CN115238653A - Report generation method, device, equipment and medium - Google Patents
Report generation method, device, equipment and medium Download PDFInfo
- Publication number
- CN115238653A CN115238653A CN202210817665.6A CN202210817665A CN115238653A CN 115238653 A CN115238653 A CN 115238653A CN 202210817665 A CN202210817665 A CN 202210817665A CN 115238653 A CN115238653 A CN 115238653A
- Authority
- CN
- China
- Prior art keywords
- report
- content
- existing
- subsequent reference
- generating
- 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.)
- Pending
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/253—Grammatical analysis; Style critique
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/258—Heading extraction; Automatic titling; Numbering
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Machine Translation (AREA)
Abstract
The invention discloses a report generating method, which comprises the following steps: receiving an instruction for representing the generation of a report by using a learning model, and reading the existing report content in the report; inputting the pre-existing report content to a pre-set learning model to obtain subsequent reference content for generating the report; determining the language style and grammar specification of the report according to the existing report content; revising the subsequent reference content according to the linguistic style and grammatical specifications of the report; and generating the report according to the existing report content and the corrected subsequent reference content.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a report generation method, an apparatus, a computer device, and a computer-readable storage medium.
Background
The information industry needs to report the latest hot spots, so that reports need to be output timely, and the reports are generally edited manually by editors, so that the efficiency is low. Automated reporting products in this regard are also currently on the market, which automatically generate reports based on the number of hotspots and words selected by the user, but which are generated entirely by the algorithm itself and which are difficult for the user to intervene in the generation of the reports.
Aiming at the technical problems that the report flexibility is poor and the idea of an author is difficult to express caused by the autonomous generation of the report through an algorithm in the prior art, an effective solution scheme does not exist at present.
Disclosure of Invention
The invention aims to provide a report generation method, a report generation device, computer equipment and a computer readable storage medium, which can solve the technical problems that the report is poor in flexibility and the thought of an author is difficult to express due to the fact that the report is autonomously generated through an algorithm in the prior art.
One aspect of the present invention provides a report generation method, including: receiving an instruction for representing the generation of a report by using a learning model, and reading the existing report content in the report; inputting the existing report content into a preset learning model to obtain subsequent reference content for generating the report; determining the language style and grammar specification of the report according to the existing report content; revising the subsequent reference content according to the linguistic style and grammatical specifications of the report; and generating the report according to the existing report content and the corrected subsequent reference content.
Optionally, the determining the language style and the grammar specification of the report according to the existing report content includes: analyzing the existing report content by using a natural language generation technology to determine a language style and a grammar specification of the existing report content; and taking the language style of the existing report content as the language style of the report, and taking the grammar specification of the existing report content as the grammar specification of the report.
Optionally, the modifying the subsequent reference content according to the linguistic style and grammatical specification of the report includes: generating a directory structure of the subsequent reference content according to the existing report content; the directory structure includes a plurality of paragraph titles; dividing the subsequent reference content under corresponding paragraph titles of the directory structure; and modifying the content divided under each paragraph title according to the language style and the grammar specification of the report.
Optionally, the generating a directory structure of the subsequent reference content according to the existing report content includes: acquiring paragraph titles of the existing report contents; and generating a directory structure of the subsequent reference content according to the paragraph titles of the existing report content and the subsequent reference content by utilizing a natural language generation technology.
Optionally, before the instruction for generating a report using a learning model reads the report content already existing in the report, the method further comprises: detecting whether a specific symbol for starting a cloud word stock exists in the content input to the report in real time in the process of editing the report; the cloud word stock is used for indexing a corresponding characteristic value according to the input characteristic words; when the specific symbol is detected, extracting a feature word limited by the specific symbol; inputting the characteristic words into the cloud word stock to obtain characteristic values of the characteristic words; replacing the feature words with the feature values in the report.
Optionally, the specific symbol includes a first label and a second label, and the extracting the feature word defined by the specific symbol includes: and extracting the content between the first label and the second label as a feature word defined by the specific symbol.
