CN117033934B - Content generation method and device based on artificial intelligence - Google Patents
Content generation method and device based on artificial intelligence Download PDFInfo
- Publication number
- CN117033934B CN117033934B CN202310964426.8A CN202310964426A CN117033934B CN 117033934 B CN117033934 B CN 117033934B CN 202310964426 A CN202310964426 A CN 202310964426A CN 117033934 B CN117033934 B CN 117033934B
- Authority
- CN
- China
- Prior art keywords
- content
- information
- content generation
- feature
- layer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 13
- 238000011176 pooling Methods 0.000 claims description 12
- 230000011218 segmentation Effects 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Machine Translation (AREA)
Abstract
The application provides a content generation method based on artificial intelligence, which relates to the field of artificial intelligence and comprises the following steps: acquiring first information for indicating information generation; obtaining second information corresponding to the first information according to the first characteristics extracted from the first information, and extracting second characteristics of the second information; inputting the first features and the second features into a pre-trained content generation model, generating first content, and classifying the first content; and extracting a third characteristic according to the category of the first content, and inputting the third characteristic into the content generation model to obtain final content. The method and the device can improve the richness of description of user input and generate more accurate and comprehensive content through feature pre-extraction and multiple content generation.
Description
Technical Field
The application relates to the field of artificial intelligence, in particular to a content generation method based on artificial intelligence. And also relates to a content generation system based on artificial intelligence.
Background
The development of artificial intelligence in the world is now developing and developing, and one of the main purposes of artificial intelligence is content creation and content evolution through artificial intelligence algorithms.
Currently, content generation based on artificial intelligence is mainly performed by inputting descriptions into an artificial intelligence model, understanding the descriptions by the artificial intelligence model, and performing content generation. This results in the need to make descriptive content as accurately as possible when artificial intelligence generation is required, otherwise the generated content will not achieve the most desirable results.
Disclosure of Invention
The application aims to solve the problem of insufficient content generation expectations in the prior art and provides a content generation method based on artificial intelligence. And also relates to a content generation system based on artificial intelligence.
The application provides a content generation method based on artificial intelligence, which comprises the following steps:
acquiring first information for indicating information generation;
Obtaining second information corresponding to the first information according to the first characteristics extracted from the first information, and extracting second characteristics of the second information;
Inputting the first features and the second features into a pre-trained content generation model, generating first content, and classifying the first content;
and extracting a third characteristic according to the category of the first content, and inputting the third characteristic into the content generation model to obtain final content.
Optionally, the first feature and the second feature, or the third feature, when input into the content generation model, make a logical relationship determination for each feature.
Optionally, the content generation model includes:
an input layer, a convolution layer, a ReLU layer, a pooling layer and a full connection layer;
and superposing the input layer, the convolution layer, the ReLU layer, the pooling layer and the full connection layer to obtain a content generation model.
Optionally, the training step of the content generation model includes:
S1, initializing network parameters of the content generation model;
s2, inputting sample data prepared in advance into the content generation model;
s3, obtaining an output value, and comparing the output value with a predetermined target value error;
And S4, the error is returned to the content generation model for parameter updating, the steps S2-S4 are repeated, and when the error meets the preset condition, training is completed.
Optionally, the acquiring of the second information based on the first feature further includes information filtering:
Wherein the said Representing the keyword, the (/ >),/>) And representing keywords in the second information, wherein f represents the association relation. Wherein i e (1, m);
In the above expression, the = indicates that it is true, and when it is not true, it is explained that the keyword in the second information needs to be deleted.
The application also provides a content generation system based on artificial intelligence, which comprises:
the receiving module is used for acquiring first information for indicating information generation;
The extraction module is used for obtaining second information corresponding to the first information according to the first characteristics extracted from the first information and extracting second characteristics of the second information;
The processing module is used for inputting the first characteristics and the second characteristics into a pre-trained content generation model, generating first content and classifying the first content;
and the generation module is used for extracting a third characteristic according to the category of the first content and inputting the third characteristic into the content generation model to obtain the final content.
