Research On Big Data Analysis Method From The Perspective Of Context Theory

As a new scientific research method,big data analysis urgently needs a comprehensive and systematic philosophy theory as support and guarantee.Therefore,we need to discuss the philosophy of big data analysis from ontology,epistemology and methodology.Context theory is a scientific world view based on the development of modern natural science and social science.At present,it has played an effective role in solving many philosophical and scientific problems.The study of big data analysis method from the perspective of contextualism can make us grasp the cognitive mechanism,methodological characteristics and significance of big data analysis more thoroughly,and help us to solve the theoretical problems encountered in the practical application of big data analysis.First of all,in the data world view,data has become the key attribute of things in the context of data science.Datifying everything becomes an effective way to understand the world.In this ontological attitude,the object of big data analysis becomes data reality,in fact,its existence is guaranteed by the objectivity of the objective world and the stipulation of its context.Secondly,the cognitive process of big data analysis is contextualized,and its cognitive characteristics come from the characteristics of the induced context: while massive data shows the context of the research problem,it also reflects the integrity and dynamics of the context,which enables us to recognize the objective world in a dynamic perspective in the complex relationship between problems and problems;the theoretical system of the current scientific context and the practical operation of the technical context provide constructive and objective contextual cognitive features for big data analysis respectively;the basic scientific rationality is relatively and contextualized in big data analysis,showing the characteristics of unity with situational rationality.Thirdly,from the perspective of knowledge production of big data analysis,the multiple cognitive subjects of data analysis lead to the change of the role of subjects in the research context,which makes human purpose and practice run through the whole process of big data analysis producing knowledge;the knowledge output from big data analysis does not have the ability to explain and demonstrate the reasons,however,in the micro and "regional" context,it can still lay a solid foundation for us to master the multidimensional and multi-faceted internal relationships among things.Fourthly,looking forward to the application prospect of big data analysis method: the advantages and disadvantages of data-driven research paradigms coexist.therefore,it is necessary to combine the multi methodology of contextualism for discussing the norms of using data paradigm in what research context;the contextualized cognitive features of big data analysis provide a theoretical and feasible way for AI to break through the limitations of Turing machine model and simulate human's irrational cognitive ability and the intentional ability of subject object interaction;the adaptability of big data analysis method to different discipline theories and the compatibility of research methods among various disciplines meet the needs of context theory to eliminate the rigid barriers between disciplines and build a unified methodology in the diversity of research methodsFifthly,looking at the application dilemma of big data analysis,the contradiction between the ideal and reality of "all data" comes from the unrestricted concept of "all data".We should re-understand and define "all data" in the limited role of context,retain its ideal and accept the "noncomprehensive" reality;in the face of the blind worship of correlativity,we need to face up to the causality hidden in the r correlativity results,which is beyond the scope of interpretation in the current human scientific context,and maintain a practical and critical attitude towards the correlativity results of big data.