User portrait determination method and device, equipment and storage medium

Abstract

The embodiment of the invention discloses a user portrait determination method and device in the field of artificial intelligence, equipment and a storage medium, and the method comprises the steps: determining an audio resource played by a target user in a target time period, and taking the audio resource as a reference audio resource; obtaining an audio coding feature corresponding to the reference audio resource, wherein the audio coding feature is obtained by processing an audio sequence of the reference audio resource through an audio coder; and according to the audio coding feature corresponding to the reference audio resource, determining an interest feature of the target user as a target audio portrait of the target user. The method can be suitable for a network platform for providing audio resources for a user, can determine the audio portrait for reflecting the preference of the user for the audio resources, and is helpful for the network platform to provide better personalized services based on the audio portrait.

Description

technical field [0001] The present application relates to the field of artificial intelligence (Artificial Intelligence, AI) technology, and in particular, to a method, device, device, and storage medium for determining a user portrait. Background technique [0002] In the context of the era of big data, providing users with corresponding personalized services based on user portraits has now become a common business model on many network platforms. In this business model, it is particularly important to build user portraits that can accurately describe user preferences. [0003] Related technologies are currently mainly used to determine user portraits in the following ways: assigning corresponding identification IDs to resources that can be provided to users on the network platform, for example, assigning corresponding IDs to resources such as commodities, videos, articles, etc.; The browsing status of the resource, and the user portrait is determined based on the ID corre...

AI Technical Summary

Problems solved by technology

[0004] However, the above-mentioned implementation method is not suitable for network platforms that provide audio resources for users, such as music network platforms; the reason is that the characteristics of audio resources are mainly reflected by audio sequences, and the user portrait determined based on the ID corresponding to the audio resources , it is usually difficult to essentially reflect the user's preferences for audio resources. Correspondingly, it is often difficult to obtain better service results for users based on the user portrait determined in this way.

Images