Processing methods and devices of multimedia information and servers

Abstract

The embodiment of the invention discloses processing methods of multimedia information. One of the methods includes: receiving a first exposure request, wherein the first exposure request is used forrequesting sending of the multimedia information for a first user set; in response to the first exposure request, sending the first multimedia information for the first user set according to an HWM (High Water Mark) algorithm, and obtaining exposure quantity of the first multimedia information by counting; sending the exposure quantity of the first multimedia information to a first server to obtain total exposure quantity of the first multimedia information by counting; receiving excess notification information from the first server, wherein the excess notification information is used for characterizing a case where the total exposure quantity of the first multimedia information is greater than or equal to a preset threshold value; and in response to the excess notification information, stopping sending of the first multimedia information. The embodiment of the invention discloses processing devices of the multimedia information and servers at the same time.

Description

technical field [0001] The present invention relates to the field of information processing, in particular to a method, device and server for processing multimedia information. Background technique [0002] Contract advertising, also known as Cost Per Mille (CPM) advertising, needs to ensure that the advertiser can obtain the traffic of the audience users agreed in the contract. Inventory estimation is the basis of the contract advertising system. It estimates the inventory of each advertising space in each targeting dimension (platform, region, gender, age, interest, etc.) at various times in the future based on historical data. The estimated results will be provided to the sales distribution and online distribution modules. [0003] At present, when contract advertisements are distributed online, in order to achieve the purpose of maintaining the quantity, the commonly used method is the High Water Mark (HWM, High Water Mark) algorithm. Since the distribution probability...

AI Technical Summary

Problems solved by technology

Since the distribution probability calculated by the HWM algorithm is based on the estimated inventory data, if the estimated inventory is too small, the distribution probability will be too high. At this time, if the distribution is made according to the distribution probability, it is easy to overplay the advertisement; if If the inventory estimate is too large, the distribution probability is too small. At this time, if the distribution is made according to the distribution probability, the advertisement playback will be short

Images