CN-114782142-B - Commodity information matching method and device, equipment, medium and product thereof

CN114782142BCN 114782142 BCN114782142 BCN 114782142BCN-114782142-B

Abstract

The application relates to a commodity information matching method, a device, equipment, a medium and a product thereof, wherein the method comprises the steps of obtaining commodity information of a commodity entity to be detected, wherein the commodity information comprises a plurality of description texts, and extracting knowledge subgraphs of the description texts; the method comprises the steps of searching a knowledge sub-graph from a knowledge graph to form a candidate commodity set by a plurality of commodity entities matched with the knowledge sub-graph of the commodity entity to be detected, obtaining a sample set corresponding to each commodity entity in the candidate commodity set, carrying out semantic matching on commodity information of the commodity entity to be detected and the sample set of the commodity entity in the candidate commodity set one by one, and determining the commodity entity matched with the commodity entity to be detected. The technical scheme of the application is suitable for commodity infringement detection scenes, and when whether the commodity entity to be detected forms infringement to some commodity entities needs to be judged, the infringement commodity can be accurately identified by further checking on the basis of checking, and not only can the omission be avoided through the action exerted by the attribute overlap ratio.

Inventors

WU ZHIDONG

Assignees

商线科技(深圳)有限公司
广州欢聚时代信息科技有限公司

Dates

Publication Date: 20260421
Application Date: 20220509
Priority Date: 20220509

Claims (10)

