Method and device for displaying target object sequence to target user

Abstract

The embodiment of the invention provides a method and device for displaying a target object sequence to a target user. The method comprises the steps of determining the initial sequence of all targetobjects in the target object sequence to be displayed; obtaining a superposition coding vector of any target object according to the attribute feature of the target object and the position of the target object in the initial sorting; inputting each superposition coding vector into a pre-trained reordering model according to an initial order, with the reordering model comprising an encoder and a decoder, the encoder encoding each target object one by one by using a multi-head attention mechanism based on each superposition coding vector to obtain an implicit feature vector corresponding to eachtarget object, enabling the decoder to select a target object from the current unselected target objects as the decoding output of the current step by using a multi-head attention mechanism accordingto each hidden feature vector and the decoding output of the previous step; and taking the decoding output sequence of each step as the reorder of each target object. User feedback is maximized.

Description

technical field [0001] One or more embodiments of this specification relate to the computer field, and in particular to a method and device for presenting target object sequences to target users. Background technique [0002] At present, the target object sequence is often displayed to the target user to recommend multiple target objects in the target object sequence to the target user. In order to achieve a specific business goal, it is necessary to filter out a large number of target objects that the user may be interested in Multiple target objects, forming a target object sequence by these multiple target objects, and determining the position of each target object when displaying the target object sequence are all performed based on the sorting of the target objects. [0003] In the prior art, when sorting the target objects, it is usually sorted according to the correlation between each target object and the target user from high to low. The above correlation reflects t...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, when sorting the target objects, it is usually sorted according to the correlation between each target object and the target user from high to low. The above correlation reflects the target user's preference for the target object. When displaying target object sequences, it is often impossible to maximize user feedback, for example, to maximize the target user's click rate for each target object, or to maximize the conversion rate, etc.

Images