Research Of Tobacco Disease Prevention And Control Model Based On Case Study Reasoning And Knowledge Map

Tobacco is an important economic crop in China,It plays an important role in the development of local economy.Pests and diseases seriously affect the tobacco production and quality.It has caused tremendous economical loss.How to control pests and diseases for reducing losses,It would make sense for growth of China’s agricultural production and economic.It was found that there are much research on the prevention and control of green technology and pest and disease diagnosis from the summary of existing research on prevention and control technology of tobacco diseases and insect pests.Formerly,lack of the pest prediction model or system for tobacco diseases and insect pests.To predict tobacco diseases and insect pests occurrence and grasp the dynamics of the occurrence of the disease.It’s important reference to use reasonable of the techniques and measures for prevention and control of tobacco diseases.This paper uses the method of case-based reasoning to set up a prevention and control model of tobacco diseases and combining the thought of Mapping Knowledge Domains in the process of case retrieval,to improve the retrieval efficiency.This model is based on the analysis of factors of external environment influence on tobacco diseases,using principal component analysis to extract the environmental factor method,as a case,attribute factor.Then use the support vector machine method to predict the future situation of these factors to get the environment of a place of tobacco.Finally according to the prediction factors to match with corresponding factors in the existing cases,searching for the most similar case.Then according to relevant information of the most similar case,judge whether the occurrence of disease and the occurrence of degree and characteristics,in order to develop reasonable and effective preventive measures.The research idea of the model is to predict firstly,control and prevent secondly,to improve the traditional methods of prevention and control of tobacco diseases,the main research contents are:(1)The principal component analysis method to screen environmental factors.Analyzing the factors of effects on the occurrence and development of tobacco disease,such as climate,soil and other external environmental factors.Take tobacco mosaic virus for example,by using the method of the principal component of the disease impact factor to reduce the dimensionality of the feature space and extract the main indexes as case attributes.(2)Use the method of support vector machine to predict the situation of factors.Using the method of support vector machine to predict the temperature range of the temperature factor in the future.The results were compared with the predicted results of all the factors that had not been screened by factor,and the factors which could express all the factors were analyzed.(3)Set up the Mapping Knowledge Domains of tobacco mosaic virus.Analyzing the data about tobacco mosaic virus,using top-down method,then set up the knowledge mapping domains of the disease.(4)The establishment of tobacco diseases prevention and control model.Based on the above research content,construction of tobacco diseases prevention and control model based on case reasoning.The knowledge map domains is applied to case retrieval,improve the comprehensive case retrieval.That is the case,attribute factor analysis using principal components extraction,calculation of attribute factor distance,get the most similar case,as prevention and control plan reference case.