Study Of Business Acquaintances Recommendation Approach Based On Workflow Log

As the enterprise business automation needs, workflow technology is a focus of research in recent years, but the rapid development of modern society, popularity of Internet and global competitive environment, traditional workflow which is centralized information processing methods can not meet the development needs of the modern enterprise, companies need to continuously adjust and optimize business processes.Migrating workflow is a new direction in workflow management in recent years. It is constructed by the migrating workflow engine, workplace and migrating instance. Migrating workflow uses information processing in a distributed way. It overcomes some deficiency in the traditional workflow management. The migrating workflow system lacks of flexibility, making the ability of adaptive workflow system is inadequate. Based on previous studies, this paper proposed business acquaintance recommendation based on workflow logIn the general enterprise workflow systems, they always have the workflow log about Information of business processes. Workflow log records the actual implementation of historical data about migrating instances. By using the workflow log, we can guide the implementation of migrating instances better, therefore, It is important to design workflow system and implement migrating instanceThis paper introduces the business acquaintance recommendation model based on workflow log. After creating migrating instance, despite of lacking the complete of the definition of business process, it has a local knowledge of the clear task and the first workplace for migrating, so it can start the workflow. When migrating instance found that the workplace can not meet the current task demands, it will put recommendation request forward to current workplace in migrating workflow system framework, the anchor server in current workplace will use recommendation module to recommend workplaces for migrating instance by workflow log and the knowledge of acquaintances of the local workplace. The migrating instance will choose the right workplace to migrating according to the result of recommendation until the completion of tasks.Based on the above model, this paper presents a business acquaintance recommendation approach based on workflow log. This method extracts the information of migrating paths and computes matching between the migrating paths based on workflow log, and get recommendation result in three steps:acquaintance set selection, service evaluation prediction difference, workplace recommendation, based on the above information and the knowledge of acquaintance of the current workplace. Acquaintance set selection is responsible to select suitable acquaintance set from the set of business acquaintances of the current workplace as the basic of recommendation. Service evaluation prediction difference is responsible to calculate the positive evaluation prediction value and negative evaluation prediction value on workplace which is in the set created by the step of acquaintance set selection, and calculate the comprehensive evaluation value. Workplace recommendation is responsible to recommend the best workplace for instance based on the evaluation result and the acquaintance matching.Finally, this paper design and implement the recommendation component in the anchor based on the platform of migrating workflow system. Then the paper describes the architecture fully and apply it on the workshop parts assemble workflow instance. The result proved that the approach is feasible and effective. It improves the flexibility of migrating workflow system in a certain.extent.Future works will include abstracting more useful information in workflow log and deal with the exception in the recommendation.