Application of the most popular analogical reasoni

  • Detail

The application of analogical reasoning in the computer management system of ink analysis

in the application field of ink composition analysis, there is often a certain ink that does not meet the use requirements. If the ink samples that do not meet the use requirements can be made into qualified ink by adding additives, blending, etc., the limited ink resources can be fully utilized, which is an ink application business with great economic benefits

due to the complexity of ink components, it is difficult to predict the change direction of ink quality after adding additives and blending. Therefore, the treatment methods for unqualified specific inks and specific indicators have great blindness. This paper presents an analogical reasoning system based on data mining technology - ink analysis computer management system (oacms) to guide the correction of ink indicators

1. Basic idea and solution model

heuristic search technology based on analogical reasoning is to obtain the solution process of past problems similar to new problems through analogy, which can be used as heuristic information to guide the solution of new problems, so as to reduce the blindness of search, narrow the search scope and reduce the difficulty of problem solving

generally, heuristic search based on current analogical reasoning directly applies the case base oxfab process, and the specific strength of forming materials is better than cast aluminum, magnesium and nylon as the search space. The result of such treatment will lead to two problems:

1. The performance of the system will continue to decline during the operation of the system. With the increase of solving times and the accumulation of solving cases, the search space will continue to expand. Although this can give the system enough opportunities to obtain solving cases of past problems similar to solving problems, it will also lead to the saturation of heuristic search information, reduce the efficiency of the system and deteriorate the performance

2. the performance of the system will largely depend on the tradeoff between the search cost and the heuristic effect of cases. The content recorded in the case base is incomplete, which can reduce the requirements for the similarity of cases, and the matching process may be easier, but the heuristic effect of case guidance in the past will be weakened; If the content recorded in the case database is too detailed, the requirements for the similarity of cases are relatively high, the matching process is difficult, and the search cost is increased. It may also cause the system to rely more on weak methods to solve

in real life, human beings do not store past experiences in their minds, but use a general data structure, that is, there is a process of mining useful information from past case data. This is an effective method to continuously solve problems, accumulate experience and enhance problem-solving ability. The solution considered in this paper is to use data mining technology to generalize and abstract hierarchical association rules from a large number of case base data, and the system searches for matching rules from the rule base to solve it. Because data mining produces hierarchical solution information, the system can reduce the search cost, minimize the impact of information loss, and maintain the search quality of the system, so as to ensure that the efficiency of problem solving will continue to improve with the increase of running time. The solution model is shown in figure (solution model)

the experience and knowledge accumulated in the past can be divided into two categories: shallow case base and rule base generated by data mining in case base. At the beginning of the system operation, there is a lack of solving cases in the case base, and there are a small number of simple past experience rules in the rule base, so the system should use the weak method to search and solve. When the system has accumulated some case experience, the solution of the new problem is first searched in the rule base. If the rule is successfully applied,

figure: solution model

figure: the case base records the solution process of past cases; The rule base stores the processing rules generated after the abstract processing of the data in the case base; The control strategy stipulates how to judge the similarity of events; The analogical reasoning mechanism stipulates how to conduct business processing under specific similarity, and completes the addition, modification and deletion of case base, rule base and control strategy

then the solution is successful. Otherwise, search the case base. If there is a solution to similar problems, use this as heuristic information to guide the solution of new problems. If there is still no similar past problem solving process, the weak method is used to solve the problem, and the new problem solving process is added to the case base. The more experience accumulated in the past, the more effective the application of rule base; The higher the similarity between the new problem and the past problem, the less the search times of the solution process

second, the system implementation

in the process of sending the angular displacement signal to the computer for data processing at the same time, we need to solve ① how to generate solution examples; ② How to define and judge the similarity of two cases; ③ How to abstract general rules from a large number of cases; ④ How to maintain rule base and case base; ⑤ How to use heuristic information to guide new problem solving

Author: Liu Zhipeng and Ke Zhi Xu Shaohua

this article is from the network, and the copyright belongs to the original author. It is only for everyone to share and realize the experiment and learning of simple supported beam and cantilever beam. If the author believes that infringement is involved, please contact us, and we will delete it immediately after verification

Copyright © 2011 JIN SHI