The Approach To Local Rough Set Of Ordered Information Systems

With the advent of the era of big data,we have to face massive,multidimensional,and constantly updated large-scale complex data.Especially in the information industry,science and technology,economic and financial industry,a large number of data fragments will be generated every day and processing these tedious data will require a lot of manpower,material resources and financial resources.Therefore,when looking for the effective information needed from these data,we need to choose the most appropriate method of data mining.How to filter data and make the required information not be missed are keys to data mining,and it is also the main content of this paper.The local rough set model in the ordered information system is to delete a large amount of redundant and unnecessary information purposefully to obtain the content related to the target information.This is also a further extension of the rough set theory method,providing a theoretical basis for the processing of intelligent information.In this paper,a local rough set model is established in the ordered information system,and the dynamic objects update method of the local rough set model is studied in the dynamic environment.Furthermore,the local rough set model is used to further make decision analysis in the target concept.The presentation of local rough set theory can speed up the search for target content and facilitate people's production and life.In this paper,the main contents of the research on local rough sets in ordered information systems are as follows:1.In ordered information systems,the lower and upper approximations of local rough set model are established based on dominance relation,and their basic properties are studied under this model.Then,we give a concrete example in our daily life.From the global and local point of view,the calculation time and results are analyzed and compared.Finally,a series of comparative experiments are carried out for target concepts of different cardinality.The experimental results fully demonstrate the effectiveness of the local method.2.In ordered information systems,a local multi-granulation rough set is constructed by combining the idea of granular computing.Furthermore,combined with the current hot spots,the updating approximation method of local multi-granulation rough sets in dynamic object environment is studied.At the same time,the local dynamic object updating algorithm and the global dynamic object updating algorithm are established,and the differences between the global multi-granulation dynamic updating method and the local multi-granulation dynamic object updating method are compared through experiments,thus giving the advantages of the local updating method of dynamic objects.3.In ordered information systems,combined with the theoretical method of three-way decision,the non-probability space is transformed into the probability approximation space by the basic set assignment function,and the local decision-theoretic rough set(DTRS)model of single-granulation and multi-granulation are established respectively.Furthermore,an example is given to explain the superiority of the local DTRS method in decision analysis.Finally,local and global comparison algorithms are established.The rationality of the local DTRS model is obtained through six sets of experimental analysis in the database.