CN112733515A - Text generation method and device, electronic equipment and readable storage medium - Google Patents
Text generation method and device, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN112733515A CN112733515A CN202011623657.5A CN202011623657A CN112733515A CN 112733515 A CN112733515 A CN 112733515A CN 202011623657 A CN202011623657 A CN 202011623657A CN 112733515 A CN112733515 A CN 112733515A
- Authority
- CN
- China
- Prior art keywords
- text
- keyword
- keywords
- initial
- judgment
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
- G06F40/186—Templates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
Abstract
The invention provides a text generation method, a text generation device, electronic equipment and a readable storage medium, wherein the method comprises the following steps: acquiring an initial text, wherein the initial text is generated based on a predefined template grammar; analyzing the initial text based on the template grammar to obtain a text structure body related to the structure of the initial text; generating a natural language text for a description target based on knowledge-graph data of the description target and the text structure. According to the invention, through predefining the template grammar, a non-development user can generate a logic meticulous initial text according to the template grammar, so that the accuracy can be effectively improved, meanwhile, the initial text is analyzed based on the template grammar to obtain symbols which can be recognized by a machine, the target text can be automatically and rapidly generated by the machine, and the development efficiency can be effectively improved.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a text generation method and apparatus, an electronic device, and a readable storage medium.
Background
With the development of artificial intelligence technology and the increase of information demand of users in the internet era, how to generate personalized content on a large scale and how to accelerate the generation become popular technical problems. Natural Language Generation (NLG) technology is widely used in many popular applications such as intelligent manuscript writing, intelligent reply, BI report generation, etc.
At present, in text document generation, a generation method based on a complete template is widely applied, and the general flow of the method is as follows:
firstly, a product manager or an operator describes product requirements, mainly the logic and conditions triggered by texts;
then, the research and development personnel translate the natural language into codes executable by the program according to the product document to generate codes;
and finally, the product manager checks the generated result and supplements or modifies the logic to unreasonable places.
However, since the code is generated by the developer, the above process has at least the following problems: research personnel may not be familiar with the product requirements, so that the product logic is not meticulous and the accuracy is not high; and when the production result is not passed through the acceptance check of the production personnel, research and development personnel need to supplement and modify repeatedly, so that the development efficiency is low.
Disclosure of Invention
The invention provides a text generation method, a text generation device, an electronic device and a readable storage medium, which are used for solving the defects of low accuracy and low development efficiency in text generation in the prior art and achieving the aim of effectively improving the text accuracy and the development efficiency.
The invention provides a text generation method, which comprises the following steps:
acquiring an initial text, wherein the initial text is generated based on a predefined template grammar;
analyzing the initial text based on the template grammar to obtain a text structure body corresponding to the structure of the initial text;
generating a natural language text for a description target based on knowledge-graph data of the description target and the text structure.
According to the text generation method of an embodiment of the present invention, before the obtaining the initial text, the method further includes:
acquiring keywords in the template grammar, wherein the keywords are keywords associated with a text structure;
receiving user-defined information input by a user, and generating the initial text based on the user-defined information and the quoted keywords;
the referred keywords are keywords in the template grammar referred when the initial text is generated, and the custom information is information corresponding to the referred keywords input by a user.
According to the text generation method of an embodiment of the present invention, the analyzing the initial text to obtain a text structure corresponding to the structure of the initial text specifically includes:
reading data of the initial text, and acquiring the quoted keywords by matching the template grammar;
and determining an analysis strategy corresponding to the type based on the type of the quoted keyword, and acquiring a text structure body corresponding to the type in the initial text based on the analysis strategy.
According to the text generation method of one embodiment of the present invention, the keywords include at least one of the following keywords:
a line structure key, a paragraph dependency key, a rename key, a sub-condition key, and a highlight key.
According to the text generation method of an embodiment of the present invention, the analyzing the initial text to obtain a text structure corresponding to the structure of the initial text specifically includes:
traversing the data of the initial text to acquire the quoted keywords;
judging whether the traversal is finished or not, if not, judging whether the quoted keyword is the line structure keyword or not, if so, analyzing the line structure to obtain a line structure, otherwise, executing paragraph dependence keyword judgment, and judging whether the quoted keyword is the paragraph dependence keyword or not;
if the result of the paragraph dependency keyword judgment is a paragraph dependency keyword, obtaining a paragraph dependency structure by performing paragraph dependency analysis, otherwise, performing sub-condition keyword judgment, wherein the sub-condition keyword judgment is used for judging whether the cited keyword is the sub-condition keyword;
if the result of the sub-condition keyword judgment is a sub-condition keyword, acquiring a sub-condition structure body by performing sub-condition analysis, otherwise, executing highlight keyword judgment, wherein the highlight keyword judgment is used for judging whether the quoted keyword is the highlight keyword;
if the result of the judgment of the highlighted keyword is the highlighted keyword, a highlighted structural body is obtained by performing highlighted analysis, otherwise, judgment of renamed keywords is executed, the judgment of renamed keywords is used for judging whether the quoted keyword is the renamed keyword, and when the result of the judgment of renamed keywords is the renamed keyword, a renamed structural body is obtained by performing rename analysis.
According to one embodiment of the present invention, the method for generating a text includes generating a natural language text for the description target, where the generating includes:
traversing the line structure body and judging whether the traversal is completed,
if not, acquiring the branch structure, and when judging that the logic judgment structure in the branch structure meets logic conditions, performing text filling by using the knowledge map data to generate a branch text;
and if so, carrying out paragraph dependence processing on the branch text based on the paragraph dependence structure, and updating the whole branch text according to a processing result to obtain the natural language text.
The text generation method according to an embodiment of the present invention further includes:
and displaying an operation interface, wherein the operation interface is used for receiving the operations of a user for writing template texts, modifying logic conditions, verifying the logic conditions and previewing the text effects.
The present invention also provides a text generating apparatus, including:
the acquisition module is used for acquiring an initial text, and the initial text is generated based on a predefined template grammar;
the analysis module is used for analyzing the initial text based on the template grammar to obtain a text structure body corresponding to the structure of the initial text;
and the generation module is used for generating a natural language text aiming at the description target based on the knowledge graph data of the description target and the text structure body.