Optionally, the processing module or the generating module performs the first feature and the second feature, or the third feature performs the logical relationship determination on the respective features when the first feature and the second feature are input into the content generating model.
Optionally, the content generation model includes:
an input layer, a convolution layer, a ReLU layer, a pooling layer and a full connection layer;
and superposing the input layer, the convolution layer, the ReLU layer, the pooling layer and the full connection layer to obtain a content generation model.
Optionally, the training step of the content generation model includes:
S1, initializing the convolutional neural network parameters;
S2, inputting sample data prepared in advance into the convolutional neural network;
s3, obtaining an output value, and comparing the output value with a predetermined target value error;
and S4, the error is returned to the convolutional neural network for parameter updating, the steps S2-S4 are repeated, and when the error meets the preset condition, training is completed.
Optionally, the acquiring of the second information based on the first feature further includes information filtering:
Wherein the said Representing the keyword, the (/ >),/>) And representing keywords in the second information, wherein f represents the association relation. Wherein i e (1, m);
In the above expression, the = indicates that it is true, and when it is not true, it is explained that the keyword in the second information needs to be deleted.
The technical scheme of the application has the beneficial effects that:
The application provides a content generation method based on artificial intelligence, which comprises the following steps: acquiring first information for indicating information generation; obtaining second information corresponding to the first information according to the first characteristics extracted from the first information, and extracting second characteristics of the second information; inputting the first features and the second features into a pre-trained content generation model, generating first content, and classifying the first content; and extracting a third characteristic according to the category of the first content, and inputting the third characteristic into the content generation model to obtain final content. The method and the device can improve the richness of description of user input and generate more accurate and comprehensive content through feature pre-extraction and multiple content generation.
Drawings
FIG. 1 is a schematic diagram of an artificial intelligence based content generation flow in accordance with the present application.
Fig. 2 is a schematic diagram of a content creation model according to the present application.
FIG. 3 is a schematic diagram of training of the content creation model in the present application.
FIG. 4 is a schematic diagram of an artificial intelligence based content generation system in accordance with the present application.
Detailed Description
The present application is further described in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art may better understand the present application and practice it.
The following is a detailed description of the embodiments of the present application, but the present application may be implemented in other ways than those described herein, and those skilled in the art can implement the present application by different technical means under the guidance of the inventive concept, so that the present application is not limited by the specific embodiments described below.
The application provides a content generation method based on artificial intelligence, which comprises the following steps: acquiring first information for indicating information generation; obtaining second information corresponding to the first information according to the first characteristics extracted from the first information, and extracting second characteristics of the second information; inputting the first features and the second features into a pre-trained content generation model, generating first content, and classifying the first content; and extracting a third characteristic according to the category of the first content, and inputting the third characteristic into the content generation model to obtain final content. The method and the device can improve the richness of description of user input and generate more accurate and comprehensive content through feature pre-extraction and multiple content generation.
FIG. 1 is a schematic diagram of an artificial intelligence based content generation flow in accordance with the present application.
Referring to the content generation flow based on artificial intelligence shown in fig. 1, the description is first enhanced in richness, and then two times of content generation are performed.
S101 acquires first information for indicating information generation.
The indication information refers to a prompt description of the generated content, and the description can be text, voice and images. Preferably the description is text, more preferably the text is a keyword.
The construction of the hint information is first required before the content generation is performed, which is determined based on the need for the generated content, or based on the intention of the generated content expression, or based on a deep understanding of the generated content.
The indication information may include a plurality of definition information. Taking a keyword as an example, a plurality of the keywords may be set, each of which is a definition regarding the generated content, thereby determining the generated content.
Regarding the indication information, analysis of the indication information may be performed, the limit information therein may be extracted, and the limit information may be combined to form the first information. In the application, text is taken as indication information, and the text is taken as an example, a part of indication information is firstly required to be obtained, the part of speech of the indication information is divided, the vocabulary is selected based on a preset rule to form a key phrase, and the first information is formed.
