Understanding Expert Systems: Bridging Human Knowledge and Technology
An expert system represents a collection of software designed to tackle problems within a specific domain that typically necessitates human expertise. It harnesses encoded knowledge to facilitate solutions, leveraging information derived from expert sources and translating it into a format conducive to the system's inference and reasoning processes. This knowledge is sourced from specialists or authoritative references like texts, journals, and databases. Frequently, this category of knowledge requires thorough education and practical immersion in particular disciplines such as medicine, geology, system configuration, or engineering design. Once a substantial repository of expert knowledge is amassed, it undergoes encoding, insertion into a knowledge base, and continuous testing and refinement throughout the system's lifecycle.
Expert systems possess distinctive characteristics that set them apart from conventional computer systems in significant ways.
1. Expert systems prioritize knowledge over data to drive the solution process, echoing the sentiment that knowledge is paramount. This underscores the significance of insightful information in guiding decisions and actions. Much of the knowledge employed tends to be heuristic, emphasizing practical insights and rules of thumb rather than rigid algorithms.
2. The knowledge within an expert system is encoded and managed independently from the control program. This means that it is not integrated into the control program during compilation. Consequently, updates and refinements to the knowledge base can be made incrementally without requiring the recompilation of the control programs. Additionally, in certain instances, it is feasible to employ different knowledge bases with the same control programs, thereby generating various types of expert systems. These systems are often called expert system shells, as they can accommodate different knowledge bases.
3. Expert systems can explain the rationale behind specific conclusions and the necessity for requested information during consultations. This functionality holds significance as it enables users to evaluate and comprehend the system's reasoning process, thereby enhancing the user's trust in the system's capabilities.
4. Expert systems utilize symbolic representations such as rules, networks, or frames to embody knowledge, and they execute their inference processes through symbolic computations that closely mimic the manipulation of natural language. (One notable exception to this pattern is expert systems founded on neural network architectures.)
5. Expert systems frequently engage in reasoning with metaknowledge, which involves the utilization of knowledge about themselves, including their own limitations and capabilities.
Applications
1. Medical Diagnoses (Internal Medicine, Pulmonary Diseases, Infections Blood Diseases)
2. Diagnosis of complex electronic and electromechanical systems.
3. Planning experiments in biology, chemistry and molecular genetics forecasting crop damage
4. Location of faults in computer and communications systems.
5. Analysis of structural systems for design or as a result of earthquake damage.
6. Estate planning for minimal taxation and other specified goals.
7. Stock and bond portfolio selection and management
8. The design of very large-scale integration (VLSI) systems.
9. Numerous military applications ranging from battlefield assessment to ocean surveillance.
10. Planning curricula for students.
No comments:
Post a Comment