# US 2026/0065053 A1 · Data processing method and related apparatus

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0065053 A1
- Document ID: US20260065053A1
- Status: Published application
- Kind: A1
- Application number: 19/382,466
- Filed: 2025-11-07
- Published: 2026-03-05
- Earliest claimed: 2023-05-10
- Assignees: Huawei Technologies Co., Ltd.
- Applicants: Huawei Technologies Co., Ltd.
- Inventors: Xiangyang Li; Bo Chen; Lu Hou; Ruiming Tang; Jun Yao
- Web page: https://patentgrep.com/patent/US20260065053A1

## CPC

- G06F 16/9535: Search customisation based on user profiles and personalisation
- G06N 3/08: Learning methods
- G06F 16/3344: {using natural language analysis}
- G06F 40/00: Handling natural language data (speech analysis or synthesis, speech recognition G10L)
- G06F 40/20: Natural language analysis (semantic analysis of natural language G06F40/30)
- G06F 40/216: using statistical methods
- G06F 40/279: Recognition of textual entities
- G06F 40/284: Lexical analysis, e.g. tokenisation or collocates
- G06F 40/295: Named entity recognition
- G06F 40/30: Semantic analysis
- G06F 40/56: Natural language generation
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/045: Combinations of networks
- G06N 3/0455: Auto-encoder networks; Encoder-decoder networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/0475: Generative networks
- G06N 3/09: Supervised learning
- G06Q 30/0631: {Recommending goods or services}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

A data processing method is provided. The method includes: obtaining first data, where the first data includes attribute information of a user or attribute information of an item in a plurality of dimensions; generating second data based on the first data, where the second data is a natural language text used to describe the attribute information; obtaining a first feature representation based on the first data through an embedding network, where the first feature representation includes a plurality of embedding vectors, and each embedding vector corresponds to one dimension; obtaining a second feature representation based on the second data through a natural language processing model, where the second feature representation includes a semantic feature of the second data; and updating the embedding network based on a difference between the second feature representation and the first feature representation, to obtain an updated embedding network.

## Claims

### Claim 1

A data processing method, wherein the method comprises:
obtaining first data, wherein the first data comprises attribute information of a user or attribute information of an item in a plurality of dimensions;
generating second data based on the first data, wherein the second data is a natural language text used to describe the attribute information;
obtaining a first feature representation based on the first data through an embedding network, wherein the first feature representation comprises a plurality of embedding vectors, and each embedding vector corresponds to one dimension;
obtaining a second feature representation based on the second data through a natural language processing model, wherein the second feature representation comprises a semantic feature of the second data; and
updating the embedding network based on a difference between the second feature representation and the first feature representation, to obtain an updated embedding network.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the second data comprises a plurality of character units, and the natural language processing model refers to each character unit and a context of each character unit in the second data when processing each character unit.

### Claim 3 (depends on 1)

The method according to claim 1, wherein the second data comprises:
a character of the dimension, a character of the attribute information, and a character indicating an association between the dimension and the attribute information.

### Claim 4 (depends on 3)

The method according to claim 3, wherein the second data further comprises a character indicating separation between attribute information in different dimensions.

### Claim 5 (depends on 3)

The method according to claim 3, wherein the first data comprises the attribute information of the user and the attribute information of the item in the plurality of dimensions, and the second data further comprises a character indicating separation between the attribute information of the user and the attribute information of the item.

### Claim 6 (depends on 1)

The method according to claim 1, wherein the first data is tabular data.

### Claim 7 (depends on 1)

The method according to claim 1, wherein the method further comprises:
mapping the first feature representation and the second feature representation to obtain a mapped first feature representation and a mapped second feature representation in a same dimension; and
updating the embedding network based on the difference between the mapped second feature representation and the mapped first feature representation, to obtain the updated embedding network comprises:
updating the embedding network based on a difference between the mapped second feature representation and the mapped first feature representation, to obtain the updated embedding network.

