Вы находитесь на странице: 1из 2

Artificial Intelligence (AI)

Systems that exhibit the characteristics associated with intelligence in human


behavior like:
Understanding language
Learning
Reasoning
Solving problems
Most Common Application of AI Concepts:
Expert Systems (ES)
These are software programs that use facts, knowledge, and reasoning
techniques to solve problems that typically require human abilities.
Steps in the Creation of an Expert System:
1. Knowledge acquisition
2. Knowledge representation
3. Computational modeling
4. Model validation
Characteristics of an Expert System
1. Relies on a rich base of specific knowledge
2. Use thought processes in solving decisions
3. Explain the reasoning process and allow for uncertainty
4. Learns
Components of an Expert System
1. Knowledge database
2. Inference engine
3. User interface
Benefits of an Expert System
1. Consistency in developing solutions
2. Efficient
3. Valuable
4. Under distribution of knowledge of many experts
5. Knowledge is retained even when employed experts leave
6. Work 24 hours a day
Risk and Limitations of Expert Systems
1. There may be legal issues when reliance is placed on Expert Systems
in decision-making
2. They are valuable only if they can perform as well or better than a
human expert
3. They are expensive
4. They cannot think and there is a need for consensus among
experts.
Neural Networks (NN)
These are software programs that solve problems by learning from
experience, and using this learning for prediction purposes.
Components of a Neural Network
1. Nodes receives and passes data

2. Connections lines of communications


Characteristics of Neural Networks
1. Much historical data are available for system training
2. Data set may be incomplete, yet describes specific examples
3. Output is dependent on multiple, interacting parameters
4. The function that would determine the output is unknown
Case-Based Reasoning Systems
These are software programs that reason by analogy and for problems
that require human to search through historical data to find similar
problems with successful solutions.
Fuzzy Logic Systems
It deals with imprecise data and problems that have many solutions.
Genetic Algorithms
These are problem-solving methods that use evolutionary processes
Intelligent Agents
These are programs that apply a built-in or learned knowledge base to
execute a specific, repetitive, and predictable task

Вам также может понравиться