The preset rule is that vocabulary is extracted through part-of-speech distinction, for example, nouns, adjectives and verbs are extracted, and other vocabularies are ignored. In the specific execution process, other set rules may be set according to actual needs, which is not described herein.
S102, obtaining second information corresponding to the first information according to the first characteristics extracted from the first information, and extracting second characteristics of the second information;
The first information is a keyword set generated by extracting keywords through the indication information and combining all the extracted keywords.
Further, each keyword is analyzed, and the associated word of each keyword is obtained.
In the application, the acquisition of the related words is carried out by firstly extracting the first characteristics of the keywords and then acquiring the related words according to the first characteristics.
Specifically, the first feature refers to the association relationship between the keyword and other words, which may be determined by a statistical manner, or may be extracted by directly setting a pre-trained neural network.
And acquiring the second information based on the first characteristic.
And acquiring second information, wherein the second information can be obtained by analyzing and judging the keywords and the first characteristics corresponding to the keywords, and in a simple way, each keyword is matched with each keyword in the second information.
The matching process is as follows:
Wherein the said Representing the keyword, the (/ >),/>) And representing keywords in the second information, wherein f represents the association relation. Wherein i is (1, m).
In the above expression, the = indicates that it is true, and when it is not true, it is explained that the keyword in the second information needs to be deleted.
Through the comparison, whether the second information can be adopted or not can be determined, and then the second characteristics of the second information are extracted in the same mode that the first characteristics are extracted from the first information.
S103, inputting the first feature and the second feature into a pre-trained content generation model, generating first content, and classifying the first content.
Based on the first feature and the second feature, an input set is first formed, and the repeated features in the first feature and the second feature are deleted to form the input feature set.
The content generation model is pre-trained, and the convolutional neural network model is adopted to construct the content generation model.
As shown in fig. 2, specifically, the content generation model includes:
an input layer, a convolution layer, a ReLU layer, a pooling layer and a full connection layer;
and superposing the input layer, the convolution layer, the ReLU layer, the pooling layer and the full connection layer to obtain a content generation model.
As shown in fig. 3, the training step of the content generation model includes:
S1, initializing the convolutional neural network parameters;
S2, inputting sample data prepared in advance into the convolutional neural network;
s3, obtaining an output value, and comparing the output value with a predetermined target value error;
and S4, the error is returned to the convolutional neural network for parameter updating, the steps S2-S4 are repeated, and when the error meets the preset condition, training is completed.
Inputting the feature set based on the pre-trained content generation model, and inputting the first content. The first content also needs to be categorized.
The classification refers to dividing the content, and dividing the content into a plurality of parts according to a preset dividing rule to form divided content. The segmentation may be performed manually or by a machine.
The machine performs segmentation, and a segmentation rule is preferably set. In general, the content generated is paragraph-specific, so that only the beginning sentence of each paragraph and the end sentence of the previous paragraph need be processed; or the comparison of the first sentence with the last sentence of the previous paragraph.
Contrast can be calculated by the following expression:
wherein P represents the coincidence degree, C represents the value corresponding to the coincident key words, and Representing the total vocabulary of the comparison, said/>Represents keyword weights, said/>Representing the i-th keyword.
Setting a threshold Y, and dividing when P is smaller than Y.
According to the above mode, the first content is divided, and a plurality of divided contents are formed. The segmented content is classified, and the categories can be set manually.
S104, extracting a third feature according to the category of the first content, and inputting the third feature into the content generation model to obtain final content.
And extracting the characteristics of the segmented content by sequentially extracting the first characteristics from the first information or extracting the second characteristics from the second information to obtain third characteristics.
Further, the third feature is input into the content generation model to obtain final content.
In the application, the prompting hours firstly define the purpose to be generated in detail through feature extraction and generation of second information, and generate the content rich in description details through twice content generation.
