Analysis Of Environment And Biology Information Based On Machine Learning And Nonlinear Theory

For any entity,its status are related to its surrounding environment,so extending information mining and utilization to analysis of environment information can benefit further reveal on laws of entities.Machine learning is one of most important technology for the basis of artifacial intelligence in this centrenry;nonlinear theory can conduct research from a holistic and systematic view.Therefore,taking advantage of these two method to study environment information can dig the laws of environment information deeply and show more comprehensive interaction between environment and entities.Based on a large amount of relevant literature at home and broad,this paper focuses on "environment information" to conduct studies and has obtained some valuable results.This paper includes four main parts:(1)Analyzing external environment information.For studing automatic drive,environment information was quantified into gravitational field,so that a basic rule and a strong rule were gained to form a guiding method for automatic drive.It shows the utilization of environment information for entities behavior.(2)Analyzing external representation of microenvironment.Since many medical indicators are the external representation of microenviroment,indicators of colorectal cancer has been studied.Based on FCM cluster,this papar put forward a variable cluster method with scene cognition,screening out meaningful variables to train a BP network.In result,a forecast model was obtained.By external representation mining,this part shows the microenvironment's influence on the whole body.(3)Analyzing microenvironment.With nonlinear and chaos theory,we went further to gene study.This paper,based on concept of landscape,put forward a neighborhood-compression gene repair method,which is of stability and compatibility,presenting a new gene repair thinking.(4)Analyzing the interaction between external environment and microenvironment.Starting with construction of a time-based gene regulatory network,gene expression time order was gained.After comparing expression time order and location order,the temporal-spatio consistency was discovered.It demonstrates that external environment,changing with evolution process,has influenced on genes' activation and production.Further,it is deduced that gene location can repaly evolution process.In summary,from external environment to microenviroment,this paper analyzed interaction between environment and entities with machine learning and nonlinear theory.The results provide an instructive method and thinking for future research.