Prediction method and device, electronic equipment and computer readable storage medium

Abstract

The invention provides a prediction method and device, and relates to the technical field of computers. The method comprises the steps of determining to-be-predicted time; according to the to-be-predicted time, obtaining target historical time sequence data in a first time period before the to-be-predicted time; extracting target characteristics according to the target historical time sequence data; Wherein the target feature comprises a time factor feature and an external factor feature, and the external factor feature is at least one feature except a time factor in the target historical timesequence data; and according to the target feature, determining a prediction result corresponding to the to-be-predicted time by using a preset target prediction model. According to the embodiment ofthe invention, by combining the time sequence and the machine learning model, prediction can be carried out by combining the characteristics of the external factors except the time factors in the target historical time sequence data, so that the prediction accuracy is improved, and the prediction accuracy of the unstable time sequence can also be improved.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a prediction method, device, electronic equipment and computer-readable storage medium. Background technique [0002] A time series is a set of numerical sequences arranged in chronological order, and each time in the sequence and the corresponding value constitute an element of the sequence. Time series analysis is to use this set of numbers and apply mathematical statistics to process them to predict the development of things in the future. [0003] The traditional time series forecasting method is to predict the future development based on the change trend of the source historical time series data. It decomposes the time series into the sum of trend items, seasonal items and random items, and the random items are generally stationary sequences. Commonly used forecasting methods include ARIMA, exponential smoothing algorithm, etc. [0004] In the process of applying...

AI Technical Summary

Problems solved by technology

[0005] Embodiments of the present invention provide a forecasting method, device, electronic equipment, and computer-readable storage medium to solve the problem that the existing time series forecasting method cannot be used for time series with large data changes and time series with large external influences. The problem of more accurate forecasting

Images