According to the text generating apparatus of an embodiment of the present invention, the obtaining module is further configured to:
acquiring keywords in the template grammar, wherein the keywords are keywords associated with a text structure;
receiving user-defined information input by a user, and generating the initial text based on the user-defined information and the quoted keywords;
the referred keywords are keywords in the template grammar referred when the initial text is generated, and the custom information is information corresponding to the referred keywords input by a user.
According to an embodiment of the present invention, the parsing module is specifically configured to:
reading data of the initial text, and acquiring the quoted keywords by matching the template grammar;
and determining an analysis strategy corresponding to the type based on the type of the quoted keyword, and acquiring a text structure body corresponding to the type in the initial text based on the analysis strategy.
According to the text generation device of one embodiment of the present invention, the keywords include at least one of the following keywords:
a line structure key, a paragraph dependency key, a rename key, a sub-condition key, and a highlight key.
According to an embodiment of the present invention, the parsing module is specifically configured to:
traversing the data of the initial text to acquire the quoted keywords;
judging whether the traversal is finished or not, if not, judging whether the quoted keyword is the line structure keyword or not, if so, analyzing the line structure to obtain a line structure, otherwise, executing paragraph dependence keyword judgment, and judging whether the quoted keyword is the paragraph dependence keyword or not;
if the result of the paragraph dependency keyword judgment is a paragraph dependency keyword, obtaining a paragraph dependency structure by performing paragraph dependency analysis, otherwise, performing sub-condition keyword judgment, wherein the sub-condition keyword judgment is used for judging whether the cited keyword is the sub-condition keyword;
if the result of the sub-condition keyword judgment is a sub-condition keyword, acquiring a sub-condition structure body by performing sub-condition analysis, otherwise, executing highlight keyword judgment, wherein the highlight keyword judgment is used for judging whether the quoted keyword is the highlight keyword;
if the result of the judgment of the highlighted keyword is the highlighted keyword, a highlighted structural body is obtained by performing highlighted analysis, otherwise, judgment of renamed keywords is executed, the judgment of renamed keywords is used for judging whether the quoted keyword is the renamed keyword, and when the result of the judgment of renamed keywords is the renamed keyword, a renamed structural body is obtained by performing rename analysis.
According to an embodiment of the present invention, the text generating apparatus includes a branch structure, the branch structure includes a logic judgment structure, and the generating module is specifically configured to:
traversing the line structure body and judging whether the traversal is completed,
if not, acquiring the branch structure, and when judging that the logic judgment structure in the branch structure meets logic conditions, performing text filling by using the knowledge map data to generate a branch text;
and if so, carrying out paragraph dependence processing on the branch text based on the paragraph dependence structure, and updating the whole branch text according to a processing result to obtain the natural language text.
The text generation apparatus according to an embodiment of the present invention further includes:
and the display module is used for displaying an operation interface, and the operation interface is used for receiving the operations of a user for writing the template text, modifying the logic conditions, verifying the logic conditions and previewing the text effect.
The invention also provides an electronic device, which comprises a memory, a processor and a program or an instruction stored on the memory and capable of running on the processor, wherein when the processor executes the program or the instruction, the steps of the text generation method are realized.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a program or instructions which, when executed by a computer, implement the steps of the text generation method as described in any of the above.
According to the text generation method and device, the electronic equipment and the readable storage medium, through predefining the template grammar, a non-development user can generate a logic meticulous initial text according to the template grammar, so that the accuracy can be effectively improved, meanwhile, the machine-recognizable symbol is obtained through analyzing the initial text based on the template grammar, a target text can be automatically and quickly generated by using a machine, and the development efficiency can be effectively improved.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a text generation method provided by the present invention;
fig. 2 is a schematic flow chart of parsing an initial text in the text generation method provided by the present invention;
fig. 3 is a schematic flow chart of generating a final text in the text generation method provided by the present invention;
FIG. 4 is a schematic structural diagram of a text generating apparatus provided in the present invention;
fig. 5 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Aiming at the problems of low accuracy and low development efficiency in text generation in the prior art, the invention enables a non-development user to generate a logic meticulous initial text by self according to the template grammar through predefining the template grammar, thereby effectively improving the accuracy, and meanwhile, the target text can be automatically and rapidly generated by a machine by analyzing the initial text based on the template grammar to obtain a symbol which can be identified by the machine, thereby effectively improving the development efficiency. The present invention will now be described and explained with reference to the drawings, in particular, by means of embodiments.
Fig. 1 is a schematic flow diagram of a text generation method provided by the present invention, and as shown in fig. 1, the method includes:
s101, acquiring an initial text.
Wherein the initial text is generated based on a predefined template grammar.
Specifically, the invention can define some template grammars in advance, and the user can call the template grammars, add self-defined information into the template grammars, combine the template grammars and input the template grammars into the system. After receiving the combined information of the user, the system can generate an initial text, or draft text, according to the combined information, wherein the text is a direct expression of the user's requirements. For example, the product staff may describe their own requirements according to a certain template grammar (or grammar rule), and generate the initial text.
And S102, analyzing the initial text based on the template grammar, and acquiring a text structure body corresponding to the structure of the initial text.
Specifically, upon obtaining the initial text, the present invention may utilize a parsing tool to parse the initial text, and therefore, a parser may be included in the parsing tool. Since the initial text contains the template grammar information used by the user, the invention can read the data of the initial text in sequence.
And then, analyzing the data by using an analyzer based on the predefined template grammar to obtain keywords which are matched with the template grammar and correspond to the text structure and user-defined information of users corresponding to the keywords. According to the keywords and the custom information, a text structure body corresponding to the structure of the initial text can be further obtained.
The text structure can be understood as a text structure framework component obtained by splitting a text according to a paragraph structure, and is a data structure which can be recognized by a machine and can be directly run, and the text structure comprises a line structure, a paragraph dependency structure, a condition structure, a highlight structure, a rename structure and the like.
S103, generating a natural language text aiming at the description target based on the knowledge graph data of the description target and the text structure.