1. The commodity information matching method is characterized by comprising the following steps: responding to a commodity release request of an online store, acquiring commodity information of a commodity entity to be detected, extracting a knowledge subgraph based on the commodity information, wherein the commodity information is commodity description information corresponding to a commodity released to an electronic commerce platform and is used for defining a corresponding commodity entity from a data layer and comprises a plurality of description texts, the description texts comprise any multiple items of title texts, detail texts and commodity attribute data, and the knowledge subgraph comprises mapping relation data between attributes and attribute values extracted from the description texts; Retrieving a knowledge subgraph from a knowledge graph, wherein the knowledge subgraph is matched with a plurality of commodity entities of the commodity entity to be detected to form a candidate commodity set, and the knowledge graph stores the knowledge subgraph corresponding to the plurality of commodity entities in a commodity database; Acquiring a sample set corresponding to each commodity entity in a candidate commodity set, wherein each attribute in a knowledge subgraph corresponding to the sample set comprises a plurality of description texts, and the description texts in all the sample sets are organized according to a unified sequence, wherein the plurality of description texts in the same sample set respectively express the commodity entity corresponding to the sample set in different forms; and carrying out semantic matching on the commodity information of the commodity entities to be detected and the sample set of the commodity entities in the candidate commodity set one by one, and determining the commodity entities matched with the commodity entities to be detected.
2. The commodity information matching method according to claim 1, wherein the acquiring commodity information of the commodity entity to be detected in response to a commodity distribution request of the online store includes the steps of: Responding to a commodity release request of an online store, and acquiring commodity information of a commodity entity to be detected corresponding to the request; extracting attributes from commodity information of commodity entity to be detected to extract corresponding attribute values from each description text according to different attributes, so as to form mapping relation data between the attributes and the attribute values; and constructing mapping relation data between the attribute and the attribute value of the commodity entity to be detected into a corresponding knowledge subgraph according to a preset knowledge graph structure.
3. The commodity information matching method according to claim 1, wherein before the step of acquiring commodity information of the commodity entity to be detected in response to a commodity distribution request of the online store, comprising the steps of: And constructing a knowledge graph corresponding to the commodity database, wherein the knowledge graph comprises knowledge subgraphs corresponding to all commodity entities in the commodity database, and the knowledge subgraphs comprise mapping relation data between attributes and attribute values extracted from description texts of commodity information of the corresponding commodity entities.
4. The commodity information matching method as claimed in claim 3, wherein constructing a knowledge-graph corresponding to the commodity database comprises the steps of: creating a knowledge graph, and acquiring commodity information of each commodity entity in a commodity database; Extracting the attribute of the commodity information of each commodity entity to extract corresponding attribute values from each description text according to different attributes to form mapping relation data between the attribute and the attribute values; According to the preset structure of the knowledge graph, mapping relation data between the attribute and the attribute value of each commodity entity is constructed into a knowledge subgraph of the corresponding commodity entity; And acquiring different description versions of commodity information of each commodity entity, wherein each description version comprises description texts matched with mapping relation data between each attribute and attribute value of the commodity entity, constructing samples corresponding to each description version according to a unified sequence, constructing a sample set of the corresponding commodity entity by all the samples, and storing the sample set in a knowledge sub-graph of the corresponding commodity entity.
5. The commodity information matching method according to claim 1, wherein a plurality of commodity entities whose knowledge subgraphs match the knowledge subgraphs of the commodity entity to be detected are retrieved from the knowledge graph to form a candidate commodity set, comprising the steps of: Acquiring mapping relation data between attributes and attribute values in a knowledge subgraph of a commodity entity to be detected, and taking the mapping relation data as an attribute set; performing coincidence matching calculation on the attribute set of the commodity entity to be detected and the attribute set corresponding to each commodity entity in the knowledge graph, and determining the attribute coincidence between each commodity entity and the commodity entity to be detected in the knowledge graph; And taking the commodity entity set with the attribute coincidence degree meeting the preset condition in the knowledge graph as a candidate commodity set.
6. The commodity information matching method according to claim 5, wherein the commodity information of the commodity entity to be detected is semantically matched with the sample set of commodity entities in the candidate commodity set one by one, and the commodity entity matched with the commodity entity to be detected is determined therefrom, comprising the steps of: calculating first document similarity between commodity information of commodity entities to be detected and a sample set of commodity entities in a candidate commodity set, taking the corresponding attribute coincidence degree of the commodity entities as the weight of the first document similarity to obtain weighted similarity, and screening commodity entities with the weighted similarity meeting preset conditions to form the first commodity set; Calculating semantic similarity between semantic vectors of commodity information of commodity entities to be detected and semantic vectors of a sample set of commodity entities in the first commodity set, and screening commodity entities with the semantic similarity meeting preset conditions to form a second commodity set; And pushing the commodity entity in the second commodity set to a terminal device for providing commodity information of the commodity entity to be detected as a commodity entity matched with the commodity entity to be detected.
7. The commodity information matching method according to claim 6, wherein the step of calculating a semantic similarity between the semantic vector of the commodity information of the commodity entity to be detected and the semantic vector of the sample set of commodity entities in the first commodity set, comprises the steps of: Adopting a sample set of commodity entities in the knowledge graph as a training sample, and performing iterative training on a preset text feature extraction model to a convergence state; And respectively extracting a sample set of each commodity entity in the first commodity set and semantic vectors of commodity information of the commodity entity to be detected by adopting the text feature extraction model.
8. A commodity information matching apparatus, comprising: The information extraction module is used for responding to a commodity release request of an online store, acquiring commodity information of a commodity entity to be detected, extracting a knowledge subgraph based on the commodity information, wherein the commodity information is commodity description information corresponding to a commodity released to an electronic commerce platform and is used for defining a corresponding commodity entity from a data layer, the information extraction module comprises a plurality of description texts, the description texts comprise any multiple items of title texts, detail texts and commodity attribute data, and the knowledge subgraph comprises mapping relation data between attributes and attribute values extracted from the description texts; the searching and matching module is used for searching out a plurality of commodity entities, of which the knowledge subgraph is matched with the knowledge subgraph of the commodity entity to be detected, from the knowledge graph to form a candidate commodity set, wherein the knowledge graph stores the knowledge subgraph corresponding to the plurality of commodity entities in the commodity database; The extraction and sorting module is arranged to acquire a sample set corresponding to each commodity entity in the candidate commodity set, wherein the sample set corresponds to each attribute in the knowledge subgraph and comprises a plurality of description texts, the description texts in all the sample sets are organized according to a unified sequence, and the plurality of description texts in the same sample set respectively express the commodity entities corresponding to the sample set in different forms; The semantic matching module is used for carrying out semantic matching on the commodity information of the commodity entities to be detected and the sample set of the commodity entities in the candidate commodity set one by one, and determining the commodity entities matched with the commodity entities to be detected.
9. A computer device comprising a central processor and a memory, characterized in that the central processor is arranged to invoke a computer program stored in the memory for performing the steps of the method according to any of claims 1 to 7.
10. A computer-readable storage medium, characterized in that it stores in the form of computer-readable instructions a computer program implemented according to the method of any one of claims 1 to 7, which, when invoked by a computer, performs the steps comprised by the corresponding method.

