Wolfgang Golubski
Thomas Feuring
Johannes-Bernhard Hundhege golubski@informatik.uni-siegen.de
Universität-GH Siegen, FB 12 Elektrotechnik & Informatik
Hölderlinstraße 3
D-57068 Siegen, Germany
Universität-GH Siegen, FB 12 Elektrotechnik & Informatik
Hölderlinstraße 3
D-57068 Siegen, Germany
GAD Münster
Weseler Str. 500
D-48163 Münster, Germany
Abstract:
Introduction
Internet and Electronic commerce (E-commerce) are of increasing
interest. It is expected that billions of USD will be earned
in this new business market the next decade. The explosive growth
of E-business will change totally and fundamentally the business
environment of the 21-th century and this process has already
began.
Right now we are focusing that more and more information is
accessible via Internet. Year ago the new medium was used by
scientist only but now most companies have their own web-sides
where they offer their products as well as many further
information. Since customers need this product information in order
to decide which product should be bought, one way of getting this
is to access the Internet. We already can buy software, books and
even products from a grocery store via the Internet. But there is
no web assistance for money investment purposes (at least to the
authors knowledge). This is the aim of this paper.
We are focusing here on the development of a web-based application
for customers who would like to invest money. Online-banking
supports the movement of money from savings account to some other
account for instance but these systems do not yet support the
money investment. So, we present how customers can be guided
interactively to find out where to invest there money. Our
interactive bank agent will select the most appropriate investment
forms based on special needs of customers. In order to find the
best saving plans several questions about the personal situation
need to be answered.
Developing such a web-based application must combine several different
technical aspects
like product presentation (in form of WWW-pages), Internet communication
(in form of HTTP, IOOP), database (managing all data), and programming the
consultant (in Java).
Our online consultant system is based on client/server [5]
and the CORBA [2,3,4,6] model. CORBA is a result of a
standardization process of the OMG (Object Management Group) in which
over 800 companies (e.g. SUN, IBM, HP) are involved. CORBA models
a distributed system based on objects on heterogeneous platforms including
the needed communication infrastructure. The communication mechanism
is transparent for the programmer and easy to use. CORBA is a middle-ware
between operating systems and application programs.
The Product
One important part of the banking business is the investment of money.
A customer usually goes to his bank and says that he would like to invest money.
A human bank employer or consultant inquires about the
concrete investment amount, the period of time, the risk-taking, the
motivation (old-age retirement plan, wealth formation, house financing,..)
and a lot of other aspects. Concerning the customer's answers the consultant
proposes an investment form selected from a number of investment forms
like savings account, savings depot, fixed interest bonds, stocks, block of
shaves, life insurance, or funds.
Exactly these steps should be provided by the developed online tool.
The customer can use this tool at home to get information before going
to the corresponding bank and making the investment.
The electronic consultant should not
replace the human but should be viewed as an assistance for the customer.
One critical point in this scenario concerning the realization of the human
consulting process is the selection of the appropriate investment form.
The human decision to propose a certain investment form is based on
a long-time experience in the consultant business. This means that
human knowledge must be simulated in an appropriate way. In our electronic
consultant we have chosen the method where each customer specific data
(or preference) is rated with respect to each investment form. Therefore
we have a matrix of customer data and investment forms. Each matrix
element has a certain value which was determined before by detailed
investigations about the human consulting process in a bank. In this way
an overall rate can be evaluated for a customer investment wish.
Altogether the customer can select from more than 30 parameters
like investment amount, the period of time, and so on. More than 40 investment
forms are compared.
Program Flow
The web-based money investment consultant tool is implemented in client/server
style with CORBA and Java [1,5].
All classes specified before are defined as CORBA-objects
so that on client- and server-side a separate ORB Object Request Broker
must run which is responsible for the object communication over the Internet.
Next a step-by-step explanation of that what takes place when a
customer uses the tool is given:
The entire tool includes more functionality like the described ones, e.g.
storage of various consultant sessions or detailed
description and help functions for the customer not familiar with investment terminology.
CORBA-ConsultantProtocol.
ConsultantProtocol
over the Internet to the bank server. This happens by using the ORB on
client-side which transmits the data to the ORB on server-side.
Some Remarks
Using CORBA and its communication infrastructure have rapidly reduced the
time for the implementation phase, since the Internet communication was
very easy to program. All programmers of the tool had no experiences
in programming distributed systems. But the clearance of the CORBA-model was
helpful so that the entire development time of the
web-based money investment consultant tool was very moderate.
The web-based money investment consultant was developed for a certain
bank. The needed knowledge about the rates and the evaluation of the grades
is very specific for a certain bank. But nevertheless these information
can easily replaced by information of other banks.
Beyond this the web-based money investment consultant can be extended
so that a complete money investment contract can be made online. But then
further aspects like security of the data transmission plays a very
important rule and must be taken into consideration.
Conclusion
In this paper we have presented a web-based money investment consultant tool
which enables
customers to find the best investment strategies according to their
personal preferences.
We have described the development of such a tool.
Bibliography
Wolfgang Golubski
2000-01-24