Research On Top-k Skyline Query In Multiple Environments

With the development of big data,Internet of Things,Internet of vehicles,quickly and accurately finding out the information that people are interested in from a large amount of data has gradually became a research hotspot.Traditional Skyline queries can lead to dimensional disasters as data dimensions increase.Returning too many results does not take advantage of user decisions.Top-k Skyline query can combine the advantages of Top-k and Skyline query.Multiple dimensions are considered for dominating comparisons and the size of the returned results can be controlled.Therefore,this paper studies the problem of Top-k Skyline query in multiple environments,and the main research contents are as follows:Firstly,the Top-k Skyline query in the road network environment is studied.Existing Skyline query mostly foucus on single-user scenarios,based on single query user model to caculate Skyline results,but less consideration is given to multi-user model in the road network environment.The existing methods cannot solve the Top-k Skyline query problem that comprehensively considers multi-user preferences and weights in the road network environment.Therefore,this paper proposes a Top-k Skyline query method MUP-TKS based on multi-user preference in the road network environment.In this environment,the different preferences and weights of multi-user are considered for Skyline calculation.The result set which conforms to the preference and weight of the query user group can be obtained quickly for the user group to make a better decision.Through the proposed pruning filter algorithm G＿DBC,the position relation of data points and query points in the road network and the new index structure Vor-R~*-DHash are used for pruning the data points.Thus,obtained the optimal distance set.Then take advantage of the invariable property of the static Skyline set,precomputed the collection,and then save it.The KPRD algorithm is performed on S set,the union of the optimal distance set and static Skyline set.Finally,the TK＿DC algorithm is used to score the candidate set.According to the score of the data points,the Top-k of the sorted set are returned to the user group.Theoretical studies and experiments show that the proposed method is efficient and reliable.The Top-k Skyline query in edge computing environment is further studied.In order to make up for the shortcomings of the existing research in dealing with the Top-k Skyline query over multiple incomplete data streams in edge computing environments,we propose the EDI-TKS framework in edge computing environments to deal with the problem of Top-k Skyline query over multiple incomplete data streams first.The framework first uses the correlation between attributes to construct a Bayesian network,filling in missing values in incomplete data.And the problem is transformed into uncertain data streams for further processing.Then the global skyline probability of an object is computed iteratively using edge nodes and cloud servers,returning the first k objects with high Skyline probability.At the same time,LR-tree index and GR-tree index are proposed to solve the problem of high computational overhead of Skyline probability.Using the dominant region of the object and index structure to obtain and update the Skyline probability of the object reduces the computational overhead,thus improves the response speed.Aiming at the problem that stream data needs to obtain Top-k Skyline sets continuously,we propose the LTST and GTST structures.Using these two structures to maintain and transfer necessary objects,thus reduce communication overhead and computational overhead.And get the updated result set of data streams at the current time quickly by dynamically maintaining these two structures.So as to improve the efficiency of query processing.