It is to be understood that the parsing tool may further comprise a generator based on the obtained text structure. With the generator, a text framework can be built based on the structures, and on the basis of building the framework, the framework is filled with specific data describing the target, and a final natural language text is generated.
The description target is an object to be described in a text, such as a certain cell, a certain specific house source and the like in the house property field. The knowledge-graph data describing the object is related information describing the object obtained from a plurality of knowledge-graphs related to the describing object.
According to the text generation method provided by the invention, through predefining the template grammar, a non-development user can generate a logic meticulous initial text according to the template grammar, so that the accuracy can be effectively improved, meanwhile, through analyzing the initial text based on the template grammar to obtain symbols which can be recognized by a machine, the target text can be automatically and rapidly generated by the machine, and the development efficiency can be effectively improved.
The text generation method provided according to the foregoing embodiments is optional, and before the obtaining the initial text, the method further includes: acquiring keywords in the template grammar, wherein the keywords are keywords associated with a text structure; and receiving user-defined information input by a user, and generating the initial text based on the user-defined information and the referred keywords.
Wherein the referred keyword is a keyword in the template grammar referred when the initial text is generated, and the custom information is information corresponding to the referred keyword and input by a user.
It will be appreciated that the present invention requires the initial text to be generated prior to obtaining the initial text for generation of the natural language text. Specifically, according to a common line structure, for example, first writing a chapter, then writing a paragraph, and finally writing a sentence, a plurality of keywords for expressing the article level are defined, which can be called as keywords in the template grammar.
Optionally, the keywords in the template grammar include at least one of the following keywords: a line structure key, a paragraph dependency key, a rename key, a sub-condition key, and a highlight key.
Specifically, the invention defines a plurality of keywords for expressing article levels, such as CHAP, SECT, PARA, BRANCH and the like according to the common line structure, such as the sequence of writing chapters, then writing paragraphs, then writing segments and finally writing sentences. And according to common logic requirements, a logic key IF is defined, so that the logic of the product when … … is described conveniently.
In addition, the invention also designs template grammars such as highlight keywords, paragraph dependent keywords and the like, thereby ensuring that the requirements of most products are covered. The keywords that can specifically define the template grammar are shown in table 1, which is a keyword list of the template grammar in the text generation method provided by the present invention.
Table 1 shows a keyword list of template grammar in the text generation method provided by the present invention
Based on the above definition, the user can refer to these keywords as needed, and these keywords can be referred to as referred keywords. After referencing the keywords, the content of the referenced keywords may be populated with the user's customization information. Thus, based on the keywords quoted by the user, such as CHAP, search and PARA, and the user-defined information filled by the user corresponding to each keyword, the corresponding initial text can be formed.
Taking a specific product requirement as an example, in a learning environment scene of a broker training court, a whole example text is expected to be given to a broker, and a cell introduction learning process is completed from a starting scene, a peripheral matching scene, a cell basic situation, a finishing word and the like. The information to be introduced in each part is shown in table 2, which is an application example table for generating an initial text in the text generation method provided by the present invention.
Table 2 shows an application example table for generating an initial text in the text generation method provided by the present invention
Based on the above application scenario, the non-developer can compose the initial text including the following:
CHAP open field
SECT Broker introduction
PARA self introduction
BRANCH
IF resblock_name,！＝,AND agent_name,！＝,
Honored customers, your good, i am your exclusive presence advisor agent name, today i am with you know about resblock name.
BRANCH
IF resblock_name,＝,AND agent_name,！＝,
Honored customers, your good, i am your proprietary marketing advisor agent name, which is today served by me.
CHAP peripheral kit
SECT traffic
PARA traffic guidance
RELY subway OR bus
BRANCH
First, we look at the perimeter complement of the cell.
PARA subway
RENAME nearest_subway_name:sub_name
BRANCH
IF sub_name,！＝,AND nearest_subway_dist,>,0AND nearest_subway_walktime,>,0
Walking about { near _ subway _ walk time } minutes away from the subway station { sub _ name } { near _ subway _ dist } kilometers.
PARA public transport
BRANCH
IF bus_station_cnt,>,0AND bus_line_cnt,>,0AND total_bus_routes,！＝
Within 1000 meters of the periphery, there are { bus _ station _ cnt } bus stations: { total _ bus _ routes }. { bus _ line _ cnt } bus lines
FMT total _ bus _ routes, concatejvalue, 3, etc.,
the CHAP, SECT, PARA and the like respectively represent keywords corresponding to the text structure, and the subsequent "opening", "broker introduction", "self introduction" and the like respectively represent user filling information corresponding to the previous keywords, that is, user-defined information of the user.
When the initial text is generated, each chapter can contain a plurality of pieces, each piece can contain a plurality of sections, each section can contain a plurality of sentences, and the number and the sequence can be controlled by a non-developer (such as a producer) by himself/herself when writing the initial text.
According to the method and the device, the keywords corresponding to the text structure are defined, so that the user can generate the required text only by referring the required keywords and filling the custom information, the whole grammar content is simple and easy to understand, and the operability is stronger.
The text generation method provided according to each of the above embodiments is optional, and the analyzing the initial text to obtain a text structure corresponding to the structure of the initial text specifically includes: reading data of the initial text, and acquiring the quoted keywords by matching the template grammar; and determining an analysis strategy corresponding to the type based on the type of the quoted keyword, and acquiring a text structure body corresponding to the type in the initial text based on the analysis strategy.
It will be appreciated that the present invention programs the initial text that satisfies the above-described template grammar into a data structure that is directly readable by subsequent machines. When the initial text is parsed, the data of the initial text needs to be read sentence by sentence, and the data is compared and matched with the keywords in the initial text, so as to obtain the data of the keywords belonging to the template grammar in the data, wherein the data is the reference of the user to the keywords in the template grammar and can be called as the referred keywords.
On this basis, the type to which the keyword belongs can be determined according to the keyword list listed in table 1. For example, if the keyword that the user refers to is CHAP obtained by matching, it belongs to the line structure keyword as shown in table 1.