Another aspect of the present invention provides a report generating apparatus, including: the reading module is used for receiving an instruction for generating a report by using a learning model, and reading the report content existing in the report; an input module for inputting the existing report content to a preset learning model to obtain subsequent reference content for generating the report; the determining module is used for determining the language style and the grammar specification of the report according to the existing report content; a revising module for revising the subsequent reference content according to the language style and grammar specification of the report; a generating module for generating the report according to the existing report content and the corrected subsequent reference content.
Optionally, the determining module is specifically configured to: analyzing the existing report content by using a natural language generation technology to determine a language style and a grammar specification of the existing report content; and taking the language style of the existing report content as the language style of the report, and taking the grammar specification of the existing report content as the grammar specification of the report.
Yet another aspect of the present invention provides a computer apparatus, comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the report generation method of any of the above embodiments when executing the computer program.
Yet another aspect of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a report generation method as described in any of the embodiments above.
According to the report generation method provided by the invention, the report is not completely generated by the algorithm, but the subsequent reference content of the report is generated according to the algorithm based on the existing content in the report, the language style and the grammar specification of the report are determined according to the existing content, and the subsequent reference content generated by the algorithm is corrected according to the language style and the grammar specification of the report, so that the overall style of the finally generated report is consistent with the style of the existing content in the report, and no error exists in grammar, and the technical problems that the report generated by the algorithm in the prior art is poor in flexibility and difficult to express the thought of an author are solved.
Drawings
Various additional advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a flow diagram illustrating a method for report generation in a first embodiment;
FIG. 2 is a schematic diagram illustrating a cloud word stock in accordance with an embodiment;
FIG. 3 is a diagram illustrating a suggested input in accordance with one embodiment;
FIG. 4 is a diagram illustrating a report generation process in a first embodiment;
fig. 5 shows a block diagram of a report generation apparatus in the second embodiment;
fig. 6 shows a block diagram of a computer device suitable for implementing the report generation method according to the third embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the 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.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, 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 one of 8230, and" comprising 8230does not exclude the presence of additional like elements in a process, method, article, or apparatus comprising the element.
Example one
Fig. 1 shows a flowchart of a report generation method in the first embodiment. As shown in fig. 1, the method includes steps S1 to S5, wherein:
step S1, receiving an instruction for representing the generation of a report by using a learning model, and reading the existing report content in the report.
The learning model of the present embodiment is capable of outputting subsequent references to the report based on the content of the report already present in the report, i.e., some content of the report must be present in the report before receiving instructions characterizing the generation of the report using the learning model.
The inventor researches and discovers that production and manufacturing enterprises generally frequently make reports in order to monitor the production condition of products and the progress of the enterprises, and the reports have a characteristic that relevant indexes in the reports do not change within a period of time, so that the automation of the reports generally makes templates in advance, and each production report generates a new report according to a fixed template. However, in the prior art, a mode of setting a template in advance is to set a fixed tag in the template, calculate a relevant tag value at a background and replace the tag value when outputting a document, and the mode has extremely poor flexibility, and if the template needs to be replaced, the tag needs to be recalculated; in addition, for the information report, this template method is completely not preferable, because the information report is followed by the temporal change, so the information report cannot predict the required label in advance. Therefore, the embodiment provides a cloud word stock to solve the problems of templates and labels in the prior art. Specifically, before step S1, the method further includes steps A1 to A4, wherein:
a1, in the process of editing the report, detecting whether a specific symbol for starting a cloud word stock exists in the content input to the report in real time; the cloud word stock is used for indexing a corresponding characteristic value according to the input characteristic words;
step A2, when the specific symbol is detected, extracting a feature word limited by the specific symbol;
step A3, inputting the characteristic words into the cloud word stock to obtain characteristic values of the characteristic words;
and A4, replacing the characteristic words with the characteristic values in the report.
