Sales effect data prediction method and device, and electronic equipment

Abstract

The embodiment of the invention provides a sales effect data prediction method, and the method comprises the steps of obtaining sales effect data of a historical market activity period according to an input channel of a time sequence model, dividing the sales effect data into time intervals to form a sequence, and determining a market activity impact factor; utilizing the time interval sequence, the influence factors and the sales effect data to construct a time sequence model based on the influence factors, and utilizing the model to predict the sales effect data in combination with future market activity task attribute information. During modeling, market activity influence factors are utilized, so that the prediction process can consider the influence of market activities on sales volume; moreover, since the development of the market activity is often complex and variable, the influence factor is directly used as an independent influence factor for modeling and prediction instead of other factors indirectly reflecting the influence of the market activity, the deviation of indirect factor conversion is avoided, and during prediction, accurate prediction can be performed by directly adjusting the model input based on a market activity, so that flexibility and accuracy are improved.

Description

technical field [0001] The present application relates to the Internet field, in particular to a method, device and electronic equipment for predicting sales effect data. Background technique [0002] At present, many sales businesses often formulate sales plans in advance for management (such as adjusting the purchase volume, etc.), which involves the prediction of sales effects. At present, the methods of predicting sales effect data are mostly based on the dimension of time. The impact is used to estimate sales, especially the impact of seasons and holidays on sales performance data. In order to accurately predict sales, the existing technology automatically realizes the prediction by building a time series model, and gets rid of human subjective influence. [0003] However, the prediction accuracy of this method needs to be improved, mainly because the time series model takes time as the influencing factor, and actually uses the sales trend in the historical period to p...

AI Technical Summary

Problems solved by technology

[0003] However, the prediction accuracy of this method needs to be improved, mainly because the time series model takes time as the influencing factor, and actually uses the sales trend in the historical period to predict the future, which is too ideal and the actual situation is complex and diverse. When encountering special actual situations, the prediction accuracy is often low because it does not adapt to the actual situation. Therefore, it is necessary to propose a new method for predicting sales performance data to improve flexibility and accuracy.

Images