Meanwhile, the invention presets analysis strategies corresponding to different key word types, for example, a line structure analysis strategy is corresponding to a line structure key word, and a paragraph dependent analysis strategy is corresponding to a paragraph dependent key word. Therefore, after the type of the referred keyword is determined through judgment, the analysis strategy for the referred keyword and the filling information thereof can be obtained through inquiring the corresponding relation between the keyword type and the analysis strategy. And then, analyzing the quoted keywords and the filling information thereof according to the analysis strategy to obtain a corresponding text structure body.
For example, for the keyword of the line structure, a line structure analysis strategy is determined, and the line structure analysis is performed according to the line structure analysis strategy to obtain the line structure.
The text generation method provided according to each of the above embodiments is optional, and the analyzing the initial text to obtain a text structure corresponding to the structure of the initial text specifically includes:
traversing the data of the initial text to acquire the quoted keywords;
judging whether the traversal is finished or not, if not, judging whether the quoted keyword is the line structure keyword or not, if so, analyzing the line structure to obtain a line structure, otherwise, executing paragraph dependence keyword judgment, and judging whether the quoted keyword is the paragraph dependence keyword or not;
if the result of the paragraph dependency keyword judgment is a paragraph dependency keyword, obtaining a paragraph dependency structure by performing paragraph dependency analysis, otherwise, performing sub-condition keyword judgment, wherein the sub-condition keyword judgment is used for judging whether the cited keyword is the sub-condition keyword;
if the result of the sub-condition keyword judgment is a sub-condition keyword, acquiring a sub-condition structure body by performing sub-condition analysis, otherwise, executing highlight keyword judgment, wherein the highlight keyword judgment is used for judging whether the quoted keyword is the highlight keyword;
if the result of the judgment of the highlighted keyword is the highlighted keyword, a highlighted structural body is obtained by performing highlighted analysis, otherwise, judgment of renamed keywords is executed, the judgment of renamed keywords is used for judging whether the quoted keyword is the renamed keyword, and when the result of the judgment of renamed keywords is the renamed keyword, a renamed structural body is obtained by performing rename analysis.
Specifically, as shown in fig. 2, a schematic flow diagram for analyzing an initial text in the text generation method provided by the present invention includes the following processing procedures:
and reading the initial text data item by adopting an overall traversal mode, and acquiring the quoted keywords contained in the initial text by matching the template grammar.
In the traversal process, after reading and analyzing one piece of initial text data, whether traversal is finished or not is judged, that is, whether the data of the initial text is completely read or not is judged. If the traversal is finished, the data is completely analyzed, and the analysis process can be finished; if the traversal is not finished, namely if the data is not read completely, the quoted keywords are obtained, and whether the quoted keywords are the line structure keywords in the template grammar or not is judged according to the types of the quoted keywords.
If the quoted keywords are known to be the line structure keywords according to the judgment, performing line structure analysis on the quoted keywords to obtain a line structure body; otherwise, the paragraph dependent key judgment is performed on the referenced key, that is, whether the referenced key is a paragraph dependent key is further judged.
If the referenced keyword is judged and known to be the paragraph dependent keyword according to the paragraph dependent keyword, performing paragraph dependent analysis on the referenced keyword to obtain a paragraph dependent structure; otherwise, the sub-condition keyword judgment is performed on the referred keyword, that is, whether the referred keyword is a sub-condition keyword is further judged.
If the quoted keyword is judged and known to be the sub-condition keyword according to the sub-condition keyword, carrying out sub-condition analysis on the quoted keyword to obtain a sub-condition structural body; otherwise, the highlighted keyword judgment is performed on the referred keyword, that is, whether the referred keyword is a highlighted keyword is further judged.
If the quoted keyword is judged and known to be the highlight keyword according to the highlight keyword, carrying out highlight analysis on the quoted keyword to obtain a highlight structure body; otherwise, the renamed keyword judgment is performed on the quoted keyword, that is, whether the quoted keyword is a renamed keyword is further judged.
If the fact that the quoted key words are the renamed key words is judged and known according to the renamed key words, renaming analysis is conducted on the quoted key words, and a renamed structure body is obtained; otherwise, turning to the step of reading the initial text data, and circularly executing the judging process until the initial text is read and analyzed.
The invention is responsible for converting the initial text composed of the template grammar into the program code which can be identified by the machine, and finally outputting the line structure, the paragraph dependence structure, the condition structure, the highlight structure, the renaming structure and the like which are all data structures which can be directly operated by the program, so that the final natural language text can be automatically and rapidly generated by operating the program, and the efficiency is higher.
The text generation method provided according to each of the above embodiments is optional, where the literary structure includes a branch structure, the branch structure includes a logic judgment structure, and the generating of the natural language text for the description target specifically includes:
traversing the line structure body and judging whether the traversal is completed,
if not, acquiring the branch structure, and when judging that the logic judgment structure in the branch structure meets logic conditions, performing text filling by using the knowledge map data to generate a branch text;
and if so, carrying out paragraph dependence processing on the branch text based on the paragraph dependence structure, and updating the whole branch text according to a processing result to obtain the natural language text.
Specifically, as shown in fig. 3, for a schematic flow diagram of generating a final text in the text generation method provided by the present invention, on the basis of obtaining an initial text structure according to the above embodiment, the structure may be directly run to generate a final natural language text. The generation process is to combine the result of the parser and the specific data describing the target to generate the final natural language text, and the processing process is as follows:
firstly, obtaining the analysis result of the analyzer, namely the structural body of the initial text, and reading the line structural bodies one by one in a whole traversal mode.
Secondly, in the traversing process, after reading one line structure body, whether traversing is finished or not is judged, namely whether reading of the line structure body is finished or not is judged. And when the traversal is not completed, continuously reading the next branch structure body as the current branch structure body, and judging whether the logic condition in the current branch structure body is met. If the target description object meets the logic conditions, text filling is carried out according to the met logic conditions, namely, the content corresponding to the branch structure body is filled by using the knowledge graph data related to the target description object to generate a branch text, and the generated text can be highlighted according to the highlight structure body.
And thirdly, updating the result of the text generated by the current branch structure, transferring to the next line structure, and circularly executing the processing process from traversing the line structure to highlighting to obtain a plurality of branch texts.