The application also provides a content generation system based on artificial intelligence, which comprises: the device comprises a receiving module 301, an extracting module 302, a processing module 303 and a generating module 304.
FIG. 4 is a schematic diagram of an artificial intelligence based content generation system in accordance with the present application.
Please refer to the artificial intelligence based content generation system shown in fig. 4.
The receiving module 301 is configured to obtain first information for indicating information generation.
The indication information refers to a prompt description of the generated content, and the description can be text, voice and images. Preferably the description is text, more preferably the text is a keyword.
The construction of the hint information is first required before the content generation is performed, which is determined based on the need for the generated content, or based on the intention of the generated content expression, or based on a deep understanding of the generated content.
The indication information may include a plurality of definition information. Taking a keyword as an example, a plurality of the keywords may be set, each of which is a definition regarding the generated content, thereby determining the generated content.
Regarding the indication information, analysis of the indication information may be performed, the limit information therein may be extracted, and the limit information may be combined to form the first information. In the application, text is taken as indication information, and the text is taken as an example, a part of indication information is firstly required to be obtained, the part of speech of the indication information is divided, the vocabulary is selected based on a preset rule to form a key phrase, and the first information is formed.
The preset rule is that vocabulary is extracted through part-of-speech distinction, for example, nouns, adjectives and verbs are extracted, and other vocabularies are ignored. In the specific execution process, other set rules may be set according to actual needs, which is not described herein.
An extracting module 302, configured to obtain second information corresponding to the first information according to the first feature extracted from the first information, and extract a second feature of the second information;
The first information is a keyword set generated by extracting keywords through the indication information and combining all the extracted keywords.
Further, each keyword is analyzed, and the associated word of each keyword is obtained.
In the application, the acquisition of the related words is carried out by firstly extracting the first characteristics of the keywords and then acquiring the related words according to the first characteristics.
Specifically, the first feature refers to the association relationship between the keyword and other words, which may be determined by a statistical manner, or may be extracted by directly setting a pre-trained neural network.
And acquiring the second information based on the first characteristic.
And acquiring second information, wherein the second information can be obtained by analyzing and judging the keywords and the first characteristics corresponding to the keywords, and in a simple way, each keyword is matched with each keyword in the second information.
The matching process is as follows:
Wherein the said Representing the keyword, the (/ >),/>) And representing keywords in the second information, wherein f represents the association relation. Wherein i is (1, m).
In the above expression, the = indicates that it is true, and when it is not true, it is explained that the keyword in the second information needs to be deleted.
Through the comparison, whether the second information can be adopted or not can be determined, and then the second characteristics of the second information are extracted in the same mode that the first characteristics are extracted from the first information.
The processing module 303 is configured to input the first feature and the second feature into a content generation model trained in advance, generate first content, and classify the first content.
Based on the first feature and the second feature, an input set is first formed, and the repeated features in the first feature and the second feature are deleted to form the input feature set.
The content generation model is pre-trained, and the convolutional neural network model is adopted to construct the content generation model.
As shown in fig. 2, specifically, the content generation model includes:
an input layer, a convolution layer, a ReLU layer, a pooling layer and a full connection layer;
and superposing the input layer, the convolution layer, the ReLU layer, the pooling layer and the full connection layer to obtain a content generation model.
As shown in fig. 3, the training step of the content generation model includes:
S1, initializing the convolutional neural network parameters;
S2, inputting sample data prepared in advance into the convolutional neural network;
s3, obtaining an output value, and comparing the output value with a predetermined target value error;
and S4, the error is returned to the convolutional neural network for parameter updating, the steps S2-S4 are repeated, and when the error meets the preset condition, training is completed.
Inputting the feature set based on the pre-trained content generation model, and inputting the first content. The first content also needs to be categorized.
The classification refers to dividing the content, and dividing the content into a plurality of parts according to a preset dividing rule to form divided content. The segmentation may be performed manually or by a machine.