Specifically, in the process of editing the report so that the existing report content exists in the report, a function of detecting whether a specific symbol is input in real time can be triggered, wherein the specific symbol is used for invoking a cloud word stock, a schematic diagram of the cloud word stock is shown in fig. 2, the cloud word stock can store data by using an inverted index, and when data is queried in the cloud word stock, the relevance between a characteristic word and a characteristic value can be calculated by a big data algorithm, and then the query result is sorted according to the relevance. For example, a user inputs "# shanghai population total #", "#. #" may be a specific symbol for invoking a cloud word stock, "shanghai population total" may be a feature word defined by the specific symbol, the cloud word stock queries data related to "shanghai population total" based on an inverted index, and queries feature values corresponding to the feature words such as "shanghai population number", "shanghai population base", "shanghai population total", "national population total", and finally, the feature values fed back by the cloud word stock based on "shanghai population number" may be "3600 million population" and "thirty-six million population". After the user selects the desired format, such as 3600 million people, "the system will automatically replace the" # Shanghai population total # "displayed in the report with" 3600 million people.
Optionally, the specific symbol includes a first index and a second index, and step A2 may include:
and extracting the content between the first label and the second label as the characteristic words defined by the specific symbols.
Specifically, the first reference number may include one or more symbols, the second reference number may also include one or more symbols, and the first reference number and the second reference number may be the same or different. For example, in connection with the above example, the first label is the first "#" and the second label is the second "#". When the system detects the first label, whether a second label appears is detected in real time within a preset time, if so, the user is considered to input a specific symbol, and the content between the two labels can be extracted as a specific word limited by the specific symbol.
And S2, inputting the existing report content into a preset learning model to obtain subsequent reference content for generating the report.
In the embodiment, the defect of preparing templates and labels in advance can be overcome through the cloud word stock, and the flexibility of writing reports is ensured, but the intellectualization of products cannot be reflected, so that the inventor further researches 'associative input'. As shown in fig. 3, the associative input can integrate technologies such as machine learning, big data analysis, and natural language processing (including natural language understanding and natural language generation), aggregate network information, and integrate a plurality of domains to perform deep analysis on data. Specifically, the input of the user can be used as a data source of machine learning, the context of the user report is analyzed through a machine learning algorithm and big data analysis, information conforming to the context of the user report is calculated from massive data, and the data is output in the language style of the user by adopting a natural language processing technology after the conforming information is found out. Although there is a scheme related to association input in the prior art, the association input in this embodiment is directed to a report, and it is capable of outputting subsequent reference content of the report according to the content of the report currently existing in the report, which is substantially different from the association input method in the prior art that associates terms that may be subsequently input by a user according to terms input by the user.
The learning model may be a part of the associative input, and the neural network model is obtained by learning a training set, where the training set is obtained by a plurality of history reports, the training set includes input samples and output samples, each input sample is a first few paragraphs in each history report, and each output sample is a second few paragraphs in each history report, where each input sample and the corresponding output sample together form a complete history report. For example, a history sample has 10 paragraphs, and if the input sample is the first 1 paragraph of the history report, the corresponding output sample is the last 9 paragraphs; for another example, if the input samples are the first 6 paragraphs of the history report, the corresponding output samples are the last 4 paragraphs.
Different types of reports correspond to different learning models. For example, the report of the financial type corresponds to a machine learning model of the financial type, and the required training set is a historical report of the financial type; the economic type report corresponds to an economic type machine learning model, and the required training set is a historical economic type report. Therefore, before step S2, the type of the report may be determined according to the existing report content, and then the learning model associated with the type of the report may be screened from all the trained learning models as the learning model to be used this time.
And S3, determining the language style and the grammar specification of the report according to the existing report content.
And S4, correcting the subsequent reference content according to the language style and the grammar specification of the report.
Since the subsequent reference content output by the learning model is not necessarily completely applicable to the report, and there may be syntax errors or style errors, the subsequent reference content output by the learning model also needs to be adjusted according to the language style and syntax specification of the report.
