# US 2023/0267142 A1 · Multimedia content publishing method and apparatus, electronic device, and storage medium

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

- Number: US 2023/0267142 A1
- Document ID: US20230267142A1
- Status: Published application
- Kind: A1
- Application number: 18/024,016
- Filed: 2021-08-18
- Published: 2023-08-24
- Earliest claimed: 2020-08-28
- Assignees: Beijing Bytedance Network Technology Co., Ltd.
- Applicants: Beijing Bytedance Network Technology Co., Ltd.
- Inventors: Kai SU; Changhu WANG
- Web page: https://patentgrep.com/patent/US20230267142A1

## CPC

- G06F 16/44: Browsing; Visualisation therefor
- G06F 16/43: Querying
- G06F 16/45: Clustering; Classification
- G06F 16/583: using metadata automatically derived from the content
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/08: Learning methods
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A multimedia content publishing method and apparatus, an electronic device and a storage medium. The method includes: determining a multimedia content to be published (S101); obtaining a plurality of candidate topics matching the multimedia content, the candidate topics being topics whose relevancy to the multimedia content meets a predetermined condition among topics included in a topic set (S102); determining a selected target topic among the plurality of candidate topics (S103); and sending multimedia content publishing information containing the target topic to a server in response to a multimedia content publishing request. According to the method, a topic having relatively high relevance to multimedia content can be automatically provided for a user, thus saving time for users to think about and edit topics and improving the accuracy of topic selection.

## Claims

### Claim 1

A multimedia content publishing method, comprising:
determining a multimedia content to be published;
obtaining a plurality of candidate topics matching the multimedia content, the candidate topics being topics whose relevancy to the multimedia content meets a predetermined condition among topics included in a topic set;
determining a selected target topic among the plurality of candidate topics; and
sending multimedia content publishing information containing the target topic to a server in response to a multimedia content publishing request.

### Claim 2 (depends on 1)

The method according to claim 1, wherein after obtaining the plurality of candidate topics matching the multimedia content and before determining the selected target topic among the plurality of candidate topics, the method further comprises:
displaying, after a target identifier input into a topic input box is detected, the plurality of obtained candidate topics matching the multimedia content in a topic selection box corresponding to the topic input box.

### Claim 3 (depends on 2)

The method according to claim 2, wherein the determining a selected target topic among the plurality of candidate topics comprises:
displaying the target topic selected from the plurality of candidate topic in the topic input box corresponding to the topic selection box in response to a selection operation in the topic selection box.

### Claim 4 (depends on 2)

The method according to claim 2, further comprising:
displaying the plurality of obtained candidate topics matching the multimedia content in a vertical scrolling manner in response to a first trigger instruction for a vertical-row display mode; or
displaying the plurality of obtained candidate topics matching the multimedia content in a horizontal scrolling manner in response to a second trigger instruction for a horizontal-row display mode.

### Claim 5

A multimedia content publishing method, comprising:
obtaining a multimedia content to be published;
selecting a plurality of candidate topics matching the multimedia content from a prestored topic set, and returning the plurality of selected candidate topics to a client;
receiving multimedia content publishing information containing a target topic, the target topic being a topic included in the plurality of candidate topics; and
publishing the multimedia content based on the multimedia content publishing information.

### Claim 6 (depends on 5)

The method according to claim 5, wherein the selecting a plurality of candidate topics matching the multimedia content from a prestored topic set comprises:
determining relevancy between the multimedia content and each topic in the stored topic set by utilizing a trained relevancy model; and
selecting the plurality of candidate topics from the topic set based on the relevancy.

### Claim 7 (depends on 6)

The method according to claim 6, wherein the step of training the relevancy model comprises:
obtaining a plurality of historical multimedia content and user-edited topics corresponding to each historical multimedia content;
taking, for each historical multimedia content, the user-edited topic corresponding to the historical multimedia content as a positive class topic corresponding to the historical multimedia content, and taking the user-edited topics corresponding to historical multimedia content other than the historical multimedia content as a negative class topic corresponding to the historical multimedia content; and
taking each historical multimedia content, the positive class topic corresponding to the historical multimedia content, and the negative class topic corresponding to the historical multimedia content as a training data group, and training the relevancy model to be trained based on the plurality of training data groups so as to obtain model parameters of the relevancy model.

### Claim 8 (depends on 7)

The method according to claim 7, wherein the training the relevancy model to be trained based on the plurality of training data groups so as to obtain model parameters of the relevancy model comprises:
inputting, for each training data group among the plurality of training data groups, the training data group into the relevancy model to be trained, so as to determine first relevancy between the historical multimedia content corresponding to the training data group and the positive class topic corresponding to the historical multimedia content, and second relevancy between the historical multimedia content corresponding to the training data group and each negative class topic corresponding to the historical multimedia content; and
adjusting, if a difference value between the first relevancy and any second relevancy in all the second relevancy is less than a predetermined threshold, the model parameters of the relevancy model, and training the relevancy model to be trained again until the difference values between the first relevancy and each second relevancy are each greater than or equal to the predetermined threshold, and stopping training to obtain model parameters of the trained relevancy model.

