                                 Quotation     Quotation   Info     Extra Info      Service        Install    Received

Sell Integrated System                                                 Extra

          [3] Y.C. Tse Raymond, "Relationship between Hong Kong
                                              Order                                      [16] B. Atkinson, G. Della-Libera, S. Hada, M. Hondo, P.
                                                             Assemble                Install          Test
              house prices and mortgage flows under deposit-rate ceiling
                             Begin           Missing
                                                              System                Software  Hallam-Baker, J. Klein, B. LaMacchia, P. Leach, J.

              and linked exchange rate", Journal of Property Finance,                         Manferdelli, H. Maruyama, A. Nadalin, N. Nagaratnam, H.
     System Vol.7, No.4, pp.54-63.
                        Prepare Service
                                                                                              Prafullchandra, J. Shewchuk and D. Simon, "Web Services
     Integrator                                                                               Security (WS-Security), IBM developerWorks, April 2002.
          [4] K. Gottschalk, S. Graham, H. Kreger, and J. Snell,
                                             Update                                           http://www-
                             Begin                             End
                                              architecture", IBM Systems
              "Introduction to Web services Catalog                                 
              Journal, Vol. 41, No. 2, 2002
                         Receive Part Info Updates                                            secure/

              [5] K. Mittal, "Web Services and the Real Estate Industry",                [17] W. Gentzsch. Grid computing: a new technology for the
                  Web Services Business Strategies and Architectures, April,                  advanced web. In Proceedings of the NATO Advanced
                  2002.                                                                       Research Workshop on Advanced Environments, Tools, and
                         Applications for Cluster Computing, Springer,
                  01.asp                                                                      LNCS2326:1-15, 2002.

              [6] M.J. Hudson, "XML @ Work", eAI Journal, February 2003.                 [18] J. Grimson, G. Stephens, B. Jung, W. Grimson, D. Berry
                                                                                              and S. Pardon, "Sharing health-care records over the
              [7] A.S. Bogdon, "Editor's Introduction: Expansion of                           internet," IEEE Internet Computing, vol. 5, no. 3, pp. 49-58,
                  Mortgage Markets and Increasing Use of Technology,"                         May/June 2001.
                  Journal of Housing Research, Vol 11, Issue 2.

              [8] S. Chaudhary, M. Green, R. Mahmoudi and V. Ting, "The
                  Impact of New Information Technologies on the US
                  Mortgage Industry", The Business and Information
                  Technologies (BIT) Research Project, April 2004.

              [9] Guldimann, M. Till, “Technology in Reshaping Finance and
                  Risks“, Business Economics, Jan 2000, pp.44-51.

              [10] Object Management Group. Foreword UML specification
                   1.5, March 2003.

              [11] IBM Web Services Architecture Team, Web Services
                   architecture overview, Sepembter 2000. http://www-
                   ovr/?dwzone=webservices - h3

              [12] D. Booth, H. Haas, F. McCabe, E. Newcomer, M.
                   Champion, C. Ferris and D. Orchard, "Web Services
                   Architecture", W3C Working Group Note, February 2004.
          - XML-infoset

              [13] T. Clements, "Overview of SOAP", Sun Develper Network,
                   January 2002.

              [14] J. Kao, "Overview of WSDL", Sun Developer Network,
                   August 2001.

              [15] K. Apshankar, "WS-Security Security for Web Servioce",
                   Web Service Architect, July, 2002.

To top