The machine performs segmentation, and a segmentation rule is preferably set. In general, the content generated is paragraph-specific, so that only the beginning sentence of each paragraph and the end sentence of the previous paragraph need be processed; or the comparison of the first sentence with the last sentence of the previous paragraph.
Contrast can be calculated by the following expression:
wherein P represents the coincidence degree, C represents the value corresponding to the coincident key words, and Representing the total vocabulary of the comparison, said/>Represents keyword weights, said/>Representing the i-th keyword.
Setting a threshold Y, and dividing when P is smaller than Y.
According to the above mode, the first content is divided, and a plurality of divided contents are formed. The segmented content is classified, and the categories can be set manually.
And the generating module 304 is configured to extract a third feature according to the category of the first content, and input the third feature into the content generating model to obtain final content.
And extracting the characteristics of the segmented content by sequentially extracting the first characteristics from the first information or extracting the second characteristics from the second information to obtain third characteristics.
Further, the third feature is input into the content generation model to obtain final content.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (8)
1. A content generation method based on artificial intelligence, comprising:
acquiring first information for indicating information generation;
Obtaining second information corresponding to the first information according to the first characteristics extracted from the first information, and extracting second characteristics of the second information;
Inputting the first features and the second features into a pre-trained content generation model, generating first content, and classifying the first content;
Extracting a third feature according to the category of the first content, and inputting the third feature into the content generation model to obtain final content;
Wherein the first feature and the second feature, or the third feature, when input into the content generation model, perform logic relationship determination on each feature;
the indication information is indicative description of generated content, the first information is information extracted by keywords through the indication information, the first characteristic is association relation between the keywords and other words, and the second information is obtained through analysis, judgment and acquisition of the keywords and the first characteristics corresponding to the keywords;
The classification refers to dividing the content, and dividing the content into a plurality of parts according to a preset dividing rule to form divided content; the preset segmentation rule determines whether segmentation is performed according to the coincidence degree, and the coincidence degree is calculated:
;
;
wherein P represents the coincidence degree, C represents the value corresponding to the coincident key words, and Representing the total vocabulary of the comparison, said/>Represents keyword weights, said/>Representing the i-th keyword.
2. The artificial intelligence based content generation method of claim 1, wherein the content generation model comprises:
an input layer, a convolution layer, a ReLU layer, a pooling layer and a full connection layer;
and superposing the input layer, the convolution layer, the ReLU layer, the pooling layer and the full connection layer to obtain a content generation model.
3. The content generation method based on artificial intelligence according to any one of claims 1 to 2, wherein the training step of the content generation model includes:
S1, initializing network parameters of the content generation model;
s2, inputting sample data prepared in advance into the content generation model;
s3, obtaining an output value, and comparing the output value with a predetermined target value error;
And S4, the error is returned to the content generation model for parameter updating, the steps S2-S4 are repeated, and when the error meets the preset condition, training is completed.
4. The artificial intelligence based content generating method according to claim 1, wherein the obtaining of the second information based on the first feature further comprises information filtering:
;
Wherein the said Represents keywords, the/>Representing keywords in the second information, wherein f represents an association relationship, and i is E (1, m);
In the above expression, the = indicates that it is true, and when it is not true, it is explained that the keyword in the second information needs to be deleted.
5. An artificial intelligence content generating system based on the content generating method according to any one of claims 1 to 4, comprising:
the receiving module is used for acquiring first information for indicating information generation;
The extraction module is used for obtaining second information corresponding to the first information according to the first characteristics extracted from the first information and extracting second characteristics of the second information;
The processing module is used for inputting the first characteristics and the second characteristics into a pre-trained content generation model, generating first content and classifying the first content;
the generation module is used for extracting a third characteristic according to the category of the first content and inputting the third characteristic into the content generation model to obtain final content;
The processing module or the generating module performs a first feature and a second feature, or performs logic relation determination on each feature when the third feature is input into the content generating model;
the indication information is indicative description of generated content, the first information is information extracted by keywords through the indication information, the first characteristic is association relation between the keywords and other words, and the second information is obtained through analysis, judgment and acquisition of the keywords and the first characteristics corresponding to the keywords;
The classification refers to dividing the content, and dividing the content into a plurality of parts according to a preset dividing rule to form divided content; the preset segmentation rule determines whether segmentation is performed according to the coincidence degree, and the coincidence degree is calculated:
;
;
wherein P represents the coincidence degree, C represents the value corresponding to the coincident key words, and Representing the total vocabulary of the comparison, said/>Represents keyword weights, said/>Representing the i-th keyword.
