Resource allocation method and device

Abstract

The invention discloses a resource allocation method and device, and relates to the technical field of computers. One specific embodiment of the method comprises the following steps: acquiring a historical reservation amount of each reservation entrance of a resource; inputting the historical reservation amount into a neural network model, and calculating a predicted reservation amount of each reservation entrance of the resource based on an optimal network weight through the neural network model; and determining the quantity of resources allocated to each reservation entrance according to thepredicted reservation quantity. According to the implementation mode, the resources allocated to each reservation entrance can be accurately and reliably predicted, the resource allocation situationis adaptively adjusted, the prediction accuracy and time-varying property are high in prediction of the service reservation number source with mutability, the service capability of a service institution is fully played, the operation efficiency of the service institution is improved, and the user experience is greatly improved.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a resource allocation method and device. Background technique [0002] At present, many service resources can be used through online appointment application, such as medical resources. Online appointment service has solved most of the basic medical needs, which greatly saves the registration time of ordinary people. For medical institutions, on the basis of assessing their own medical output capabilities, it is necessary to set the number source for multiple appointment portals of different channels of cooperation, and allocate the total reception capacity within a fixed period to multiple appointment entrances. [0003] Among the existing solutions, one is to manually adjust the current number source allocation based on experience. This method has too much error. If the number source setting is too large, it will cause a waste of medical resources. Other reservation en...

AI Technical Summary

Problems solved by technology

[0003] Among the existing solutions, one is to manually adjust the current number source allocation based on experience. This method has too much error. If the number source setting is too large, it will cause a waste of medical resources. Other reservation entrances have reservation needs but cannot Make an appointment; if the number source setting is too small, the user experience will be poor. Medical institutions can provide medical services, but cannot make an appointment because the number source pool is exhausted

The second is to predict the number of possible medical visits in the future, and allocate the number source according to the number of medical visits, but this method ignores the causal relationship between the past and the future, and the reliability of long-term appointments is not high

The third is to approach any complex nonlinear problem through BP (BackPropagation) neural network, but the accuracy and time-varyingness of the prediction are not high in the prediction of the abrupt change of the dynamic medical service appointment number source

[0005] Existing solutions have defects such as large prediction error, low reliability, accuracy, and low time-varying, which reduce user experience

Images