Method and equipment for predicting quantity of daily active objects

Abstract

The embodiment of the invention discloses a method and equipment for predicting the quantity of daily active objects. The method provided by the embodiment of the invention comprises the steps of predicting a first object reflux amount of a (M+X)-th day via an object reflux model, wherein the object reflux module is acquired through modeling after statistically analyzing a reflux ratio of a firstobject in the daily active objects of continuous (N+1) days from the M-th day to the (M-N)-th day, wherein the first object is the newly added object before the (M-N)-th day; predicting the retainingamount of a second object of the (M+X)-th day via an object retaining model, wherein the object retaining model is acquired through performing data analysis according to the retaining amount of the second object in the quantity of the daily active objects within (N+1) days; and determining the quantity of daily active objects of the (M+X)-th day according to the reflux amount of the first object of the (M+X)-th day in the quantity of the daily active objects of the (M+X)-th day and the retaining amount of the second object of the (M+X)-th day.

Description

technical field [0001] The invention relates to the field of computers, in particular to a method and device for predicting the number of daily active objects. Background technique [0002] The number of daily active user terminals (Daily Active User, abbreviation: DAU), usually counts the number of user terminals that log in or use a certain product within one day (statistical day) (excluding repeated login user terminals), and is often used to reflect the website , the operation of Internet applications or online games. Different products may have different definitions of activity. For example, if it is a client product with an account, account login is usually used as an activity indicator, such as QQ, IM, and online games. If it is some tool application, it can be regarded as active by starting, for example, Meitu Xiuxiu, etc. Such applications need to take at least one photo or use the photo editing function once to be considered active. [0003] DAU is a very importa...

AI Technical Summary

Problems solved by technology

However, due to the impurity of old user terminals, in a relatively mature Internet product, the existing old user terminals may include periods such as launch period, shock period, and stable period, and the trend of existing old user terminals at each stage in the existing old user terminals Therefore, a linear model is established by obtaining the number of old user terminals in memory for a certain period of time. This linear model is not accurate in predicting the stock of old users. The proportion is relatively high, so the overall DAU prediction accuracy is very low

Images