Method and apparatus for constructing object relationship network, and electronic device

Patent metadata

Field	Value
Publication number	US-12417345-B2
Application number	US-202217939271-A
Country	US
Kind code	B2
Filing date	Sep 7, 2022
Priority date	Jan 17, 2022
Publication date	Sep 16, 2025
Grant date	Sep 16, 2025

Abstract

A method and an apparatus for constructing an object relationship network and an electronic device are provided by the present disclosure, relating to the field of artificial intelligence technologies, such as deep neural networks, deep learning, etc. A specific implementation solution is: extracting keywords in respective text contents corresponding to a plurality of objects to obtain keywords corresponding to respective objects; and according to the keywords corresponding to the objects, a similarity between the plurality of objects is determined; and then according to the similarity between the plurality of objects, an object relationship network between the plurality of objects is constructed. Since the object relationship network constructed by means of the similarity between the plurality of objects can accurately describe a closeness degree of a relationship between the objects, thus, the plurality of objects can be managed effectively by means of the constructed object relationship network.

First claim

What is claimed is:

1. A method for constructing an object relationship network, applied to an electronic device and comprising:
extracting keywords in respective text contents corresponding to a plurality of objects to obtain keywords corresponding to respective objects;
determining a similarity between the plurality of objects according to the keywords corresponding to the respective objects; and
constructing the object relationship network between the plurality of objects according to the similarity between the plurality of objects;
wherein the constructing the object relationship network between the plurality of objects according to the similarity between the plurality of objects comprises:
determining target objects from the plurality of objects according to the similarity between the plurality of objects, wherein a similarity corresponding to the target objects is greater than a pre-set threshold value;
determining degrees of centrality corresponding to the target objects, wherein the degrees of centrality are used for indicating positions of the target objects in the object relationship network to be generated; and
constructing the object relationship network according to the degrees of centrality corresponding to the target objects;
wherein the constructing the object relationship network according to the degrees of centrality corresponding to the target objects comprises:
clustering the target objects to obtain a plurality of clustering results, wherein the target objects in different clustering results have different node identifiers in the object relationship network to be generated; and
constructing the object relationship network according to the degrees of centrality and the node identifiers corresponding to the target objects.

2. The method according to claim 1 , wherein the extracting the keywords in the respective text contents corresponding to the plurality of objects to obtain the keywords corresponding to the respective objects comprises:
for each object of the plurality of objects, inputting a text content corresponding to the each object into a keyword extraction model, and obtaining respective vector representations corresponding to a plurality of word combinations by means of a word segmentation model in the keyword extraction model; and
inputting the respective vector representations corresponding to the plurality of word combinations into a classification model in the keyword extraction model to obtain the keywords corresponding to the respective objects.

3. The method according to claim 2 , wherein the obtaining the respective vector representations corresponding to the plurality of word combinations by means of the word segmentation model in the keyword extraction model comprises:
extracting a plurality of word segmentations in the text content by means of the word segmentation model;
determining, according to word embedding vectors and part-of-speech vectors corresponding to the word segmentations, vector representations corresponding to the word segmentations; and
determining, according to the vector representations corresponding to the word segmentations, the respective vector representations corresponding to the plurality of word combinations, wherein the plurality of word combinations are formed by a plurality of adjacent word segmentations.

4. The method according to claim 3 , wherein the plurality of objects comprise a first object and a second object, and determining a similarity between the first object and the second object according to the keywords corresponding to the respective objects comprises:
determining, according to respective keywords corresponding to the first object and the second object, an intersection keyword and a union keyword corresponding to the first object and the second object;
determining, according to a keyword corresponding to the first object, a first vector representation corresponding to the first object, and determining, according to a keyword corresponding to the second object, a second vector representation corresponding to the second object; and
determining the similarity between the first object and the second object according to the intersection keyword, the union keyword, the first vector representation, and the second vector representation.

5. The method according to claim 2 , wherein the plurality of objects comprise a first object and a second object, and determining a similarity between the first object and the second object according to the keywords corresponding to the respective objects comprises:
determining, according to respective keywords corresponding to the first object and the second object, an intersection keyword and a union keyword corresponding to the first object and the second object;
determining, according to a keyword corresponding to the first object, a first vector representation corresponding to the first object, and determining, according to a keyword corresponding to the second object, a second vector representation corresponding to the second object; and
determining the similarity between the first object and the second object according to the intersection keyword, the union keyword, the first vector representation, and the second vector representation.

6. The method according to claim 5 , wherein the determining the similarity between the first object and the second object according to the intersection keyword, the union keyword, the first vector representation, and the second vector representation comprises:
determining a first similarity between the first object and the second object according to a ratio of a quantity of the intersection keyword to a quantity of the union keyword;
determining a second similarity between the first object and the second object according to the first vector representation and the second vector representation; and
determining the similarity between the first object and the second object according to the first similarity and the second similarity.

7. The method according to claim 1 , wherein the plurality of objects comprise a first object and a second object, and determining a similarity between the first object and the second object according to the keywords corresponding to the respective objects comprises:
determining, according to respective keywords corresponding to the first object and the second object, an intersection keyword and a union keyword corresponding to the first object and the second object;
determining, according to a keyword corresponding to the first object, a first vector representation corresponding to the first object, and determining, according to a keyword corresponding to the second object, a second vector representation corresponding to the second object; and
determining the similarity between the first object and the second object according to the intersection keyword, the union keyword, the first vector representation, and the second vector representation.

8. The method according to claim 7 , wherein the determining the similarity between the first object and the second object according to the intersection keyword, the union keyword, the first vector representation, and the second vector representation comprises:
determining a first similarity between the first object and the second object according to a ratio of a quantity of the intersection keyword to a quantity of the union keyword;
determining a second similarity between the first object and the second object according to the first vector representation and the second vector representation; and
determining the similarity between the first object and the second object according to the first similarity and the second similarity.

9. The method according to claim 1 , wherein the keywords are

Assignees

• Beijing Baidu Netcom Sci & Tech Co Ltd

Inventors

• Wang Peng
• Zhu Hengshu
• Dong Zheng
• Yao Kaichun
• Qin Chuan

Classifications

• G06F40/253

  Grammatical analysis; Style critique · CPC title

• G06F16/3347

  using vector based model · CPC title

• G06F16/355

  Creation or modification of classes or clusters · CPC title

• G06F40/268

  Morphological analysis · CPC title

• G06F40/216

  using statistical methods · CPC title