Research On Traffic Information Service Composition Algorithm Based On Graph Planning Via Knowledge Graph

With the rapid development of Internet technology,Traffic Information System has gradually increased their level of intelligence through the integration of Internet technology.However,in the face of ever-increasing traffic demand and an increasingly complex traffic environment,a single simple service can no longer meet the passenger's need for convenient travel.Therefore,how to reuse existing fine-grained services,through specific regulations for service composition to construct the compound service conform to the diversification of user requirements,is the focus of research in the field of Intelligent Traffic Information Services.In this paper,through analyzes the characteristics of the current traffic information service environment,constructs a Knowledge Graph-based traffic information service think tank,and builds a traffic information service composition method based on graph planning service composition framework to provide a high-quality traffic information service solutions that meet the needs of users.The key work and innovations in this paper can be summarized by the following three points:(1)For the problem that the graph planning service composition algorithm can not effectively reuse the service relationship,this paper introduces the Knowledge Graph of the latest research results of the Semantic Network into the field of traffic information services,and proposes a traffic information service description framework based on OWL-S through the research on the construction of the Knowledge Graph.And then on the basis of the description framework proposes the construction of traffic information service think tank.Through the Neo4 j graph database,the traffic information service think tank based on the Knowledge Graph is built to support the graph planning service composition algorithm to reuse the service relationship and improve the execution efficiency.(2)For the problem that the graph planning composition algorithm cannot effectively measure the quality of the plan solution.This paper designs a heuristic graph planning service composition method based on service multi-index fusion model with the support of service think tank.This method guides the algorithm's service optimization selection through state-quality heuristic functions.(3)For the problem that the graph planning service composition algorithm solves theproblem with low efficiency under the large-scale service set,this paper applies the ant colony algorithm to the solution extraction algorithm of the graph planning service composition algorithm to improve the parallelism of the solution process.In the traffic information service environment,in order to overcome the shortcomings of traditional ant colony algorithm such as slower initial convergence,easy to fall into local optimal solution and inability to measure service quality,the optimal trajectory enhancement strategy,pheromone smoothing strategy,and service metrics evaluation model were proposed.A graph planning composition service composition optimization algorithm based on improved ant colony algorithm is designed.Based on the above research,this paper proposes a traffic information service composition framework based on service think tank.Through a specific traffic travel scenario,the feasibility and effectiveness of the service composition framework are verified.The experimental results of the composition algorithm involved in the paper are tested.