Introduction to AI key facts and definitions

Backward Chaining Mechanism

- First, the knowledge base is searched to find rules that might have the desired solution. - Such rules must have the goal in their THEN (action) parts. - If such a rule is found and its IF (condition) part matches data in the database, then the rule is fired and the goal is proved. - However, this is rarely the case. - Thus the inference engine puts aside the rule it is working with (the rule is said to stack) and sets up a new goal, a subgoal, to prove the IF part of this rule. - Then the knowledge base is searched again for rules that can prove the subgoal. - The inference engine repeats the process of stacking the rules until no rules are found in the knowledge base to prove the current subgoal.

Expert System Major Tasks

- Obtain required knowledge from expert - Formulate knowledge as a set of rules in the form of logical implications (knowledge base) - Extract conclusions (reasoning)

Programmer

A person who is responsible for the actual programming, describing the domain knowledge in terms that a computer can understand.

End-user

A person who uses the expert system when it is developed.

Database

A set of facts about the current situation.

Domain Knowledge

A set of if-then production rules.

Knowledge Base

A set of rules describing knowledge of a specific domain.

Forward Chaining

A technique for gathering information and then inferring from it whatever can be inferred. This is the data-driven reasoning.

Reasoning

A mechanism for selecting the relevant facts and extracting conclusions from them in a logical way.

Conflict Resolution

A method for choosing a rule to fire when more than one rule can be fired in a given cycle.

Rule

Can have multiple antecedents, conjunctions (AND), disjunctions (OR) or a combination of both.

Consequent (1)

Can have multiple clauses (e.g. IF w, THEN x, y, ... , z).

Inference Engine (1)

Carries out the reasoning whereby the expert system reaches a solution - It links the rules given in the knowledge base with the facts given in the database.

Antecedent (1)

Consists of an object (linguistic object) and its value that is linked by an operator.

Consequent (2)

Consists of an object (linguistic object) and its value that is linked by an operator.

Aggregation of Rule Outputs

Finding the output of x number of rules.

Computers

Have high computation, but low recognition and reasoning.

Humans

Have high recognition and reasoning, but low computation

Operator

Identifies the object and assigns the value (e.g. is, are, is not, are not, =, >, <)

Metaknowledge

Knowledge about the use and control of domain knowledge in an expert system. It is represented by metarules.

Knowledge Representation

Knowledge that is formulated into a series of statements in the form of if-then expressions or production rules.

Conduct decision making in a fuzzy environment

Model goals and constraints by fuzzy sets.

Experts

People who possess knowledge. They have deep knowledge and strong practical experience of the subject or domain they specialise in.

Hedges

Terms that modify the shape of fuzzy sets: very, more, quite, less, slightly, likely, very likely

Intelligence

The ability to learn, understand and make judgements or have opinions that are based on reason.

Project Manager

The leader of the expert system development team, responsible for keeping the project on track.

Uncertainty

The lack of exact knowledge that would enable us to reach a perfectly reliable solution.

Production Rules

These are a set of if-then expressions.

Programming Language

This is means of representing knowledge.

Consequent (THEN)

This is the conclusion or action as a result of the antecedent being met.

Procedural Knowledge

This is the knowledge about how to perform some task.

Antecedent (IF)

This is the premise or condition of an event being met.

Defuzzification

This is the process of evaluating the rules and producing an aggregate output fuzzy set and is a single number.

Firing a rule

This occurs when its condition part is satisfied and its action part is executed.

Sources of Uncertainty

Weak implications - Vague associations between IF (condition) and THEN (action) parts of the rules. Imprecise language - It can be difficult to express knowledge in the precise IF-THEN from of rules. Unknown data - When data are incomplete or missing, the only solution is to accept the value "unknown" and proceed to an approximate reasoning with this value. Combining the views of different experts - - Experts seldom reach exactly the same conclusions. - They often have contradictory opinions and produce conflicting rules.

