A service recommendation method for service supply chain network

Abstract

The invention discloses a service recommendation method oriented to a service supply chain network. Firstly, a transfer matrix is ​​constructed based on the link relationship between all nodes in the service supply chain network, and the reputation values ​​of all nodes are calculated according to the transfer matrix; and then two-two nodes are calculated. The similarity between reputation enhancements is used to construct the neighbors of all nodes; finally, for users who need recommended services, predict the service quality evaluation value of the user for the required service according to the neighbors of the corresponding nodes, and recommend high service quality evaluation values ​​to them services. The service recommendation method of the present invention can successfully predict the user's service quality evaluation value for the required service when the historical data is insufficient or some nodes do not have historical data, and overcomes the data sparsity and cold start existing in the existing collaborative filtering technology problem, improving the accuracy of service recommendation.

Description

technical field [0001] The invention belongs to the technical field of supply chain configuration, and in particular relates to a service recommendation method oriented to a service supply chain network. Background technique [0002] In the field of supply chain configuration, it is generally believed that its development trend is to increasingly emphasize the sharing and integration of various external high-quality services (such as resources and knowledge). The exponential growth of the Internet and the accelerated globalization of manufacturing enterprises have significantly affected the interoperability of cross-enterprise collaboration services. For example, the SOA architecture and the corresponding Web services not only provide a dynamic implementation for the communication between manufacturing enterprises and their partners (including suppliers and customers) through the seamless integration of various complex distributed manufacturing resources and knowledge techn...

AI Technical Summary

Problems solved by technology

However, the recommendation effect of collaborative filtering technology is easily affected by data sparsity and cold start problem.

When historical data is insufficient to find similar users or items, data sparsity problems will arise at this time

When a new user or item does not have historical data, it is impossible to find similar users or similar items, which will cause a cold start problem

Images