### Claim 8 (depends on 1)

The method according to claim 1, wherein the method further comprises:
updating the natural language processing model based on the difference between the second feature representation and the first feature representation, to obtain an updated natural language processing model.

### Claim 9

A data processing method, wherein the method comprises:
obtaining attribute information of a second user, attribute information of a second item, and an operation log of the second user on the second item;
obtaining, based on the attribute information of the second user and the attribute information of the second item through a recommendation model, prediction operation information of the second user on the second item, wherein the recommendation model comprises an updated embedding network; and
fine-tuning the recommendation model based on the operation log and the prediction operation information;
wherein the updated embedding network is obtained by a method, wherein the method comprises:
obtaining first data, wherein the first data comprises attribute information of a user or attribute information of an item in a plurality of dimensions;
generating second data based on the first data, wherein the second data is a natural language text used to describe the attribute information;
obtaining a first feature representation based on the first data through an embedding network, wherein the first feature representation comprises a plurality of embedding vectors, and each embedding vector corresponds to one dimension;
obtaining a second feature representation based on the second data through a natural language processing model, wherein the second feature representation comprises a semantic feature of the second data; and
updating the embedding network based on a difference between the second feature representation and the first feature representation, to obtain the updated embedding network.

### Claim 10 (depends on 9)

The method according to claim 9, wherein the second data comprises a plurality of character units, and the natural language processing model refers to each character unit and a context of each character unit in the second data when processing each character unit.

### Claim 11 (depends on 9)

The method according to claim 9, wherein the second data comprises:
a character of the dimension, a character of the attribute information, and a character indicating an association between the dimension and the attribute information.

### Claim 12 (depends on 11)

The method according to claim 11, wherein the second data further comprises a character indicating separation between attribute information in different dimensions.

### Claim 13

A computing device, wherein the computing device comprises a memory and a processor, the memory stores code, and the processor is enabled to: obtain the code:
obtain first data, wherein the first data comprises attribute information of a user or attribute information of an item in a plurality of dimensions;
generate second data based on the first data, wherein the second data is a natural language text used to describe the attribute information;
obtain a first feature representation based on the first data through an embedding network, wherein the first feature representation comprises a plurality of embedding vectors, and each embedding vector corresponds to one dimension;
obtain a second feature representation based on the second data through a natural language processing model, wherein the second feature representation comprises a semantic feature of the second data; and
update the embedding network based on a difference between the second feature representation and the first feature representation, to obtain an updated embedding network.

### Claim 14 (depends on 13)

The computing device according to claim 13, wherein the second data comprises a plurality of character units, and the natural language processing model refers to each character unit and a context of each character unit in the second data when processing each character unit.

### Claim 15 (depends on 13)

The computing device according to claim 13, wherein the second data comprises:
a character of the dimension, a character of the attribute information, and a character indicating an association between the dimension and the attribute information.

### Claim 16 (depends on 15)

The computing device according to claim 15, wherein the second data further comprises a character indicating separation between attribute information in different dimensions.

### Claim 17 (depends on 15)

The computing device according to claim 15, wherein the first data comprises the attribute information of the user and the attribute information of the item in the plurality of dimensions, and the second data further comprises a character indicating separation between the attribute information of the user and the attribute information of the item.

### Claim 18 (depends on 13)

The computing device according to claim 13, wherein the first data is tabular data.

### Claim 19 (depends on 13)

The computing device according to claim 13, wherein the processor is further enabled to:
map the first feature representation and the second feature representation to obtain a mapped first feature representation and a mapped second feature representation in a same dimension; and
update the embedding network based on a difference between the mapped second feature representation and the mapped first feature representation, to obtain the updated embedding network.

### Claim 20 (depends on 13)

The computing device according to claim 13, wherein the processor is further enabled to:
update the natural language processing model based on the difference between the second feature representation and the first feature representation, to obtain an updated natural language processing model.