And finally, after the traversal of the line text structure is completed, performing paragraph dependence processing on the obtained multiple branch texts according to different branch structures, updating the result of the whole paragraph dependence processing, outputting the final natural language text result, and ending the text generation flow.
For example, for the example of table 2 above, the final generated text for data with broker name, cell name, subway station name, subway line travel time, number of buses, and bus line is as follows:
honored customers, good, i am your exclusive business advisor shellfish, and today I am with you know to melt the garden. First, we look at the perimeter complement of the cell. The subway station melting station is 1.5 kilometers away, and the walking time is about 3 minutes. Within 1000 meters of periphery, there are 2 bus stations, 10 bus lines: 671 routes, 38 routes for fast direct special line, 921 routes, etc.
If there is no subway or bus around the cell, the final generated text is as follows:
honored customers, good, i am your exclusive business advisor shellfish, and today I am with you know to melt the garden.
It can be seen that the present invention triggers different logics according to different cell conditions, and finally generates different texts.
According to the method, paragraph dependence processing is performed on the basis of traversing the line structure, so that the condition that contents of different levels of the text are inconsistent can be effectively avoided, and the accuracy is higher.
Further, on the basis of the text generation method provided by each of the above embodiments, the method further includes: and displaying an operation interface, wherein the operation interface is used for receiving the operations of a user for writing template texts, modifying logic conditions, verifying the logic conditions and previewing the text effects.
Specifically, the invention facilitates the operation, encapsulates the whole text generation process into a service form for external provision, and provides and displays an operation interface at the front end, so that a user can directly write template texts, modify logics, verify logics, finally generate effect display of texts and the like through the operation interface. For example, the service may be a Tornado service with Django as a framework, that is, an operation interface is provided and exposed to the user in the form of the Tornado service.
By providing and displaying the operation interface, the invention enables non-developers to conveniently carry out the related operation of the text generation process through the operation interface, for example, products or operators can write logic and texts themselves and know the text generation effect in real time, thus not only improving the development efficiency of research and development personnel, but also helping the product logic to be more meticulous, and further improving the quality and efficiency of the whole natural language processing process.
Based on the same inventive concept, the present invention provides a text generation apparatus according to the above embodiments, which is used for implementing text generation in the above embodiments. Therefore, the description and definition in the text generation method of each embodiment described above can be used for understanding each execution module in the text generation device of the present invention, and reference may be specifically made to each method embodiment described above, which is not described herein again.
According to an embodiment of the present invention, a structure of a text generating apparatus is as shown in fig. 4, which is a schematic structural diagram of a text generating apparatus provided by the present invention, and the apparatus may be used to implement generation of a text in the foregoing method embodiments, and the apparatus includes: an acquisition module 401, an analysis module 402 and a generation module 403.
The obtaining module 401 is configured to obtain an initial text, where the initial text is generated based on a predefined template grammar; the parsing module 402 is configured to parse the initial text based on the template grammar to obtain a text structure corresponding to the structure of the initial text; the generating module 403 is configured to generate a natural language text for a description target based on the knowledge-graph data of the description target and the text structure.
Specifically, the present invention may define some template grammars in advance, and the user may call the template grammar, add self-defined information, combine the template grammar and input the combined information to the system acquisition module 401. After receiving the combined information of the user, the obtaining module 401 may generate an initial text, or referred to as a draft text, according to the combined information, where the text is a direct expression of the user's requirements. For example, the product staff may describe their own requirements according to a certain template grammar (or grammar rule), and generate the initial text.
Thereafter, upon obtaining the initial text, the parsing module 402 may parse the initial text using a parsing tool, and thus a parser may be included in the parsing tool. Since the initial text includes the template grammar information used by the user, the parsing module 402 reads the data of the initial text in sequence, and parses the data by using a parser based on the predefined template grammar to obtain keywords which are matched with the template grammar and are related to the text structure, and the user-defined information of the user corresponding to the keywords. Based on these keywords and the custom information, the parsing module 402 can further obtain a text structure related to the structure of the initial text.
Finally, on the basis of obtaining the text structure, the generating module 403 may build a text framework based on the structure, and on the basis of building the framework, fill the framework with specific data describing the target, and generate a final natural language text.
The text generation device provided by the invention can enable a non-development user to generate a logic meticulous initial text according to the template grammar by predefining the template grammar, thereby effectively improving the accuracy, and meanwhile, a machine can be used for automatically and rapidly generating a target text by analyzing the initial text based on the template grammar to obtain a symbol which can be identified by the machine, thereby effectively improving the development efficiency.
Optionally, the text generating apparatus provided according to the foregoing embodiments, the obtaining module is further configured to:
acquiring keywords in the template grammar, wherein the keywords are keywords associated with a text structure;
receiving user-defined information input by a user, and generating the initial text based on the user-defined information and the quoted keywords;
the referred keywords are keywords in the template grammar referred when the initial text is generated, and the custom information is information corresponding to the referred keywords input by a user.
The text generation device provided according to each of the above embodiments is optional, and the parsing module is specifically configured to:
reading data of the initial text, and acquiring the quoted keywords by matching the template grammar;
and determining an analysis strategy corresponding to the type based on the type of the quoted keyword, and acquiring a text structure body corresponding to the type in the initial text based on the analysis strategy.
The text generation device provided according to the foregoing embodiments is optional, and the keyword includes at least one of the following keywords: a line structure key, a paragraph dependency key, a rename key, a sub-condition key, and a highlight key.
The text generation device provided according to each of the above embodiments is optional, and the parsing module is specifically configured to:
traversing the data of the initial text to acquire the quoted keywords;
judging whether the traversal is finished or not, if not, judging whether the quoted keyword is the line structure keyword or not, if so, analyzing the line structure to obtain a line structure, otherwise, executing paragraph dependence keyword judgment, and judging whether the quoted keyword is the paragraph dependence keyword or not;
if the result of the paragraph dependency keyword judgment is a paragraph dependency keyword, obtaining a paragraph dependency structure by performing paragraph dependency analysis, otherwise, performing sub-condition keyword judgment, wherein the sub-condition keyword judgment is used for judging whether the cited keyword is the sub-condition keyword;
if the result of the sub-condition keyword judgment is a sub-condition keyword, acquiring a sub-condition structure body by performing sub-condition analysis, otherwise, executing highlight keyword judgment, wherein the highlight keyword judgment is used for judging whether the quoted keyword is the highlight keyword;
if the result of the judgment of the highlighted keyword is the highlighted keyword, a highlighted structural body is obtained by performing highlighted analysis, otherwise, judgment of renamed keywords is executed, the judgment of renamed keywords is used for judging whether the quoted keyword is the renamed keyword, and when the result of the judgment of renamed keywords is the renamed keyword, a renamed structural body is obtained by performing rename analysis.
