Granted Patent B2

US 12,511,341 · App. 18/246,000 · Granted Dec 30, 2025

Inventors: Wenli Yu (Beijing, CN); Wei Liu (Beijing, CN); Bo Zhang (Beijing, CN)

Assignee: Beijing Baidu Netcom Science Technology Co., Ltd.

G06F16/958

Abstract

Claims (49)

1 . A method of identifying a webpage, comprising:

acquiring structural data of a target webpage, a first association relationship between the target webpage and a historical webpage, and historical graph data for the historical webpage, wherein the historical graph data is a knowledge graph, the historical graph data comprises node elements representing historical webpages and edge elements representing relationships between the historical webpages based on a second association relationship, the first association relationship represents that the historical webpage jumps to the target webpage within a first predetermined time period, or that the target webpage jumps to the historical webpage within the first predetermined time period;

determining target graph data for the target webpage and the historical webpage based on the structural data of the target webpage, the first association relationship and the historical graph data;

determining a similarity between the target webpage and the historical webpage based on the target graph data; and

determining a category of the target webpage based on a category of the historical webpage and the similarity, wherein the determining target graph data for the target webpage and the historical webpage based on the structural data of the target webpage, the first association relationship and the historical graph data comprises:

associating the structural data of the target webpage with the historical graph data based on the first association relationship, such that the target graph data includes node elements corresponding to the target and historical webpages, and edge elements corresponding to the first association relationship.

2 . The method according to claim 1 , wherein:

the determining a similarity between the target webpage and the historical webpage based on the target graph data comprises: processing the target graph data by using a graph neural network, so as to obtain the similarity between the target webpage and the historical webpage; and

the determining a category of the target webpage based on a category of the historical webpage and the similarity comprises: determining, by using the graph neural network, the category of the target webpage based on the similarity and the category of the historical webpage.

3 . The method according to claim 2 , wherein the graph neural network is obtained by:

obtaining, by using the graph neural network, the category of the historical webpage based on the historical graph data; and

updating a network parameter of the graph neural network based on a reference category and the category of the historical webpage.

4 . The method according to claim 3 , wherein a plurality of historical webpages are acquired; and the historical graph data is obtained by:

acquiring structural data of the plurality of historical webpages and the second association relationship between the plurality of historical webpages; and

determining the historical graph data based on the structural data of the historical webpages and the second association relationship.

5 . The method according to claim 4 , wherein the plurality of historical webpages comprise a first historical webpage and a second historical webpage; the second association relationship represents that the first historical webpage jumps to the second historical webpage within a second predetermined time period, or that the second historical webpage jumps to the first historical webpage within the second predetermined time period.

6 . An electronic device, comprising:

at least one processor; and

a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions, when executed by the at least one processor, cause the at least one processor to:

acquire structural data of a target webpage, a first association relationship between the target webpage and a historical webpage, and historical graph data for the historical webpage, wherein the historical graph data is a knowledge graph, the historical graph data comprises node elements representing historical webpages and edge elements representing relationships between the historical webpages based on a second association relationship, the first association relationship represents that the historical webpage jumps to the target webpage within a first predetermined time period, or that the target webpage jumps to the historical webpage within the first predetermined time period;

determine target graph data for the target webpage and the historical webpage based on the structural data of the target webpage, the first association relationship and the historical graph data;

determine a similarity between the target webpage and the historical webpage based on the target graph data; and

determine a category of the target webpage based on a category of the historical webpage and the similarity, wherein the at least one processor is further configured to:

associate the structural data of the target webpage with the historical graph data based on the first association relationship, such that the target graph data includes node elements corresponding to the target and historical webpages, and edge elements corresponding to the first association relationship.

7 . The electronic device according to claim 6 , wherein:

wherein the at least one processor is further configured to: process the target graph data by using a graph neural network, so as to obtain the similarity between the target webpage and the historical webpage; and

wherein the at least one processor is further configured to: determine, by using the graph neural network, the category of the target webpage based on the similarity and the category of the historical webpage.

8 . The electronic device according to claim 7 , wherein the at least one processor is further configured to obtain the graph neural network by:

obtaining, by using the graph neural network, the category of the historical webpage based on the historical graph data; and

updating a network parameter of the graph neural network based on a reference category and the category of the historical webpage.

9 . The electronic device according to claim 8 , wherein a plurality of historical webpages are acquired; and the at least one processor is further configured to obtain the historical graph data by:

acquiring structural data of the plurality of historical webpages and the second association relationship between the plurality of historical webpages; and

determining the historical graph data based on the structural data of the historical webpages and the second association relationship.

10 . The electronic device according to claim 9 , wherein the plurality of historical webpages comprise a first historical webpage and a second historical webpage; the second association relationship represents that the first historical webpage jumps to the second historical webpage within a second predetermined time period, or that the second historical webpage jumps to the first historical webpage within the second predetermined time period.

11 . A non-transitory computer-readable storage medium having computer instructions therein, wherein the computer instructions are configured to cause a computer to:

acquire structural data of a target webpage, a first association relationship between the target webpage and a historical webpage, and historical graph data for the historical webpage, wherein the historical graph data is a knowledge graph, the historical graph data comprises node elements representing historical webpages and edge elements representing relationships between the historical webpages based on a second association relationship, the first association relationship represents that the historical webpage jumps to the target webpage within a first predetermined time period, or that the target webpage jumps to the historical webpage within the first predetermined time period;

determine target graph data for the target webpage and the historical webpage based on the structural data of the target webpage, the first association relationship and the historical graph data;

determine a similarity between the target webpage and the historical webpage based on the target graph data; and

determine a category of the target webpage based on a category of the historical webpage and the similarity, wherein the computer instructions are further configured to cause the computer to:

associate the structural data of the target webpage with the historical graph data based on the first association relationship, such that the target graph data includes node elements corresponding to the target and historical webpages, and edge elements corresponding to the first association relationship.

12 . The non-transitory computer-readable storage medium according to claim 11 , wherein:

wherein the computer instructions are further configured to cause the computer to: process the target graph data by using a graph neural network, so as to obtain the similarity between the target webpage and the historical webpage; and

wherein the computer instructions are further configured to cause the computer to: determine, by using the graph neural network, the category of the target webpage based on the similarity and the category of the historical webpage.

13 . The non-transitory computer-readable storage medium according to claim 12 , wherein the computer instructions are further configured to cause the computer to obtain the graph neural network by:

obtaining, by using the graph neural network, the category of the historical webpage based on the historical graph data; and

updating a network parameter of the graph neural network based on a reference category and the category of the historical webpage.

14 . The non-transitory computer-readable storage medium according to claim 13 , wherein a plurality of historical webpages are acquired; and the computer instructions are further configured to cause the computer to obtain the historical graph data by:

acquiring structural data of the plurality of historical webpages and the second association relationship between the plurality of historical webpages; and

determining the historical graph data based on the structural data of the historical webpages and the second association relationship.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (35)