《人工智能（智能系统指南，第二版）》读书笔记——10、第九章

1. introduction to knowledge-based intelligent systems(summary / questions for review / references)

2. rule-based expert systems

3. uncertainty management in rule-based expert systems

4. fuzzy expert systems

5. frame-based expert systems

6. artificial neural networks

7. evolutionary computation

8. hybrid intelligent systems

9. knowledge engineering and data mining

9. knowledge engineering and data mining

Knowledge engineering is the process of building intelligent knowledge-based systems. There are six main steps:

1)Problem assessment:

2)Data and knowledge acquisition

incompatible data; inconsistent data; missing data;//数据不兼容; 数据不一致; 数据缺失

3)Development of a prototype system(原型系统)

helps us to test how well we understand the problem domain and to make sure that the problem-solving strategy, the tool selected for building system, and the techniques for representing acquired data and knowledge are adequate to the task

Will an expert system work for my problem?

case study 1: diagnostic expert system(diagnosis and troublesshooting problems)

case study 2: classification expert system(classification problems, 有先验知识作依据)

Will a fuzzy expert system work for my problem?

case study 3: decision-support fuzy system(assessing mortgage applications<评估抵押申请> problems, 有精确知识作依据)

Will a neural network work for my problem?

case study 4: character recognition neural networks(character recognition problems)

case study 5: prediction neural networks(real-estate appraisal<房地产估价> problems)

case study 6: classification neural networks with competitive learning(character recognition problems, 无先验知识作依据)

Will genetic algorithms work for my problem?

case study 7: the travelling salesman problem(produce optimal itineraries<最优路线> problems)

Will a hybrid intelligent system work for my problem?

case study 8: neuro-fuzzy decision-support systems(diagnose myocardial perfusion from cardiac images, clinical notes, physician's interpretation, 无精确知识作依据)

case study 9: time-series prediction(predit an aircraft's trajectory at least 2 seconds in advance, ANFIS)

Data is what we collect and store, and knowledge is what helps us to make informed decisions. The extraction(抽取) of knowledge from data is called data mining. Data mining can also be defined as the exploration and analysis of large quantities of data in order to discover meaningful patterns and rules.

The most popular tool for data mining is a decision tree - a tool that describes a data set by a tree-like structure. Decision trees are particularly good at solving classification problems. The main advantage of the decision-tree approach to data mining is that it visualises the solution; it is easy to follow any path through the tree. The tree's ability to produce clear sets of rules makes it particularly attractive for business professionals.

case study 10: decision trees for data mining(results of a health survey, which people are at a greater risk of having high blood pressure)