Research On Resource Service Sequence Mining And Optimization Method In Collaborative Tasks

With the acceleration of the global economic integration process,the collaborative task model has been more and more widely applied.Supported by emerging information technologies such as cloud computing and big data,the collaborative task platform manages business processes and resource services more efficiently because it coordinates different organizations and accomplishes a task together.The provision of resource services in the form of "service flows" to business processes is more conducive to improving the overall efficiency of the business process.However,the distribution of resource services and the autonomy of each organization's selection of resource services are bad for the efficiency of the business process.In order to improve the efficiency of resource service selection and further improve the overall efficiency of the business process,with the support of workflow technology,we proposed the resource service sequence mining and optimization methods considered the correlation between resource services,the similarity between resource service sequences(RSS)and the classification of resource services.The main research works of this paper are as follows:(1)Mining resource service sequences based on similarity for collaborative tasks.For collaborative tasks,in order to improve the efficiency of resource service selection,a method of mining resource service sequences based on similarity is proposed.First,since the dependencies of resource services can be represented by the frequency of use of resource services,an equation is presented to calculate the distances between resource services according to the frequency of use of resource services.Then,according to the distance of the resource services,the relationship of the resource services in the two resource service sequences is obtained through dynamic programming,we propose a recursive calculation formula for the similarity between resource service sequences.Finally,the method of judging the frequent resource service sequence is introduced.The experimental results show that the proposed method can improve the efficiency of resource service selection.(2)Resource service sequence mining method for abstract granularity optimization.Aiming at the variety of resource services and the difficulty of mining useful resource service sequences,this paper proposes a resource service sequence mining method for abstract granularity optimization.Firstly,according to the resource service classification tree,the resource service is represented by a vector under different abstract categories.Secondly,according to the hierarchical relationship of classification trees in different abstract categories,the similarity and discrimination of resource services are analyzed.Finally,based on the similarity and discrimination between the resource services and the frequency of use of resource services,we propose the best abstract class representation of the resource service.The experimental results show that the proposed method can improve the mining effect of resource service sequences.Finally,in order to prove the feasibility of the proposed method,the proposed method is verified on the platform based on the service platform.