CF(R1,R2) = [CF(R1)] x [CF(R2)]
CF(R1,R2) = [CF(R1)] x [CF(R2)]
Here we have deleted this value from the sum of the two certainty factors, assuming an independent relationship between the rules. For example,
Given CF(R1) = 0.7 AND CF(R2) = 0.6, CF(R1,R2) = 0.7 + 0.6 - (0.7)(0.6) = 0.88
That is, the ES tells us that there is an 88 percent chance that stock prices will increase. Note: If we just added the CFs of R1 and R2, their combined certainty would be larger
than 1. We modify the amount of certainty added by the second certainty factor by multiplying it by (1 minus the first certainty factor). Thus, the greater the first CF, the less the certainty added by the second. But additional factors always add some certainty.
CHAPTER 11 KNOWLEDGE ACQUISITION, REPRESENTATION, AND REASONING
For a third rule to be added, the following formula can be used: CF(R1,R2,R3) = CF(R1,R2) + [CF(R3)] [1 - CF (R1,R2)]
= CF(R1,R2) + CF(R3) - [CF(R1,R2)][CF(R3)]
Assume that a third rule is added: R3: IF bond price increases,
THEN stock prices go up (CF = 0.85). Now, assuming all the rules are true in their IF part, the chance that stock prices will go
up is CF(R1,R2,R3) = 0.88 + 0.85 - (0.88)(0.85) = 0.982 That is, there is a 98.2 percent chance that stock prices will go up. Note that CF(R1,R2) was
computed earlier as 0.88. For a situation with more rules, we can apply the same formula incrementally. See Russell and Norvig (2002) for more information on uncertainty and probabilistic ES.
An expert system is basically computer software-an information system; so itsdevelopment follows a software development process. The goal of a development process is to maximize the probability of developing viable, sustainable software within cost limitations and on schedule, while managing change (because introducing a new, computerized approach or even implementing software to perform an existing method of working involves change). The availability of expert system shells and other development tools has enabled rapid prototyping in system development (see Chapter 6). The idea of rapid prototyping, which involves performing the analysis, design, and implementation phases concurrently and repeatedly in short cycles, ultimately leading to deployment of the system, is feasible. Prototyping is also known as iterative design or evolutionary development. It is the most commonly used DSS development methodology. One of its advantages is that decision-maker(s) and user(s) learn more about the problem they are trying to solve while performing the prototyping process, as occurred in the Opening Vignette and in the case applications.
Using the prototyping approach to developing ES includes the following six major phases (as illustrated in Figure 11.16).
Phase I: project initialization Phase II: system analysis and design Phase III: rapid proto typing and a demonstration prototype Phase TV: system development Phase If: implementation Phase VI: postimplementation.
PHASE I: PROJECT INITIATION
Project initialization is the first step in developing expert systems. Its main goal is to identify problems and prepare for further actions. Major tasks that must be done in this
PART IV INTELLIGENT DECISION SUPPORT SYSTEMS
Problem definition Need assessment
Evaluation of alternative solutions Verification of
Phase I
an expert systems approach Feasibility study
Project initialization
Cost-benefit analysis
Consideration of managerial issues Organization of the development team
Phase II System
Conceptual design and plan
analysis and
Development strategy
design
Sources of knowledge Computing resources
Building a small prototype
Phase III
Testing, improving. expanding
Rapid prototyping
Demonstrating and analyzing feasibility Completing design
Phase IV
Completing the knowledge base
System development
Testing, evaluating, and improving knowledge base Planning for integration
Acceptance by users Installation. demonstration, deployment
Phase V
Orientation. training
Implementation
Security Documentation Integration, field-testing
Phase VI
Operation
Postimplementation
Maintenance and upgrades Periodic evaluation
Problem definition. Identify problems that are suitable for systems development. Need assessment. Determine the functional requirements of the system.
Evaluation of alternative solutions. Examine the availability of experts, education and training, packaged knowledge, conventional software. Verification of an expert system approach. Futher verify the requirements, justify the needs, and assess the appropriateness of developing such a system. Feasilibity study. Evaluate whether the project is technologically feasible, man- agerially feasible, and change can be initiated and sustained. Cost-benefit analysis. This last part of the feasibility study involves determining
L economic feasibility: formally identifying and estimating the potential costs and benefits of the system. Many criteria including internal rate of return (IRR) or net present value (NPV) can be used for evaluating ES projects. For details, see Chapter
CHAPTER 11
KNOWLEDGE ACQUISITION, REPRESENTAtiON, AND REASONING
Managerial consideration. Examine managerial issues, including project'initiator,
financing, resources, legal and other constraints, selling the project, identifying a champion, potential of gaining access to users and user support.
Organizing the development team. The team must be assembled and commitment established before' starting formal development because of the nature of the
prototyping process. Ending milestone. Approval of the project in principle.
PHASE II: SYSTEM ANALYSIS AND DESIGN
Once the concept of the project has been approved, a detailed system analysis must be conducted to estimate system functionality. The major tasks in this phase are explained
below.