Description

Commodity information matching method and device, equipment, medium and product thereof Technical Field The present application relates to the field of electronic commerce information technology, and in particular, to a method for matching merchandise information, a corresponding apparatus, a computer device, a computer readable storage medium, and a computer program product. Background With the increasing maturity of internet technology, the electronic commerce industry also rapidly develops, and a large number of merchants select to sell commodities through channels of the electronic commerce. The commodities are various, the quality is uneven, and some merchants can sell infringed commodities. The act of selling infringing goods seriously compromises the legitimate interests of the original brand. Thus, detecting a commodity in which an infringement condition exists from among a large number of commodities is an effective means for maintaining the legitimate interests of brand merchants. Although the e-commerce platform has a plurality of methods for detecting whether the infringed commodity is sold by a merchant, the infringed commodity cannot be completely detected due to various evasion means of the merchant. In the existing infringement detection method, a brand word stock is usually maintained, and fuzzy matching is carried out on commodities by using the brand word stock. However, this method cannot detect a commodity in which a brand word does not exist in the text, resulting in a large number of missed detection situations. For example, some sellers may intentionally remove the brand name of the merchandise, leaving only information such as the description of the merchandise, the model number, etc. to circumvent the infringement merchandise detection of the platform. Meanwhile, a plurality of brand names are also some common phrases, and other non-brand commodity texts can also appear, so that the traditional detection method can cause false detection on the conditions. Disclosure of Invention The present application aims to solve at least one of the above problems and provide a commodity information matching method and corresponding device, computer equipment, computer readable storage medium, computer program product, and the following technical solutions are adopted to adapt to the respective purposes of the present application: in one aspect, the present application provides a commodity information matching method, which includes the following steps: acquiring commodity information of a commodity entity to be detected, wherein the commodity information comprises a plurality of description texts, extracting knowledge subgraphs of the commodity information, and the knowledge subgraphs comprise mapping relation data between attributes and attribute values extracted from the description texts; Retrieving a knowledge subgraph from a knowledge graph, wherein the knowledge subgraph is matched with a plurality of commodity entities of the commodity entity to be detected to form a candidate commodity set, and the knowledge graph stores the knowledge subgraph corresponding to the plurality of commodity entities in a commodity database; Acquiring a sample set corresponding to each commodity entity in a candidate commodity set, wherein each attribute in a knowledge subgraph corresponding to the sample set comprises one or more description texts, and the description texts in all the sample sets are organized according to a unified sequence; and carrying out semantic matching on the commodity information of the commodity entities to be detected and the sample set of the commodity entities in the candidate commodity set one by one, and determining the commodity entities matched with the commodity entities to be detected. In some embodiments, the method for acquiring the commodity information of the commodity entity to be detected includes the following steps: responding to a commodity release request of an online store, and acquiring commodity information of a commodity entity to be detected corresponding to the request, wherein the commodity information comprises any one or more items of commodity titles, commodity details and description texts corresponding to commodity attribute data of the commodity entity; extracting attributes from commodity information of commodity entity to be detected to extract corresponding attribute values from each description text according to different attributes, so as to form mapping relation data between the attributes and the attribute values; and constructing mapping relation data between the attribute and the attribute value of the commodity entity to be detected into a corresponding knowledge subgraph according to a preset knowledge graph structure. In some embodiments, before the step of obtaining the commodity information of the commodity entity to be detected, the method includes the following steps: And constructing a knowledge graph corresponding to the commodity