μ^slightly(x)

[μ(x)]^1.7

μ^very (x)

[μ(x)]^2

μ^extremely(x)

[μ(x)]^3

μ^more or less (x)

√[μ(x)]

Backward Chaining

An expert system has a goal and the inference engine attempts to prove it. This is the goal-driven reasoning. This is the most common inference method found in expert systems. An inference engine backchains by determining the highest priority goal from user specifications, then asking questions (collecting facts) about rules in order to find a rule or rules that lead to the goal

Boolean Logic

Only allows for truth or falsehood (1 or 0)

Disadvantages of rule-based expert systems

Opaque relations between rules. Although the individual production rules are relatively simple and self-documented, their logical interactions within the large set of rules may be opaque. Rule-based systems make it difficult to observe how individual rules serve the overall strategy. Ineffective search strategy. The inference engine applies an exhaustive search through all the production rules during each cycle. Expert systems with a large set of rules (over 100 rules) can be slow, and thus large rule-based systems can be unsuitable for real-time applications. Inability to learn. In general, rule-based expert systems do not have an ability to learn from the experience. Unlike a human expert, who knows when to "break the rules", an expert system cannot automatically modify its knowledge base, or adjust existing rules or add new ones. The knowledge engineer is still responsible for revising and maintaining the system.

Domain Expert

A knowledgeable and skilled person capable of solving problems in a specific area or domain.

Inference Engine (2)

The part of the rule-based expert system that compares each rule stored in the knowledge base with facts in the database.

Decision Making

The process of problem solving involving pursuing of goals under constraints.

Knowledge

The theoretical or practical understanding of a subject or domain - the sum of what is currently known.

Artificial Intelligence

The science of making machines do tasks that would require intelligence if done by humans.

Factors affecting the decision making process

incomplete and imprecise information subjectivity linguistics

Forward Chaining Mechanism

- The reasoning starts from the known data and proceeds forward with that data. - Each time only the topmost rule is executed. - When fired, the rule adds a new fact in the database. - Any rule can be executed only once. - The match-fire cycle stops when no further rules can be fired.

Outcome

A decision which should result in an action.

Linguistic Variable

A type of fuzzy variable (e.g. the linguistic variable John takes the value tall).

Fuzzy logic

Determined as a set of mathematical principles for knowledge representation based on the degree of membership rather than on crisp membership of classical binary logic.

Metarule

Determines a strategy for the use of task-specific rules in the expert system.

Heuristics

If A and A < x and A is B, then A is C.

Recommendations

If A and B and C, then advice is D.

Directive

If A and B, then action is C.

Relations

If A, then B.

Strategies

If A, then action is B. Step 1 is complete. If step 1 is complete and B is 0, then C. Step 2 is complete.

Cons of Forward Chaining

Many rules may be executed that have nothing to do with the established goal. This means that if our goal is to infer only one particular fact, then the forward chaining inference technique would not be efficient.

Advantages of rule-based expert systems

Natural knowledge representation. An expert usually explains the problem-solving procedure with such expressions as this: "In such-and-such situation, I do so-and-so". These expressions can be represented quite naturally as IF-THEN production rules. Uniform structure. Production rules have the uniform IF-THEN structure. Each rule is an independent piece of knowledge. The very syntax of production rules enables them to be self-documented. Separation of knowledge from its processing. The structure of a rule-based expert system provides an effective separation of the knowledge base from the inference engine. This makes it possible to develop different applications using the same expert system shell. Dealing with incomplete and uncertain knowledge. Most rule-based expert systems are capable of representing and reasoning with incomplete and uncertain knowledge.

Knowledge Engineer

Someone who is capable of designing, building and testing an expert system.

Expert Systems

Software packages designed to assist humans in situations in which an expert in a specific area is required.

Inference Chains

The constant matching of if parts to the facts

Probability

The probability of an event is the proportion of cases in which the event occurs. It's a scientific measure of chance. It can be expressed mathematically as a numerical index with a range between zero to unity (absolute certainty).

Universe of Discourse

The range of all possible values applicable to a chosen variable.

Rule is fired and its then part is executed

When the if part of the rule matches a fact.

Fuzzy Set

You belong to a certain set to a certain degree (e.g. an element x belongs to a set with a certain degree of membership).

Crisp Set

You either belong to a certain set or not (e.g. an element x belongs to X or does not belong to X).