Since the existing report contents are inputted by the user himself, the language style of the existing report contents may be used as the language style of the report and the grammar specification of the existing report contents may be used as the grammar specification of the report in order to make the generated report closer to the idea of the user himself. Alternatively, step S3 may include steps S31 to S32, where:
step S31, analyzing the existing report content by using a natural language generation technology to determine the language style and the grammar specification of the existing report content;
step S32, the language style of the existing report content is used as the language style of the report, and the grammar specification of the existing report content is used as the grammar specification of the report.
The Natural Language Generation (NLG) technology refers to converting structured data into text and writing information in human Language by a computer. The main benefit of natural language generation technology is that it can transform a data set into a clear narrative that is understood by humans, and it can generate rich information when processing statistical data present in a spreadsheet, unlike natural language processing, which evaluates only text to form insights through which data can be accurately evaluated, analyzed, and conveyed.
Alternatively, step S4 may include steps S41 to S43, where:
step S41, generating a directory structure of the subsequent reference content according to the existing report content; the directory structure includes a plurality of paragraph titles;
the directory structure is for example: abstract, background introduction, preamble, etc., and the directory structure may include a plurality of paragraph titles at the same level and/or a plurality of paragraph titles at different levels. Alternatively, step S41 may include steps S411 to S42, in which:
step S411, obtaining paragraph titles of the existing report content;
step S412, generating a directory structure of the subsequent reference content according to the paragraph header of the existing report content and the subsequent reference content by using a natural language generation technology.
The generated directory structure does not include the paragraph headers of the already existing report content. The natural language generation technology can analyze subsequent reference content, divide paragraphs used for describing the same or similar content into a group, then refine main body ideas of all paragraphs divided in the same group to serve as paragraph titles of the group, repeat the steps until all the paragraph titles are obtained, and finally generate a target structure of the subsequent reference content according to the front-back sequence and the level relation of the paragraph titles.
Step S42, dividing the subsequent reference content into corresponding paragraph titles of the directory structure;
and S43, correcting the content divided under each paragraph title according to the reported language style and grammar specification.
By understanding the subsequent reference content output by the machine learning model, the subsequent reference content is divided under the corresponding paragraph titles. Since the subsequent reference content may have syntax errors and/or differ from the user writing style, the present embodiment can adjust the content under each paragraph title according to the recognized language style and syntax specification by understanding the language style and syntax specification in the report front. For example, when a certain paragraph under a certain paragraph title appears with a plurality of words or sentences of the same level, the above paragraph is usually used as a proportional sentence.
And S5, generating the report according to the existing report content and the corrected subsequent reference content.
The existing report content and the corrected subsequent reference content can be directly used as the complete content of the report, or the user can secondarily correct the corrected subsequent reference content and then use the secondarily corrected content of the user and the existing report content as the complete content of the report.
The technical scheme of the application has the following advantages: 1. the problem of setting up the template in advance can be solved in this application: the template setting is cancelled, the report flexibility is increased, the user can freely edit, and related functions are called when automatic output is required. 2. The accuracy of the data is improved by using the cloud word stock: the cloud word stock is accessed into big data, the integrity of the data is guaranteed, the data depth is improved, and meanwhile a plurality of results are returned according to the association degree, so that a user can independently select a required result. 3. Using the "associative input" function: accurately calculating characters behind the user according to machine learning and big data and context; the output result conforms to the grammar specification and the language style of an author by using natural language processing, and the intellectualization of the system is ensured.
As shown in fig. 4, in the process of editing the report by the user, a cloud word stock may be evoked by using a specific symbol, that is, a special tag, and then the cloud word stock requests the big data center to query the data of the relevant tag, that is, the characteristic value, and then the big data center returns the ranked query result through the cloud word stock, and the user autonomously selects the required characteristic value; and further calling an association function in the process of editing the report by the user, analyzing the context of the report by the machine learning model to output subsequent reference contents, analyzing the language style and the grammar specification of the front by natural language processing, particularly natural language generating technology, and then correcting the subsequent reference contents according to the analyzed language style and the grammar specification.