The text generating device provided according to each of the above embodiments is optional, where the literary structure includes a branch structure, the branch structure includes a logic judgment structure, and the generating module is specifically configured to:
traversing the line structure body and judging whether the traversal is completed,
if not, acquiring the branch structure, and when judging that the logic judgment structure in the branch structure meets logic conditions, performing text filling by using the knowledge map data to generate a branch text;
and if so, carrying out paragraph dependence processing on the branch text based on the paragraph dependence structure, and updating the whole branch text according to a processing result to obtain the natural language text.
Further, on the basis of the text generation apparatus provided in the foregoing embodiments, the method further includes: and the display module is used for displaying an operation interface, and the operation interface is used for receiving the operations of a user for writing the template text, modifying the logic conditions, verifying the logic conditions and previewing the text effect.
It is understood that the relevant program modules in the devices of the above embodiments can be implemented by a hardware processor (hardware processor) in the present invention. Moreover, the text generation apparatus of the present invention can implement the text generation process of each method embodiment by using each program module, and when the apparatus of the present invention is used to implement the text generation of each method embodiment, the beneficial effects produced by the apparatus of the present invention are the same as those of each corresponding method embodiment, and reference may be made to each method embodiment, which is not described herein again.
As a further aspect of the present invention, the present embodiment provides an electronic device according to the above embodiments, where the electronic device includes a memory, a processor, and a program or an instruction stored in the memory and executable on the processor, and the processor executes the program or the instruction to implement the steps of the text generation method according to the above embodiments.
Further, the electronic device of the present invention may further include a communication interface and a bus. Referring to fig. 5, a schematic structural diagram of an electronic device provided in the present invention includes: at least one memory 501, at least one processor 502, a communication interface 503, and a bus 504.
The memory 501, the processor 502 and the communication interface 503 complete mutual communication through the bus 504, and the communication interface 503 is used for information transmission between the electronic device and the user input device; the memory 501 stores a program or instructions that can be executed on the processor 502, and when the processor 502 executes the program or instructions, the steps of the text generation method described in the above embodiments are implemented.
It is understood that the electronic device at least comprises a memory 501, a processor 502, a communication interface 503 and a bus 504, and the memory 501, the processor 502 and the communication interface 503 are connected in communication with each other through the bus 504, and can complete communication with each other, for example, the processor 502 reads a program or an instruction of a text generation method from the memory 501. In addition, the communication interface 503 may also implement a communication connection between the electronic device and a user input device, and may complete mutual information transmission, such as reading of initial text through the communication interface 503.
When the electronic device is running, the processor 502 calls the program or instructions in the memory 501 to perform the methods provided by the above-mentioned method embodiments, including for example: acquiring an initial text, wherein the initial text is generated based on a predefined template grammar; analyzing the initial text based on the template grammar to obtain a text structure body corresponding to the structure of the initial text; and generating a natural language text and the like for the description target based on the knowledge-graph data of the description target and the text structure.
The program or instructions in the memory 501 may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand-alone product. Alternatively, all or part of the steps for implementing the method embodiments may be implemented by hardware related to a program or instructions, where the program may be stored in a computer-readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The present invention also provides a non-transitory computer-readable storage medium according to the above embodiments, on which a program or instructions are stored, the program or instructions, when executed by a computer, implementing the steps of the text generation method according to the above embodiments, for example, including: acquiring an initial text, wherein the initial text is generated based on a predefined template grammar; analyzing the initial text based on the template grammar to obtain a text structure body corresponding to the structure of the initial text; and generating a natural language text and the like for the description target based on the knowledge-graph data of the description target and the text structure.
As a further aspect of the present invention, the present embodiment further provides a computer program product according to the above embodiments, the computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising a program or instructions, which when executed by a computer, the computer is capable of executing the text generation method provided by the above method embodiments, the method comprising: acquiring an initial text, wherein the initial text is generated based on a predefined template grammar; analyzing the initial text based on the template grammar to obtain a text structure body corresponding to the structure of the initial text; generating a natural language text for a description target based on knowledge-graph data of the description target and the text structure.
According to the electronic device, the non-transitory computer readable storage medium and the computer program product provided by the invention, through executing the steps of the text generation method described in the embodiments, template grammar is predefined, so that a non-development user can generate a logic meticulous initial text according to the template grammar, thereby effectively improving accuracy, and meanwhile, a machine can be used for automatically and quickly generating a target text by analyzing the initial text based on the template grammar to obtain a machine recognizable symbol, thereby effectively improving development efficiency.
It is to be understood that the above-described embodiments of the apparatus, the electronic device and the storage medium are merely illustrative, and that elements described as separate components may or may not be physically separate, may be located in one place, or may be distributed on different network elements. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on such understanding, the technical solutions mentioned above may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a usb disk, a removable hard disk, a ROM, a RAM, a magnetic or optical disk, etc., and includes several instructions for causing a computer device (such as a personal computer, a server, or a network device, etc.) to execute the methods described in the method embodiments or some parts of the method embodiments.
In addition, it should be understood by those skilled in the art that the terms "comprises," "comprising," or any other variation thereof, in the specification of the present invention, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
In the description of the present invention, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description. Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A text generation method, comprising:
acquiring an initial text, wherein the initial text is generated based on a predefined template grammar;
analyzing the initial text based on the template grammar to obtain a text structure body corresponding to the structure of the initial text;
generating a natural language text for a description target based on knowledge-graph data of the description target and the text structure.
