Personnel recommendation method based on heterogeneous graph nerves

Abstract

The invention provides a personnel recommendation method based on heterogeneous graph nerves, and relates to the technical field of recommendation methods. According to the method, the event characteristics of the event and the ability characteristics of the personnel are combined, the heterogeneous graph of the event and the personnel is constructed, and through training of the heterogeneous graph neural network, a plurality of to-be-recommended personnel can be recommended for the new event finally. The event allocation strategy is used for allocating events to a large number of new events, so that the event handling efficiency can be effectively improved, and the event overstock is avoided. Meanwhile, the system provides potential handling ability of mining personnel, and can perform priority allocation under certain conditions for event types which are never handled or less handled by the personnel, so that the learning enthusiasm of the personnel is effectively aroused, and the multi-dimensional handling ability of the personnel is improved.

Description

technical field [0001] The present invention relates to the technical field of recommendation methods, in particular to a person recommendation method based on heterogeneous graph neural. Background technique [0002] The event distribution system is an important system for event management and distribution. The random distribution event mode is to randomly distribute the event to specific undertakers according to certain random distribution event rules after the event is established; the designated distribution event mode is established by the According to the specific circumstances of the event, special personnel assign the event to specific undertakers. The former has the advantage of being open and transparent in the process of assigning events, and the results of the assigning events are objective and random. It can eliminate the possible impact of human factors on the process of assigning events. It is simple, fast, objective and fair, but cannot flexibly identify the ...

AI Technical Summary

Problems solved by technology

Many scholars have conducted research on how to achieve the optimal matching of events and personnel, but the current research is only at the stage of qualitative analysis. In addition to listing the specific goals that intelligent distribution events need to accomplish, they have not specifically given the impact of intelligence. What are the factors that assign event outcomes

Images