Research And Application Of Cloud Manufacturing Service Recommendation Method Based On Domain

Cloud manufacturing is a new model of service-oriented networked manufacturing based on cloud computing model.It has the characteristics of service-oriented,high efficiency and low consumption.It is a hot topic in the field of advanced manufacturing in recent years.According to the lifecyele of cloud manufacturing service,cloud service model can be divided into service description,service composition,service matching,service recommendation and so on.The purpose of the service recommendation is to solve the problem that users can efficiently and accurately obtain the resources that meet the requirements in the massive data.However,the current service recommendation in the manufacturing field has not achieved the expected effect.How to improve the accuracy of service recommendation in the cloud manufacturing environment is an urgent problem to be solved.Based on the characteristics of the cloud manufacturing environment,combining the service recommendation process and on the basis of relevant literature research,this paper analyzes the method and accuracy of service recommendation.Through the similarity solving process of supplier service description and requirement description information,the inaccurate factors of service recommendation in the cloud manufacturing environment are analyzed,and then the targeted solutions are put forward to these factors to improve the accuracy of service recommendation in the manufacturing field.The main work of this article is as follows:Firstly,taking the traditional manufacturing service recommendation process as an example,this paper analyzes the characteristics and types of the recommended accuracy and summarizes the factors that affect the accuracy of recommendation so as to solve these problems more pertinent.Considering the ununified quality information of cloud manufacturing service and the incomplete description of manufacturing service in practical application,a domain based quality of service(QoS)description model is built on the basis of the traditional general evaluation index.Secondly,it improves the accuracy of recommendation from three aspects: different forms of service quality,lack of quality of service data,and weight setting.In the first aspect,according to the different forms of the data of service quality,standardized processing using triangular fuzzy numbers;In the second aspect,aiming at missing data of QoS, predict the missing values by classification processing.To a certain extent,the degree of sparsity is reduced objectively and effectively;In the third aspect,considering the influence of weight settings in the multi-attribute decision-making of service quality scoring,at the same time,combining the advantages and disadvantages of the subjective and objective attribute setting method,similarity computation method of QoS matching combined subjective and objective weights is proposed,which avoids the problem that the service recommendation is inaccurate caused by the unreasonable weight setting caused by the single factor.Finally,according to the improved method of service recommendation in the manufacturing field,the corresponding experimental scheme is designed,taking the recommendation of the sluice valve in the mechanical field as an example,the feasibility of the scheme design is verified,the validity of similarity computation method of QoS matching combined subjective and objective weights proposed in this paper is proved,the results are in accordance with the actual requirements and have a certain improvement in the accuracy of the optimal service recommendation results.The method also has a certain adaptability and expansibility.