Method and device for forecasting changes of feature information

Abstract

The invention aims at providing a method and device for forecasting changes of feature information. The method includes the steps of obtaining historical feature data of multiple pieces of the first feature information corresponding to a forecasting model in at least one counting period and current incremental data, wherein the current incremental data are used for indicating a ratio of feature data of the first feature information on the day before the forecasting day to the historical feature data in the at least one counting period; obtaining first change information of the second feature information on the forecasting day, wherein the second feature information is determined by the forecasting model based on the historical feature data of the first feature information corresponding to the forecasting model and the current incremental data after forecasting processing; determining change forecasting information of the second feature information on the forecasting day according to the first change information to prompt a user to execute corresponding operation based on the change forecasting information.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to an estimation device for predicting changes in feature information. Background technique [0002] In the prior art, the prediction of the future change trend of some information is usually determined based on the historical data of the information. However, since the change of information is usually affected by many other factors, for example, the retrieval of some keywords The amount of clicks may be affected by factors such as the placement area and time of these keywords. For example, the number of clicks on some links may be related to factors such as the search volume of the keywords contained in the link and the presentation area of ​​the link. Therefore, only based on the historical data of the information itself, it is impossible to make a good prediction of its future change trend. Especially when the user frequently adjusts some of these factors, for example,...

AI Technical Summary

Problems solved by technology

[0002] In the prior art, the prediction of the future change trend of some information is usually determined based on the historical data of the information. However, since the change of information is usually affected by many other factors, for example, the retrieval of some keywords The volume may be affected by factors such as the placement area and time of these keywords. For example, the number of clicks on some links may be related to the retrieval volume of keywords contained in the link and the presentation area of ​​the link.

Therefore, only based on the historical data of the information itself, it is impossible to make a good prediction of its future change trend

Especially when the user frequently adjusts some of these factors, for example, the user adjusts the desired keyword placement area and time, etc. At this time, it will be more difficult to predict and estimate the future trend of relevant information

Images