### Claim 9 (depends on 7)

The method according to claim 7, wherein the step of determining the negative class topic corresponding to each historical multimedia content comprises:
taking, for each historical multimedia content, the positive class topic corresponding to historical multimedia content other than the historical multimedia content, whose topic similarity with the positive class topic of the historical multimedia content is less than a predetermined threshold, as the negative class topic corresponding to the historical multimedia content.

### Claim 10 (depends on 9)

The method according to claim 9, wherein the topic similarity comprises a word overlapping degree, and the steps of determining the word overlapping degree comprises:
performing, for two topics for which the word overlapping degree is to be calculated, word segmentation processing on each of the two topics to obtain a plurality of topic words corresponding to each topic;
performing intersection processing on the plurality of topic words respectively corresponding to the two topics to obtain a processed first topic word group, and performing union processing on the plurality of topic words respectively corresponding to the two topics to obtain a processed second topic word group; and
determining a proportion of the first topic word group in the second topic word group, and taking the determined proportion as the word overlapping degree between the two topics.

### Claim 11 (depends on 7)

The method according to claim 7, wherein the training the relevancy model to be trained based on the plurality of training data groups so as to obtain model parameters of the relevancy model comprises:
performing at least one round of training on the relevancy model to be trained based on the plurality of training data groups, wherein the step of one round of training comprises:
dividing the plurality of training data groups into a plurality of training data sets randomly before the current round of training of the relevancy model to be trained, and each training data set comprising multiple training data groups; and
traversing each training data set among the plurality of training data sets sequentially, and performing the current round of model training on the relevancy model to be trained based on the traversed training data set until all the training data sets are traversed, so as to obtain the model parameters of the relevancy model trained in the current round.

### Claim 12 (depends on 11)

The method according to claim 11, wherein the performing the current round of model training on the relevancy model to be trained based on the traversed training data set comprises:
inputting, for each training data group in the traversed training data set, the training data group into the relevancy model to be trained, so as to determine the first relevancy between the historical multimedia content corresponding to the training data group and the positive class topic of the historical multimedia content, and the second relevancy between the historical multimedia content corresponding to the training data group and each negative class topic of the historical multimedia content; and
selecting the predetermined sampling number of second target relevancy from all the second relevancy; and
adjusting the model parameters of the current round of relevancy model to be trained based on the first relevancy and the predetermined sampling number of selected second target relevancy.

### Claim 13 (depends on 12)

The method according to claim 12, wherein the selecting the predetermined sampling number of second target relevancy from all the second relevancy comprises:
selecting a first number of second relevancy corresponding to a first predetermined proportion randomly from all the second relevancy based on the first predetermined proportion of the second target relevancy to be randomly selected in the predetermined sampling number of second target relevancy;
ranking all the second relevancy in an order from large to small, and selecting a second number of second relevancy corresponding to a second predetermined proportion from all the ranked second relevancy based on the second predetermined proportion of the second target relevancy to be sequentially selected in the predetermined sampling number of second target relevancy, wherein a sum value of the second predetermined proportion and the first predetermined proportion is 1; and
determining the predetermined sampling number of second target relevancy based on the first number of second relevancy and the second number of second relevancy selected.

### Claim 14

A multimedia content publishing apparatus, comprising:
at least one processor; and
at least one memory communicatively coupled to the at least one processor and storing instructions that upon execution by the at least one processor cause the apparatus to:
determine a multimedia content to be published;
obtain a plurality of candidate topics matching the multimedia content, the candidate topics being topics whose relevancy to the multimedia content meets a predetermined condition among topics included in a topic set;
determine a selected target topic among the plurality of candidate topics; and
send multimedia content publishing information containing the target topic to a server in response to a multimedia content publishing request.

### Claim 15

A multimedia content publishing apparatus, comprising:
at least one processor; and
at least one memory communicatively coupled to the at least one processor and storing instructions that upon execution by the at least one processor cause the apparatus to:
obtain a multimedia content to be published;
select a plurality of candidate topics matching the multimedia content from a prestored topic set, and returning the plurality of selected candidate topics to a client;
receive multimedia content publishing information containing a target topic, the target topic being included in the plurality of candidate topics; and
publish the multimedia content based on the multimedia content publishing information.

### Claim 16

(canceled)

### Claim 17

(canceled)