6. The artificial intelligence based content generation system of claim 5, wherein the content generation model comprises:
an input layer, a convolution layer, a ReLU layer, a pooling layer and a full connection layer;
and superposing the input layer, the convolution layer, the ReLU layer, the pooling layer and the full connection layer to obtain a content generation model.
7. The artificial intelligence based content generation system of any one of claims 5 to 6, wherein the training step of the content generation model comprises:
S1, initializing the convolutional neural network parameters;
S2, inputting sample data prepared in advance into the convolutional neural network;
s3, obtaining an output value, and comparing the output value with a predetermined target value error;
and S4, the error is returned to the convolutional neural network for parameter updating, the steps S2-S4 are repeated, and when the error meets the preset condition, training is completed.
8. The artificial intelligence based content generating system according to claim 5, wherein the obtaining of the second information based on the first feature further comprises information filtering:
;
Wherein the said Represents keywords, the/>Representing keywords in the second information, wherein f represents an association relationship, and i is E (1, m);
In the above expression, the = indicates that it is true, and when it is not true, it is explained that the keyword in the second information needs to be deleted.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310964426.8A CN117033934B (en) | 2023-08-02 | 2023-08-02 | Content generation method and device based on artificial intelligence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310964426.8A CN117033934B (en) | 2023-08-02 | 2023-08-02 | Content generation method and device based on artificial intelligence |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117033934A CN117033934A (en) | 2023-11-10 |
CN117033934B true CN117033934B (en) | 2024-04-19 |
Family
ID=88621930
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310964426.8A Active CN117033934B (en) | 2023-08-02 | 2023-08-02 | Content generation method and device based on artificial intelligence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117033934B (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110134796A (en) * | 2019-04-19 | 2019-08-16 | 平安科技(深圳)有限公司 | Clinical test search method, device, computer equipment and the storage medium of knowledge based map |
CN111414735A (en) * | 2020-03-11 | 2020-07-14 | 北京明略软件***有限公司 | Text data generation method and device |
CN111414736A (en) * | 2020-03-23 | 2020-07-14 | 腾讯科技(深圳)有限公司 | Story generation model training method, device, equipment and storage medium |
CN111783455A (en) * | 2020-07-13 | 2020-10-16 | 网易(杭州)网络有限公司 | Training method and device of text generation model and text generation method and device |
CN112417846A (en) * | 2020-11-25 | 2021-02-26 | 中译语通科技股份有限公司 | Text automatic generation method and device, electronic equipment and storage medium |
CN113362045A (en) * | 2021-08-10 | 2021-09-07 | 北京明略软件***有限公司 | Conference schedule generation method and device, electronic equipment and readable storage medium |
CN113420553A (en) * | 2021-07-21 | 2021-09-21 | 北京小米移动软件有限公司 | Text generation method and device, storage medium and electronic equipment |
CN116468009A (en) * | 2023-04-11 | 2023-07-21 | 北京百度网讯科技有限公司 | Article generation method, apparatus, electronic device and storage medium |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10642872B2 (en) * | 2016-10-21 | 2020-05-05 | Salesforce.Com, Inc. | System for optimizing content queries |
US11526808B2 (en) * | 2019-05-29 | 2022-12-13 | The Board Of Trustees Of The Leland Stanford Junior University | Machine learning based generation of ontology for structural and functional mapping |
-
2023
- 2023-08-02 CN CN202310964426.8A patent/CN117033934B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110134796A (en) * | 2019-04-19 | 2019-08-16 | 平安科技(深圳)有限公司 | Clinical test search method, device, computer equipment and the storage medium of knowledge based map |
