Expert System Experiment (draft)

Today, I got caught up with a duck. MM asked me to help her write an "Expert System" lab report. This is really hard for me. It took me a few hours to prepare the first draft. Ah ~ The phrase "books are used to hate less" is the truth! No matter how bad it is, let's first let ggjj give us some comments. My understanding of the expert system is still superficial. This time, I shot my head ...... Help me more ~~~

Expert System Experiment

Expert systems are an important branch of early artificial intelligence. They can be seen as a kind of computer intelligence with special knowledge and experience.ProgramSystems generally use knowledge representation and Knowledge Reasoning techniques in artificial intelligence to simulate complex problems that can be solved by field experts.

Application fields:
Expert systems are suitable for the diagnosis, interpretation, monitoring, prediction, planning and design of unrecognized theories and methods, inaccurate data or incomplete information, a shortage of human experts, or very expensive expertise.

Generally, the expert system = knowledge base + inference engine. Therefore, the expert system is also known as a knowledge-based system. An expert system must have three elements:
1. expert knowledge in the field
2. simulate expert thinking
3. Reaching the expert level

According to the above introduction, the "Expert System" experiment is divided into three parts:
1. Create a knowledge base
2. Database Management System implementation
3. Simple Inference Engine Design

Lab 1 Knowledge Base Creation

Knowledge in the knowledge base comes from domain experts. It is a collection of domain knowledge required for problem solving, including basic facts, rules, and other relevant information.

Steps:
1. Knowledge Acquisition
The knowledge involved in the professional field.
2. Expression of knowledge
The content in the Knowledge set obtained by step 1, including qualitative information and quantitative information. For qualitative information, rules must be formulated to express the meaning of knowledge. For example: the expression of "missing element" must be properly converted, and the final expression is "when the element content is small, it is missing element "; for quantitative information, rules need to be formulated to express the meaning of knowledge. For example, "ph6.9" can be expressed as "acidic ". It does not mean that all information needs to be converted. Here, it is only for the information to be converted. Therefore, this step is also the focus of creating a knowledge base.
3. Create a knowledge base
Create a knowledge base in the database management system.

Experiment 2 database management function implementation

Database management mainly includes routine database operations (query, addition, modification, and deletion), database maintenance, database backup, and disaster recovery. The following describes how to perform routine operations.

Steps:
1. Open the visual programming environment. Visual FoxPro is used here.
2. Implement the "Create Database Connection" Function
3. Implement the "database query" Function
Including: (1) Single Table query; (2) Multi-Table query
3. Implement the "database update" Function
Including: (1) Add; (2) modify; (3) Delete

Experiment 3 simple Inference Engine Design

The inference engine is the core implementing mechanism for implementing problem solving. It is a program for interpreting knowledge and interpreting and executing the knowledge found according to certain strategies according to the meaning of knowledge, and record the results to the appropriate space of the dynamic library.

Inference strategies include:
1. Positive (data-driven)
2. Reverse (target driver)
3. Bidirectional
In this experiment, a forward inference policy is used to implement the inference engine program, which has nothing to do with the specific content of the Knowledge Base. Modifications to the knowledge base do not need to be modified.

Steps:
1. fully understand the existing knowledge of the knowledge base and propose the problems to be solved
2. Create a rule repository
3. Obtain the available rule set from the rule repository
4. Determine search control policies (for example, evaluate function policies)
5. Select the optimal rule from the rule set through the search control policy
6. Execute the optimal rules and update the knowledge base.
7. Repeated steps
8. obtain a solution or no solution