Based On Association Rule Mining Data Analysis And Discussion Of Agricultural Pests

With the development of large databases and the increasing cooperation between data warehouses, database researchers are faced with more and more new challenges. Meteorological department and forestry department have accumulated a large amount of observational data, while how to discover knowledge from the data and how to make it better serve for government decision-making are not only the research hotspot for experts in this field, but also the important point for decision making. Climate conditions are very important and have a great influence on plant diseases and insect pests. For instance, under certain climate condition there may be a sudden outbreak of plant diseases and pests and it may be in small pattern.This thesis studies and analyzes outlier data mining technology and its application in the data of meteorology and plant diseases and pests based on data mining technology and related technology. The research procedures are as follows:first, it introduces the related data and analyses of outlier data mining technology based on association rules, and then summarizes the types of association rules; second, it introduces an important data mining algorithm-the genetic algorithm including its emergence, development, main theories, parallel features and the wide range of application; third, researchers have discussed the specific application of the algorithms in association rules extraction; finally, it proposes the encoding method of the array of real numbers and the reason for the encoding. It has constructed the fitness function and improved the mutation, crossover and selection algorithm. In addition, it provides an example illustrating how to use the genetic algorithm to do association rule mining based on the meteorological data and the database of plant diseases and pests.