CN111414735A (en) * | 2020-03-11 | 2020-07-14 | 北京明略软件***有限公司 | Text data generation method and device |
CN111414736A (en) * | 2020-03-23 | 2020-07-14 | 腾讯科技(深圳)有限公司 | Story generation model training method, device, equipment and storage medium |
CN111783455A (en) * | 2020-07-13 | 2020-10-16 | 网易(杭州)网络有限公司 | Training method and device of text generation model and text generation method and device |
CN112417846A (en) * | 2020-11-25 | 2021-02-26 | 中译语通科技股份有限公司 | Text automatic generation method and device, electronic equipment and storage medium |
CN113420553A (en) * | 2021-07-21 | 2021-09-21 | 北京小米移动软件有限公司 | Text generation method and device, storage medium and electronic equipment |
CN113362045A (en) * | 2021-08-10 | 2021-09-07 | 北京明略软件***有限公司 | Conference schedule generation method and device, electronic equipment and readable storage medium |
CN116468009A (en) * | 2023-04-11 | 2023-07-21 | 北京百度网讯科技有限公司 | Article generation method, apparatus, electronic device and storage medium |
Non-Patent Citations (3)
Title |
---|
Efficient estimation of word representations in vector space;Mikolov T等;《arXiv:1301.3781v3》;20130930;1-12 * |
PLANET: Dynamic content planning in autoregressive transformers for long-form text generation;Hu Z等;《arXiv:2203.09100v1》;20220331;1-18 * |
The survey: Text generation models in deep learning;Iqbal T等;《Journal of King Saud University-Computer and Information Sciences》;20221231;第34卷(第6期);2515-2528 * |
Also Published As
Publication number | Publication date |
---|---|
CN117033934A (en) | 2023-11-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110096570B (en) | Intention identification method and device applied to intelligent customer service robot | |
CN108304372B (en) | Entity extraction method and device, computer equipment and storage medium | |
CN110347787B (en) | Interview method and device based on AI auxiliary interview scene and terminal equipment | |
US20060136208A1 (en) | Hybrid apparatus for recognizing answer type | |
CN110232923B (en) | Voice control instruction generation method and device and electronic equipment | |
WO2022042297A1 (en) | Text clustering method, apparatus, electronic device, and storage medium | |
CN116306317B (en) | Automatic AIGC modeling system based on artificial intelligence | |
CN113378563B (en) | Case feature extraction method and device based on genetic variation and semi-supervision | |
CN112199501A (en) | Scientific and technological information text classification method | |
KR20090004216A (en) | System and method for classifying named entities from speech recongnition | |
CN112183106B (en) | Semantic understanding method and device based on phoneme association and deep learning | |
JP2010256498A (en) | Conversion model generating apparatus, voice recognition result conversion system, method and program | |
CN114491062B (en) | Short text classification method integrating knowledge graph and topic model | |
CN116881470A (en) | Method and device for generating question-answer pairs | |
CN113076749A (en) | Text recognition method and system | |
CN112632956A (en) | Text matching method, device, terminal and storage medium | |
CN117033934B (en) | Content generation method and device based on artificial intelligence | |
CN116070642A (en) | Text emotion analysis method and related device based on expression embedding | |
CN114117069B (en) | Semantic understanding method and system for intelligent knowledge graph questions and answers | |
CN112071304B (en) | Semantic analysis method and device | |
CN114911922A (en) | Emotion analysis method, emotion analysis device and storage medium | |
CN117892735B (en) | Deep learning-based natural language processing method and system | |
CN114926272B (en) | Behavior overdue prediction method, system, equipment and medium based on end-to-end model | |
CN117235137B (en) | Professional information query method and device based on vector database | |
CN113254658B (en) | Text information processing method, system, medium, and apparatus |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |