Promote information effect analysis method and device, and computer readable medium

Abstract

The invention discloses a promote information effect analysis method and device, and a computer readable medium. The method comprises the following steps: constructing a network structure of a Bayesian network based on factors influencing the promote information effect and a dependence relation among the factors; constructing training data based on a value of each factor; determining a joint probability distribution of the factors based on the training data; querying the joint probability distribution of the factors to obtain the first joint probability when the query factor value of the to-be-analyzed promote information is a given value and the evidence factor value of the to-be-analyzed promote information is an observation value, and obtaining the second joint probability when the value of the evidence factor of the to-be-analyzed promote information is the observation value; and querying the influence degree on the promote information effect of the to-be-analyzed promote information by the factor when obtaining that the query factor of the to-be-analyzed promote information is the given value by computing a specific value of the first joint probability and the second joint probability. By implementing the promote information effect analysis method disclosed by the invention, the influence degree on the promote information effect by different factors can be accurately analyzed.

Description

technical field [0001] The present invention relates to communication technology, in particular to a promotional information effect analysis method and device, and a computer-readable medium. Background technique [0002] Internet promotion information (referred to as promotion information in this article, for example, advertisement), as a typical technology for information promotion on the Internet, has become an important way for users to obtain information, and for promotion information owners to increase their popularity and promote products. [0003] Taking promotion information as an example, in order to optimize the effect of advertising, the advertising system needs to know various factors that affect the effect of the advertisement, so that it can adjust the advertisement according to various factors such as bid (high or low bid), material (high or low quality), etc. Effects (such as click-through rate, advertising investment, etc.), on the one hand, for advertisers...

AI Technical Summary

Problems solved by technology

[0008] In the process of advertising, the advertising effect is affected by multiple factors, and the factors themselves are often in constant change. Related technologies cannot accurately distinguish whether the change in advertising effect is affected by one factor or multiple factors, and can only give A possible estimate, not instructive for adjusting factors later to improve ad performance

[0009] To sum up, there is no effective solution for the related technologies to accurately and efficiently determine the influence of different factors on the effect of promotional information

Images