Cross-modal commodity classification method and device, equipment, medium and product

Abstract

The invention discloses a cross-modal commodity classification method and device, equipment, a medium and a product, and the method comprises the steps: obtaining an image-text splicing vector of a commodity object, and enabling the image-text splicing vector to be formed by splicing a picture coding vector of a commodity picture and a text coding vector of a commodity title; performing multi-level coding on the image-text splicing vector based on a multi-head attention mechanism, realizing first feature interaction of the image coding vector and the text coding vector, and obtaining a first image-text fusion vector; performing second feature interaction on the first image-text fusion vector and a label coding vector representing a commodity label of the commodity object in a preset category tree label structure based on a multi-head attention mechanism to obtain a second image-text fusion vector; and performing classification according to the second image-text fusion vector, and judging a classification label of the commodity object according to a classification result. The image-text fusion vector is obtained through deep interaction of the features of the multiple modes, classification is carried out according to the image-text fusion vector, and the classification accuracy can be improved.

Description

technical field [0001] The present application relates to the field of e-commerce information technology, and in particular to a cross-modal commodity classification method and its corresponding device, computer equipment, computer-readable storage medium, and computer program product. Background technique [0002] With the sudden emergence of the cross-border e-commerce model, more and more overseas merchants have joined the e-commerce industry. Due to the different sources of goods, how to better integrate product information to provide users with a better shopping experience poses a big challenge. . One of the basic technologies to improve the user's shopping experience is to correctly classify commodity objects and implement various downstream services based on the classification. [0003] Most of the early product classification algorithms were based on the feature extraction of product titles. However, the features that can be extracted from product titles are relativ...

AI Technical Summary

Problems solved by technology

[0003] Most of the early product classification algorithms were based on the feature extraction of product titles. However, the features that can be extracted from product titles are relatively simple. With the increase of product categories, the accuracy of classification will decrease.

With the development of deep learning, algorithms that combine text and image features to classify products have emerged. However, there are information barriers between the features of different modalities, and simple feature splicing cannot integrate multi-modal information well.

[0004] In addition, with the gradual enrichment of product categories, the product labeling system is becoming more and more complex, and the product labeling system can also assist product classification to a certain extent. However, the traditional algorithm fails to integrate the information of these three modes Well put together, so need to explore separately

Images