Research On Dynamic Three-way Decision Of Incomplete Hybrid Systems In The Background Of Streaming Data

Stream computing paradigm is a new form of dynamic computing in the background of streaming data,and it focuses on the efficient knowledge updating given immigration and emigration occur simultaneously.Existing stream computing for reasoning with three-way decisions problems is primarily aimed at symbolic and complete data,whereas the studies on more realistic cases of numerical and symbolic types with potentially incomplete data are preliminary.In addition,existing researches mainly focus on object stream computing,while dynamic three-way decisions in the setting of attribute stream computing are rarely studied.Therefore,in the background of streaming data,we study the dynamic three-way decisions in incomplete hybrid information systems for object streaming and attribute streaming,respectively.The main work of this paper is as follows.(1)Aiming at the three-way decisions problem in the object stream computing for incomplete hybrid dynamic information systems,we propose a three-way single-object neighborhood stream computing algorithm(ONS3WD)for incomplete hybrid information systems.In this algorithm,we first generate the granulation by calculating the neighborhood classes and decision equivalence classes on incomplete hybrid data.Then,with flowing into one object and flowing out of another object setting,we dynamically model all neighborhood classes and decision equivalence classes in the data streaming window.Next,we deduce the principles of variations of conditional probabilities and update three-way regions.Finally,extensive comparisons against the classical single-object non-incremental three-way decisions algorithm(ONI3WD)demonstrate that the proposed algorithm can efficiently leverage the previously acquired knowledge and show obvious advantages in time consumption.(2)Aiming at the three-way decisions problem in the attribute stream computing for incomplete hybrid dynamic information systems,we propose a three-way single-attribute neighborhood stream computing algorithm(ANS3WD)for incomplete hybrid information systems.In this algorithm,we first construct a neighborhood relationship matrix,a decision relationship matrix,a diagonal matrix,and an intermediate matrix;Then,with flowing into one attribute and flowing out of another attribute setting,we dynamically model the relationship matrix.Next,based on the modeled relationship matrix,the diagonal matrix and the intermediate matrix are dynamically updated.Finally,extensive comparisons against the classical single-attribute non-incremental three-way decisions algorithm(ANI3WD)demonstrate that the proposed algorithm can efficiently leverage the previously acquired knowledge and show obvious advantages in time consumption.The dynamic three-way decisions of incomplete hybrid systems under the background of streaming data can not only develop the theory of three-way decisions,but also provide a new research direction for solving the problem of quickly obtaining valuable information from streaming data.