Research On Self-construction Algorithm Of IoT Service System Based On Keyword Search

With the repid development of the Internet,it has gradually evolved from an information release platform in the traditional sense to an open distributed computing infrastructure.In2002,the service-oriented architecture(SOA)was put forward,making "service" the core concept of resource encapsulation and sharing in an open network environment.However,with the increasing complexity of cross-enterprise applications,a single web service can no longer meet the needs of complex business.It is necessary to build a complex service system by combining multiple services that meet different functions to meet logic requirements on certain business process.In addition,with the development of light-weight smart devices,embedded technology for web services,and data integration technology,people's research interest has also begun to turn to Internet of Things(Io T)services.The Io T service can connect objects in reality to the Internet,and can be operated and transmitted in the form of network services through a combination of service-oriented middleware and Internet protocols.However,the process of building a service-based system is more complicated,and only experts can complete it.Therefore,simplifying the process of building a service-based system and automatically building a system for service composition is a problem worthy of study.The automatically-built algorithm of the service-based system based on keyword search is a better solution.Users only need to enter a set of simple keywords to get a set of service-based systems that reflect their preferences.The existing keyword-based search solutions are divided into two categories: one is graph theory-based service composition solution;the other is database-based solution.The existing solutions generally have some problems,such as low algorithm time efficiency and poor solution shrinkability.In response to the above problems,this paper designs an automatically-built algorithm for Io T service-based systems based on Skyline query and group steiner tree,and an automatically-built algorithm for lightweight Io T service-based systems based on database.The main research contents of this paper are as follows:(1)An automatically-built of Io T service-based system based on Skyline query and group steiner tree is proposed.The solution is based on graph theory,which enables users to realize the function of automatically constructing a service-based system when only a set of keywords(including the set mandatory keywords)are provided.Compared with the traditional service composition solution based on keyword search,the solution in this paper is more adaptable and less restrictive,and can achieve functions equivalent to a complete service combination containing all query keywords.In addition,in order to improve the time efficiency of the algorithm,the algorithm also designs 3 pruning strategies,which reduces unnecessary time consumption to a certain extent.Finally,a comparative experiment proves the effectiveness of the pruning strategy.(2)An automatically-built algorithm for a lightweight Io T service-based system based on a database is proposed.The solution takes into account the space pressure of the storage system and query time loss,and a relational database is designed for storage services.Secondly,in order to meet the requirement of recommending several service-based systems by providing user-given keywords and Qo S parameters,this algorithm constructs a special scoring mechanism--kscore,which evaluates the degree of matching between candidate services and input keywords;Similarly,in order to evaluate the quality of the automatically constructed service-based system,the algorithm also designs a quality scoring mechanism--qscore.Finally,in order to meet the needs of quickly building a service-based system,on the basis of database query technology,three progressive automatically-built algorithms for service compositions based on database query are designed to produce an optimal algorithm.Comparative experiments show that the algorithm has a significant improvement in time efficiency.