Transcript The Intellectual Development and Structure of Decision

The Changing Structure of DSS Research:
An Empirical Investigation through Author
Cocitaion Mapping
The 2004 IFIP International Conference on Decision Support Systems (DSS2004)
Decision Support in an Uncertain World
Introduction
 DSSs are a relatively young field of study. As a field of study
continues to grow and becomes a coherent field, study of the
intellectual development of the field is important (Culnan (1986).
 Researchers can benefit by understanding this process and its
outcomes because it
 reveals the vitality and the evolution of thought in a discipline
 gives a sense of its future
 identifies the basic commitments that will serve as the
foundations of the field
 It atrophies if it cuts itself off from curiosity, diversity, and reflection"
For DSS to become a coherent and substantive field, a continuing
line of research must be built on the foundation of previous work.
Without it, there may be good individual fragments rather than a
cumulative tradition (Keen, 1980).
This research assesses on-going changes
in the intellectual development and
structure of DSS research.
 Emphasis on contrasting the structural
changes over the period of:
- 1969 through 1990 and
- 1991 through 1999
DATA
 A total of about 1600 citing articles in the DSS
area
632 -- 1969 through 1990 (28.7 articles/year)
984 -- 1990 through 1999 (98.4 articles/year)
A total of about 25,000 cited references taken from the
citing articles
Research methodology
Author cocitation analysis (ACA)
"a set of data gathering,, analytical, and graphical display
techniques that can be used to produce empirical maps
of prominent authors in various areas of scholarship"
The tools used in ACA
factor analysis
multidimensional scaling
cluster analysis.
Research methodology
The Assumptions in Author cocitation analysis
(ACA) “cocitation is a measure of the perceived
similarity, conceptual linkage, or cognitive
relationship between two cocited items
(documents or authors).”
“Cocitation studies of specialties and fields yield
valid representations of intellectual structure.”
Steps in ACA
1. Selection of Authors
2. Retrieval/compilation of Cocitation Frequencies
3. Multivariate Analysis
4. Interpretation
When does the cocitation occur?
When a citing paper cites any work of
authors in reference lists
 Ref. of Paper #1







Ackoff
Bonczek
Bonczek
Blanning
Blanning
Blanning
Whinston
Ref. of Paper #2
Ackoff
Ackoff
Applegate
Applegate
Whinston
Ref. of Paper #3
Ackoff
Ackoff
Blanning
Sample Cocitation Matrix
Ackoff
 Ackoff
 Applegate
 Bonczek
 Blanning
 Whinston
Applegate
Bonczek
Blanning
Whinston






Ackoff
Ackoff
Applegate
Bonczek
Blanning
Whinston
1
1
2
2
Applegate Bonczek Blanning
0
0
1
2
1
2
Whinston
Author Cocitation Analysis
"a set of data gathering, analytical, and
graphical display techniques that can be
used to produce empirical maps of
prominent authors in various areas of
scholarship" (McCain 1990)
Can you see the trunk, branches, and the roots of the tree?
ACA: A Tool for Digging Up the Roots, Trunks,
Branches
Results
Factor analysis of the data (1990-1999) extracted
11 factors
Six major areas of DSS research
1. group support systems 2. design 3. model
management 4. implementation/user interfaces
5. Evaluation 6. multiple criteria DSS
Five contributing disciplines
1. cognitive science 2. computer supported
cooperative works 3. organizational science
4. social psychology 5. MCDM
Systems Science
Organization Science
Cognitive Science
Implementation
User Interface
Foundations
Model
Management
Artificial Intelligence
MCDSS
MCDM
GDSS
Communication
Organization Science
Artificial Intelligence
Psychology
Theory ,
Applications,
and Contributing
Disciplines of DSS
Organizational goals
Improving effectiveness of the DM’s problem-solving process
A specific DSS
J Contributing disciplin
Decision maker(s)
J1.Artificial
Intelligence
J2.Cognitive
Science
B Functional app.
Interface
Marketing
DSS
POM
DSS
Financial
DSS
Data
C
Design
F
Model
D
Decision
maker(s)
SDSS
GDSS
ODSS
E
Implementation
G
J3.Communication
Theory
H
Evaluation
I
Model
Interfaces
(Dialogue)
MCDSS
J5.Organization
Science
J6.Systems
Science
J7.Psychology
data
KBDSS
J4.
MCDM
What has happened/is happening in DSS
research since 1991?
the DSS area has undergone profound structural changes
in the intellectual structure over the past 10 years (19901999).
-- steady, strengthening, emerging, dying, and slowly
growing areas.
The steady areas
model management
organizational science
multiple criteria decision making
Artificial Intelligence
What has happened in DSS research?
The Strengthening Areas
GSS
The Emerging Areas
Design
Implementation
Cognitive science
The Dying Areas
Foundation and Individual Differences
(appeared to be no longer active)
Group DSS
The result of this study clearly shows that GDSS has become a central
part of DSS research area over the past five years.
Some of the important recent developments
(1) There have been continuing developments and
enhancements of GDSS tools to support and augment
the existing group DSS and electronic meeting systems
such as the following:
 -- An idea consolidator
 -- An optimization-based group decision tool for
combining subjective estimates and extracting the
underlying knowledge of group members
Group DSS
-- A group software for modeling and analyzing
business process re-engineering;
 -- An interactive videodisk-based GDSS for
directing the pattern, timing, and contents in
group decision making;
 -- A prototype GDSS for multicultural and
multilingual communication to translate among
several foreign languages such as English,
German, Korean and Spanish;

