The Internet Encyclopedia (Volume 3)

P1: IML/FFX P2: IML/FFX QC: IML/FFX T1: IML

WL040-Schalkoff WL040/Bidgolio-Vol I WL040-Sample.cls July 16, 2003 15:52 Char Count= 0

244 RULE-BASED ANDEXPERTSYSTEMS

CORVID Logic Blocks may be integrated with forward

or backward chaining. Furthermore, Logic Blocks are

built and maintained in a visual development environ-

ment. The underlying knowledge representation in the

block is an if–then rule.

CommonRules

Although not strictly a simple ES, the IBM Common-

Rules framework is highly related to both e-commerce

and rule-based systems. In 1999, IBM developed an

application called CommonRules 1.0: Business Rules for

E-Commerce. Its goal was to make a program available

that would allow businesses to communicate their “busi-

ness policy rules about pricing, promotions, customer

service provisions for refunds and cancellation, ordering

lead time, and other contractual terms & conditions, to a

customer application/agent.” An overview may be found

at http://www.research.ibm.com/rules/commonrules-

overview.html. According to IBM,

CommonRules 3.1 is a rule-based framework for

developing rule-based applications with empha-

sis on maximum separation of business logic and

data, conflict handling, and interoperability of

rules.

... it provides a platform that enables the
rapid development of rule-based applications
through its situated rule engine via dynamic and
real-time connection with business objects. Com-
monRules can be integrated with existing appli-
cations at a specific point of interest, or it can
be used to create applications composed only of
rules. (IBM Common Rules, n.d.)

Each business involved in e-commerce probably im-

plements its own set of rules, implemented using many

different languages. Because these representations are

different, communicating business policy to customers

or other businesses becomes challenging. The purpose

of CommonRules is to eliminate this difficulty by us-

ing a semantically rich rule language called CLP (Cour-

teous Logic Program). CLP incorporates many rule sets

and a rule-based computational model. XML (extensible

markup language) is the string format used for CLP, which

makes it a natural format for the Web.

CommonRules 1.0 was developed in Java and runs

on the user’s computer. It accesses the Internet when in-

voked. When the user wishes to look at a seller’s avail-

able business rules, the user downloads these rules and

runs them through CommonRules to output a file with

a set of user-readable rules. This eliminates the cus-

tomer’s needing to have the same rule representation of

the seller to view its business policies. The current ver-

sion of CommonRules is 3.0 and can be found at http://

alphaworks.ibm.com/aw.nsf/download/commonrules.

CONCLUSION

Many internet applications may be enhanced with the per-

ception of intelligent behavior. This “intelligent” enhance-

ment may be achieved using a rule-based or agent-based

computational paradigm. Examples include e-commerce,

diagnosis, and compliance monitoring.

GLOSSARY

Agent Encapsulated and task-focused software struc-

ture.

Antecedent The “if” part of a rule.

Certainty factor One method to incorporate uncer-

tainty in the inference process.

Conflict resolution Process of choosing which rule or

rules to use in the inference process.

Consequent The “then” part of a rule.

Expert system Software intended to emulate human ex-

pertise.

Heuristics “Rules of thumb” intended, but not guaran-

teed, to aid in the solution of a problem.

Inference engine Production system controlling soft-

ware.

Knowledge representation The paradigm chosen to

encode knowledge (e.g., rules and facts).

Modus ponens A logical basis for rule-based inference.

Multivalued logic A logical system with more than two

truth values.

Production (rule-based) system A conceptual and

computational paradigm useful for building intelligent

and expert systems.

Recommender system Software that tracks previous

purchases and extrapolates to future recommenda-

tions.

Rule If–Then structure used to represent expertise or

knowledge.

CROSS REFERENCES

SeeIntelligent Agents; Machine Learning and Data Mining

on the Web.

REFERENCES

Bolc, L., & Borowik, P. (1992).Many valued logics—1. The-

oretical foundations.Berlin: Springer.

Buchanan, B. G., & Shortliffe, E. H. (1984).Rule-based

expert systems: The MYCIN experiments of the Stanford

Heuristic Programming Project.Reading, MA: Addison-

Wesley.

Burke, R. (2000). Knowledge-based recommender

systems. In A. Kent (Ed.),Encyclopedia of Library and

Information Systems(Vol. 69, Suppl. 32). New York:

Marcel Dekker.

Grosof, B., Levine, D., Chan, H., Parris C., & Auerbach

J. (1995). Reusable architecture for embedding rule-

based intelligence in information agents.Proceedings

of the Workshop on Intelligent Information Agents, ACM

Conference on Information and Knowledge Management

(CIKM-95).New York: ACM Press.

Harmon, P., & King, D. (1985).Expert systems—artificial

intelligence in business.New York: Wiley.

Hicks, R. (2000, January/February). New Trends in ES de-

velopment and implementation.PC AI Magazine, 14,

37.