Multimedia file recommendation system and delivery information storage method

Abstract

The invention discloses a multimedia file recommendation system and a releasing information storage method and belongs to the technical field of network. The system comprises a service server cluster, a recommendation engine cluster, a pushing server cluster and a distributed storage cluster. The service server cluster is used for receiving releasing information of a multimedia file, and sending the releasing information to the pushing s server cluster and the distributed storage cluster. The pushing server cluster is used for pushing the received releasing information to the recommendation engine cluster. The recommendation engine cluster is used for storing the received releasing information into the local cache of the recommendation engine cluster, and carrying out multimedia file recommendation according to the releasing information. The distributed storage cluster is used for storing the releasing information. According to the invention, by adding the pushing server cluster in the multimedia file recommendation system, storage modes of the releasing information are changed, multimedia file recommendation effects are improved and multimedia file recommendation accuracy is improved.

Description

technical field [0001] The invention relates to the field of network technology, in particular to a multimedia file recommendation system and a storage method for delivery information. Background technique [0002] With the continuous development of network technology, the way to display multimedia files such as advertisements is no longer limited to traditional media such as TV and newspapers, but can also be extended to network media such as websites and network applications. When displaying multimedia files on the network, in order to improve the pertinence of multimedia file display, multimedia files may be recommended for users. For example, for advertisement recommendation, when determining which advertisement to provide on which advertisement position, it can be based on some user characteristics such as user age and gender. Time period, target audience, etc., for more purposeful recommendations. [0003] see figure 1 , the existing multimedia file recommendation s...

AI Technical Summary

Problems solved by technology

[0004] However, when the local cache stores delivery information, it will set a timeout period. When a piece of delivery information reaches the timeout period, the delivery information will be set as invalid, resulting in no query and low query efficiency. As a result, multimedia files The recommendation effect is poor, and the recommendation accuracy is also greatly reduced

Images