Commodity category prediction method and device, equipment and storage medium

Abstract

The invention discloses a commodity category prediction method and device, equipment and a storage medium, and the method comprises the steps: inputting preset commodity information into a pre-training model for processing, and obtaining a commodity category corresponding to the preset commodity information; wherein the pre-training model is an existing trained model used for predicting a commodity category; carrying out distribution statistics on commodity categories corresponding to the preset commodity information to obtain a corresponding category distribution result, and carrying out distribution alignment processing on the preset commodity information according to the category distribution result; and performing model fine adjustment on the pre-training model by using the preset commodity information after distribution alignment to obtain a target prediction model, and predicting a commodity category of the to-be-predicted commodity information by using the target prediction model. According to the method, the pre-training model trained by large-scale corpora is migrated to the target prediction model, a large number of manual labeling processes are avoided, and the commodity category prediction efficiency and accuracy are improved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, and in particular, to a commodity category prediction method, device, equipment and storage medium. Background technique [0002] Big data is everywhere, especially in the rapidly developing e-commerce platform, data maintenance is particularly important. Among them, a large number of commodity categories are the top priority of governance. Since each e-commerce platform has its own unique commodity categories, how to adaptively organize and multi-level categories of all commodities according to different platforms Carding is the only way to improve platform competitiveness and work efficiency. Most of the existing commodity category governance schemes combine "direct manual labeling" and "fine-tuning model". However, commodity category labeling is relatively complex, manual labeling costs are high and efficiency is low, and the accuracy rate cannot be guaranteed. [0003] There...

AI Technical Summary

Problems solved by technology

Most of the existing product category governance schemes combine "direct manual labeling" and "fine-tuning model". However, product category labeling is more complicated, manual labeling is costly and inefficient, and the accuracy rate cannot be guaranteed

Images