Data processing method, device and equipment and storage medium

Abstract

The invention provides a data processing method, device and equipment, and a storage medium. The method comprises the steps of obtaining internal historical information of a target object at a first time and external historical associated information having a mutual relationship with the target object within the first time; performing classification processing on the internal historical information and the external historical associated information based on a first neural network model to obtain a classification result of whether the target object executes a preset operation within a second time; performing prediction processing on the internal historical information and the external historical associated information based on a second neural network model to obtain an operation result of executing a preset operation by the target object in a second time; and based on the classification result and the operation result, obtaining operation information of the target object in second time,wherein the first time is time before the current time, and the second time is time after the current time. The method can reduce the maintenance cost of daily operation data, and achieves the precise quantitative prediction of recent operation indexes.

Description

technical field [0001] The invention belongs to the technical field of computers, and in particular relates to a data processing method, device, equipment and storage medium. Background technique [0002] Existing technologies usually use a time series model (prophet) for operational forecasting. A prophet model usually collects data for several months or even years, and predicts the data for the next year or several months. The model considers the influence of seasonal factors (periodical changes) s(t), the influence of holiday factors h(t), the influence of historical trend changes g(t), and introduces ∈ t Fit the abnormal points that cannot be described by the model, and finally predict the future changes of this indicator based on a known operational data indicator y. The calculation formula is as follows: [0003] y(t)=g(t)+s(t)+h(t)+∈ t , [0004] Among them, g(t) is expressed by nonlinear function and weighted linear function, and describes the most important tren...

AI Technical Summary

Problems solved by technology

[0005] The existing operation prediction method of the prophet model requires manual intervention in time: for each operation index, it is necessary to manually design holidays (periodic rules) to intervene, and to manually correct after a period of time, and the maintenance cost of operation prediction is relatively high

In addition, the prophet model needs to collect data for several months or even years, and make predictions for the next year or several months. The data collected by this model has a long interval, which is suitable for long-term trend prediction, but not for the near future (nearly Precise quantitative forecast of operating indicators in the immediate week or two or a month)

Images