2. The text generation method according to claim 1, further comprising, before the obtaining the initial text:
acquiring keywords in the template grammar, wherein the keywords are keywords associated with a text structure;
receiving user-defined information input by a user, and generating the initial text based on the user-defined information and the quoted keywords;
the referred keywords are keywords in the template grammar referred when the initial text is generated, and the custom information is information corresponding to the referred keywords input by a user.
3. The method according to claim 2, wherein the parsing the initial text to obtain a text structure corresponding to the structure of the initial text specifically includes:
reading data of the initial text, and acquiring the quoted keywords by matching the template grammar;
and determining an analysis strategy corresponding to the type based on the type of the quoted keyword, and acquiring a text structure body corresponding to the type in the initial text based on the analysis strategy.
4. The text generation method according to claim 2 or 3, wherein the keyword includes at least one of:
a line structure key, a paragraph dependency key, a rename key, a sub-condition key, and a highlight key.
5. The method according to claim 4, wherein the parsing the initial text to obtain a text structure corresponding to the structure of the initial text specifically includes:
traversing the data of the initial text to acquire the quoted keywords;
judging whether the traversal is finished or not, if not, judging whether the quoted keyword is the line structure keyword or not, if so, analyzing the line structure to obtain a line structure, otherwise, executing paragraph dependence keyword judgment, and judging whether the quoted keyword is the paragraph dependence keyword or not;
if the result of the paragraph dependency keyword judgment is a paragraph dependency keyword, obtaining a paragraph dependency structure by performing paragraph dependency analysis, otherwise, performing sub-condition keyword judgment, wherein the sub-condition keyword judgment is used for judging whether the cited keyword is the sub-condition keyword;
if the result of the sub-condition keyword judgment is a sub-condition keyword, acquiring a sub-condition structure body by performing sub-condition analysis, otherwise, executing highlight keyword judgment, wherein the highlight keyword judgment is used for judging whether the quoted keyword is the highlight keyword;
if the result of the judgment of the highlighted keyword is the highlighted keyword, a highlighted structural body is obtained by performing highlighted analysis, otherwise, judgment of renamed keywords is executed, the judgment of renamed keywords is used for judging whether the quoted keyword is the renamed keyword, and when the result of the judgment of renamed keywords is the renamed keyword, a renamed structural body is obtained by performing rename analysis.
6. The text generation method according to claim 5, wherein the literary structure comprises a branch structure comprising a logical judgment structure, and the generating of the natural language text for the description target specifically comprises:
traversing the line structure body and judging whether the traversal is completed,
if not, acquiring the branch structure, and when judging that the logic judgment structure in the branch structure meets logic conditions, performing text filling by using the knowledge map data to generate a branch text;
and if so, carrying out paragraph dependence processing on the branch text based on the paragraph dependence structure, and updating the whole branch text according to a processing result to obtain the natural language text.
7. The text generation method according to any one of claims 2 to 3 and 5 to 6, further comprising:
and displaying an operation interface, wherein the operation interface is used for receiving the operations of a user for writing template texts, modifying logic conditions, verifying the logic conditions and previewing the text effects.
8. A text generation apparatus, comprising:
the acquisition module is used for acquiring an initial text, and the initial text is generated based on a predefined template grammar;
the analysis module is used for analyzing the initial text based on the template grammar to obtain a text structure body corresponding to the structure of the initial text;
and the generation module is used for generating a natural language text aiming at the description target based on the knowledge graph data of the description target and the text structure body.
9. An electronic device comprising a memory, a processor and a program or instructions stored on the memory and executable on the processor, wherein the steps of the text generation method according to any one of claims 1 to 7 are implemented when the program or instructions are executed by the processor.
10. A non-transitory computer readable storage medium having stored thereon a program or instructions, wherein the program or instructions, when executed by a computer, implement the steps of the text generation method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011623657.5A CN112733515B (en) | 2020-12-31 | 2020-12-31 | Text generation method and device, electronic equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011623657.5A CN112733515B (en) | 2020-12-31 | 2020-12-31 | Text generation method and device, electronic equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112733515A true CN112733515A (en) | 2021-04-30 |
| CN112733515B CN112733515B (en) | 2022-11-11 |
Family
ID=75609359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011623657.5A Active CN112733515B (en) | 2020-12-31 | 2020-12-31 | Text generation method and device, electronic equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112733515B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113673966A (en) * | 2021-09-03 | 2021-11-19 | 海尔数字科技(青岛)有限公司 | Information security construction scheme generation method and device, electronic equipment and storage medium |
| CN117113961A (en) * | 2023-10-20 | 2023-11-24 | 中电数创(北京)科技有限公司 | Agent-based document writing method and system |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104572744A (en) * | 2013-10-23 | 2015-04-29 | 北大方正集团有限公司 | Structured document generating method and device |
| US20160078102A1 (en) * | 2014-09-12 | 2016-03-17 | Nuance Communications, Inc. | Text indexing and passage retrieval |
| CN106156365A (en) * | 2016-08-03 | 2016-11-23 | 北京智能管家科技有限公司 | A kind of generation method and device of knowledge mapping |
| CN107526717A (en) * | 2017-07-27 | 2017-12-29 | 山东科技大学 | A kind of method by structured process auto-building model natural language text |
| CN108763171A (en) * | 2018-04-20 | 2018-11-06 | 中国船舶重工集团公司第七〇九研究所 | A kind of document automation generation method based on format module |