Group DSS
(2) A wide
range of GDSS/electronic meeting
systems/decision conferencing system
applications has been reported to
support/facilitate the following areas:
-- Strategic management meetings
 -- Quality improvement process
 -- Knowledge acquisition for multiple experts
 -- Distributed decision making involving fairly
large numbers of participants (tens to
hundreds);

Group DSS
-- Developing a cognitive map of users of
object-oriented techniques for understanding
individual and group perceptions [48];
 -- Developing national economic policy [49];
 -- Expediting the requirements specification
in the system development process
 -- Facilitating the United States Army’s group
decision making in geographically distributed
environments ; and

Group DSS
(3) GDSS
is being integrated with other
technologies such as ES and case-based
reasoning, etc.
A prototype system that embedded ES into
GDSS is developed to make a GDSS a more
user-friendly and powerful tool
The distributed artificial intelligence approach for
designing and developing group problem
solving systems is being investigated to
coordinate organizational activities in a
distributed environment
GDSS Empirical Research Model (U. of
AZ)
Group
Task
Context
EMS
Process
Outcome
Model management
Some of notable approaches include the following:
(1) Development of Graph-based modeling:
Jones presented a prototype system of graphbased modeling, NETWORKS, which allows
the user to represent a wide variety of decision
problems in a graphical form such as bar chart,
decision tree, decision network, etc.. Further,
the users manipulate the models (e.g..,
deleting/adding subtrees for decision trees)
using a graph-grammar by applying a set of
operations (or productions).
Model management
(2) Object-oriented approach:
Using the object-oriented framework, Muhanna
[37] designed and implemented a prototype
model management system to build, store,
retrieve composite models and to maintain the
integrity of model bases through providing the
functional capability of model sharing,
integration, and reusability. e.g., a DSS formulates
and integrates optimization and simulation modeling and
heuristic reasoning for non-expert users through an
object-oriented domain-specific knowledge base.
Model management
(3) Modeling by analogy (Analogical modeling) is
suggested as potentially fruitful avenues for increasing the
productivity of model formulation. This approach is using a
process by which model X is constructed based on a
known model for problem X and the similarity
between problems X and Y [31].
(4) In addition to a knowledge base which stores facts
and rules, case- based reasoning systems maintain a
case base which is a repository of all previous cases
solved.
To find a solution for a new problem, the system
identifies the most similar case from its case base to
be applied.
Model management
(4) Active modeling systems are expert
systems embedded modeling systems
which provide intelligent support to the
modelers. e.g. a knowledge-based linear
programming (LP) model construction
system
Model management
(5) Integrating model management and
inductive machine learning in an adaptive
decision support system.
by incorporating machine learning capabilities
for model management .
the system adapts itself to the environment
through continuously updating and refining the
knowledge-base .
(6) Model integration using metagraph: the
process of model integration can be
significantly expedited by utilizing certain
connectivity properties in metagraph.
User Interfaces
Intelligent agents (a.k.a. intelligent
interfaces, adaptive interfaces) research
is an emerging interdisciplinary research area
involving researchers from such fields as expert
systems, decision support systems, cognitive
science, psychology, databases, etc.
The primary purpose of agent research is
to "develop software systems which engage
and help all types of end users" in order to
reduce work and information overload, teach,
learn, and perform tasks for the user.”
A NEW GENERATION OF
ACTIVE/INTELLIGENT DSS
WHAT A
GREAT
IDEA!
A SYSTEM THAT
BRINGS
TOGETHER THE
ANALYTICAL
STYLE OF DSS
AND THE
JUDGMENTAL
CONCLUSION AND IMPLICATIONS
FOR FUTURE DSS RESEARCH
This study identified a dynamic dimension
of DSS research areas to account for the
ongoing changes in its "disciplinary matrix"
-- the four emerging areas
(Implementation, Design, and Cognitive
science); continuously growing areas
(GDSS, Model management, MCDM, and
Organization science); and dying areas
(Individual difference and Foundations).
 Focus of DSS research appears to be shifting
from the study of DSS components (data, model,
individual differences of decision makers) during
the periods of 1970 through 1990 to the design,
implementation, and user-interface management
(which have not been shown to be substantive
DSS research subspecialties in the previous
research), to provide useful guiding principles for
practitioners in the integrated processes of
design, implementation, and evaluation of
decision support systems.
CONCLUSION AND IMPLICATIONS FOR
FUTURE DSS RESEARCH
World Wide Web-based DSS is another emerging topic in the
DSS area. The World Wide Web is increasingly being used as
the client/server platform of many business organizations due
to its network and platform-independence and very low
software/ installation/maintenance costs. The web-based
solutions are low cost vehicles for easily accessing, analyzing,
and distributing timely business information from corporate
databases through OLAP.
The Internet and corporate intranets opened a wide possibility of
building global DSS/interorganizational DSS to deal with
problems of global natures. As we enter the age of the global
village where geographical and temporal boundaries are
shrinking rapidly, global DSS/Interorganizational DSS support
systems are emerging as the new frontiers in management
information systems area.
Number of
Operating
Countries
1970's
1980's
1990's
2000's
Multiple
Global
DSS
Group
Support
Systems
Single
single-user
DSS
Single Decision Maker,
Single Organization
Enterprise/
Organizational
DSS
Web-based DSS
Inter-Organizational
DSS
Group DSS
group decision makers,
single organization
Cross-functional decision
Support for an
organization
Extended enterprises
decision support
Number of
Organizations