Audio album recommendation method and system based on parameter configuration

Abstract

The invention discloses an audio album recommendation method, a system and a device based on parameter configuration, and a storage medium, and the method comprises the steps: generating a recall dataset of a current Batch in near real time when the ith Batch triggered by a user arrives, and carrying out the sorting according to a number return score; fusing the recall data sets of the user in the past n times, and sorting the recall data sets according to the sequence of the generation time of each recall set; setting a fusion sorting weight parameter of each recall data set according to thedistance from the ith recall time, and rearranging all audio albums in the fused recall data set; adjusting the weight of each fusion sorting weight parameter, performing data deduplication based onan initial sorting result, and performing descending sorting again according to a final score to obtain a final recall set of the TopN; and performing fine sorting on the final recall set based on a preset fine sorting model to obtain a recommendation result. Through parameter configuration, the real-time interest trend and the historical interest of the user can be considered at the same time, sothat the recommendation result is more reasonable.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to an audio album recommendation method and system based on parameter configuration. Background technique [0002] In online music platforms, audio albums that users are interested in are usually recommended to users through the audio album recommendation system, and the most important link in the audio album recommendation system is the recommendation algorithm. The recommendation algorithm includes coarse sort recall and fine sort Ranking. According to different business characteristics, it is necessary to comprehensively consider the relationship between recall rough sorting and fine sorting. In the part of recall set generation, it is necessary to consider the user's long-term behavior, that is, the historical behavior, as well as the user's real-time behavior requirements. Both have their own advantages and disadvantages. The historical behavior update time is slow, bu...

AI Technical Summary

Problems solved by technology

Both have their own advantages and disadvantages. The historical behavior update time is slow, but the data can be considered from the overall situation. The real-time behavior update speed is fast, which can quickly reflect the real-time interest preferences of users, but the data is relatively small.

Images