| CN109783787A (en) * | 2018-12-29 | 2019-05-21 | 远光软件股份有限公司 | A kind of generation method of structured document, device and storage medium |
| US20190163745A1 (en) * | 2017-11-30 | 2019-05-30 | International Business Machines Corporation | Document preparation with argumentation support from a deep question answering system |
| CN110069763A (en) * | 2019-03-16 | 2019-07-30 | 平安科技（深圳）有限公司 | Contract text method for customizing, device, equipment and readable storage medium storing program for executing |
| CN110705244A (en) * | 2019-10-12 | 2020-01-17 | 武汉灯塔之光科技有限公司 | Method and device for automatically generating article based on template |
| CN110765753A (en) * | 2019-12-27 | 2020-02-07 | 广东博智林机器人有限公司 | Method, system, computer device and storage medium for generating file |
| CN110826340A (en) * | 2019-11-06 | 2020-02-21 | 广东三维家信息科技有限公司 | Evaluation text generation method and device and electronic equipment |
| CN111176650A (en) * | 2018-11-09 | 2020-05-19 | 阿里巴巴集团控股有限公司 | Parser generation method, search method, server, and storage medium |
| CN111177184A (en) * | 2019-12-24 | 2020-05-19 | 深圳壹账通智能科技有限公司 | Structured query language conversion method based on natural language and related equipment thereof |
| CN111832275A (en) * | 2020-09-21 | 2020-10-27 | 北京百度网讯科技有限公司 | Text creation method, device, equipment and storage medium |
| CN111859950A (en) * | 2020-06-18 | 2020-10-30 | 达而观信息科技（上海）有限公司 | Method for automatically generating lecture notes |
-
2020
- 2020-12-31 CN CN202011623657.5A patent/CN112733515B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104572744A (en) * | 2013-10-23 | 2015-04-29 | 北大方正集团有限公司 | Structured document generating method and device |
| US20160078102A1 (en) * | 2014-09-12 | 2016-03-17 | Nuance Communications, Inc. | Text indexing and passage retrieval |
| CN106156365A (en) * | 2016-08-03 | 2016-11-23 | 北京智能管家科技有限公司 | A kind of generation method and device of knowledge mapping |
| CN107526717A (en) * | 2017-07-27 | 2017-12-29 | 山东科技大学 | A kind of method by structured process auto-building model natural language text |
| US20190163745A1 (en) * | 2017-11-30 | 2019-05-30 | International Business Machines Corporation | Document preparation with argumentation support from a deep question answering system |
| CN108763171A (en) * | 2018-04-20 | 2018-11-06 | 中国船舶重工集团公司第七〇九研究所 | A kind of document automation generation method based on format module |
| CN111176650A (en) * | 2018-11-09 | 2020-05-19 | 阿里巴巴集团控股有限公司 | Parser generation method, search method, server, and storage medium |
| CN109783787A (en) * | 2018-12-29 | 2019-05-21 | 远光软件股份有限公司 | A kind of generation method of structured document, device and storage medium |
| CN110069763A (en) * | 2019-03-16 | 2019-07-30 | 平安科技（深圳）有限公司 | Contract text method for customizing, device, equipment and readable storage medium storing program for executing |
| CN110705244A (en) * | 2019-10-12 | 2020-01-17 | 武汉灯塔之光科技有限公司 | Method and device for automatically generating article based on template |
| CN110826340A (en) * | 2019-11-06 | 2020-02-21 | 广东三维家信息科技有限公司 | Evaluation text generation method and device and electronic equipment |
| CN111177184A (en) * | 2019-12-24 | 2020-05-19 | 深圳壹账通智能科技有限公司 | Structured query language conversion method based on natural language and related equipment thereof |
| CN110765753A (en) * | 2019-12-27 | 2020-02-07 | 广东博智林机器人有限公司 | Method, system, computer device and storage medium for generating file |
| CN111859950A (en) * | 2020-06-18 | 2020-10-30 | 达而观信息科技（上海）有限公司 | Method for automatically generating lecture notes |
| CN111832275A (en) * | 2020-09-21 | 2020-10-27 | 北京百度网讯科技有限公司 | Text creation method, device, equipment and storage medium |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113673966A (en) * | 2021-09-03 | 2021-11-19 | 海尔数字科技(青岛)有限公司 | Information security construction scheme generation method and device, electronic equipment and storage medium |
| CN113673966B (en) * | 2021-09-03 | 2024-03-08 | 卡奥斯数字科技(青岛)有限公司 | Information security construction scheme generation method and device, electronic equipment and storage medium |
| CN117113961A (en) * | 2023-10-20 | 2023-11-24 | 中电数创(北京)科技有限公司 | Agent-based document writing method and system |
| CN117113961B (en) * | 2023-10-20 | 2024-02-09 | 中电数创(北京)科技有限公司 | Agent-based document writing method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112733515B (en) | 2022-11-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110192210B (en) | Construction and processing of computational graphs for dynamically structured machine learning models | |
| Magnini et al. | The excitement open platform for textual inferences | |
| CN107797991B (en) | Dependency syntax tree-based knowledge graph expansion method and system | |
| Leopold et al. | Generating natural language texts from business process models | |
| US7143026B2 (en) | Generating rules to convert HTML tables to prose | |
| CN110532573A (en) | A kind of interpretation method and system | |
| CN103678285A (en) | Machine translation method and machine translation system | |
| RU2011122784A (en) | METHOD FOR SEMANTIC PROCESSING OF NATURAL LANGUAGE USING GRAPHIC INTERMEDIARY LANGUAGE | |
| CN112733515B (en) | Text generation method and device, electronic equipment and readable storage medium | |
| CN107798123A (en) | Knowledge base and its foundation, modification, intelligent answer method, apparatus and equipment | |
| CN114625844B (en) | Code searching method, device and equipment | |
| CN108563488A (en) | Model training method and device, the method and device for building container mirror image | |
| Bautista et al. | Aratraductor: text to pictogram translation using natural language processing techniques | |
| CN107908743B (en) | Artificial intelligence application construction method and device | |
| WO2023278052A1 (en) | Automated troubleshooter | |
| CN108664464B (en) | Method and device for determining semantic relevance | |
| CN111553138A (en) | Auxiliary writing method and device for standardizing content structure document | |
| Chevelu et al. | ROOTS: a toolkit for easy, fast and consistent processing of large sequential annotated data collections. | |
| Bender | Language CoLLAGE: Grammatical Description with the LinGO Grammar Matrix. | |
| El-ghobashy et al. | A proposed framework for Arabic semantic annotation tool | |
| CN115906818A (en) | Grammar knowledge prediction method, grammar knowledge prediction device, electronic equipment and storage medium | |
| CN115115984A (en) | Video data processing method, apparatus, program product, computer device, and medium | |
| CN113421657A (en) | Construction method and device of knowledge representation model of clinical practice guideline | |
| Sibeko et al. | An overview of Sesotho BLARK content | |
| CN112749553A (en) | Text information processing method and device for video file and server |
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 |