Example two
The second embodiment of the present invention provides a report generating device, which corresponds to the method provided by the first embodiment, and corresponding technical features and technical effects are not described in detail in this embodiment, and reference may be made to the first embodiment for related points. Specifically, fig. 5 shows a block diagram of the report generation apparatus in the second embodiment. As shown in fig. 5, the report generating apparatus 500 may include:
a reading module 501, configured to receive an instruction for characterizing generation of a report by using a learning model, and read report content existing in the report;
an input module 502, configured to input the existing report content into a preset learning model to obtain a subsequent reference content for generating the report;
a determining module 503, configured to determine a linguistic style and a grammatical specification of the report according to the existing report content;
a revising module 504 for revising the subsequent reference content according to the linguistic style and grammatical specifications of the report;
a generating module 505, configured to generate the report according to the existing report content and the corrected subsequent reference content.
Optionally, the determining module is specifically configured to: analyzing the existing report content by using a natural language generation technology to determine a language style and a grammar specification of the existing report content; and taking the language style of the existing report content as the language style of the report, and taking the grammar specification of the existing report content as the grammar specification of the report.
Optionally, the correction module includes: a generating unit, configured to generate a directory structure of the report according to the existing report content; the directory structure includes a plurality of paragraph titles; a dividing unit, configured to divide the subsequent reference content into the corresponding paragraph titles of the directory structure; and the correcting unit is used for correcting the content divided under each paragraph title according to the language style and the grammar specification of the report.
Optionally, the generating unit is specifically configured to: acquiring paragraph titles of the existing report contents; and generating a directory structure of the subsequent reference content according to the paragraph titles of the existing report content and the subsequent reference content by utilizing a natural language generation technology.
Optionally, the apparatus further comprises: the detection module is used for detecting whether a specific symbol for starting a cloud word stock exists in the content input to the report in real time during the process of editing the report and before the instruction responding to the instruction for generating the report by using the learning model reads the existing report content in the report; the cloud word stock is used for indexing a corresponding characteristic value according to the input characteristic words; the extracting module is used for extracting the feature words limited by the specific symbols when the specific symbols are detected; the index catalogue is used for inputting the characteristic words into the cloud word stock so as to obtain characteristic values of the characteristic words; a replacement module for replacing the feature words with the feature values in the report.
Optionally, the specific symbol includes a first label and a second label, and the extraction module is specifically configured to: and extracting the content between the first label and the second label as a feature word defined by the specific symbol.
EXAMPLE III
Fig. 6 shows a block diagram of a computer device suitable for implementing a report generation method according to the third embodiment. In this embodiment, the computer device 600 may be a smart phone, a tablet computer, a notebook computer, a desktop computer, a rack server, a blade server, a tower server, or a rack server (including an independent server or a server cluster composed of a plurality of servers), and the like that execute programs. As shown in fig. 6, the computer device 600 of the present embodiment includes at least but is not limited to: a memory 601, a processor 602, a network interface 603, which may be communicatively coupled to each other via a system bus. It should be noted that fig. 6 only shows a computer device 600 with components 601-603, but it should be understood that not all of the shown components are required to be implemented, and more or fewer components may be implemented instead.
In this embodiment, the memory 603 includes at least one type of computer-readable storage medium, which includes flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a Programmable Read Only Memory (PROM), a magnetic memory, a magnetic disk, an optical disk, and the like. In some embodiments, the storage 601 may be an internal storage unit of the computer device 600, such as a hard disk or a memory of the computer device 600. In other embodiments, the memory 601 may also be an external storage device of the computer device 600, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), etc. provided on the computer device 600. Of course, the memory 601 may also include both internal and external storage units for the computer device 600. In the present embodiment, the memory 601 is generally used for storing an operating system and various types of application software installed in the computer apparatus 600, such as program codes of report generation methods and the like. The report generation method comprises the following steps: receiving an instruction for representing the generation of a report by using a learning model, and reading the existing report content in the report; inputting the existing report content into a preset learning model to obtain subsequent reference content for generating the report; determining the language style and grammar specification of the report according to the existing report content; revising the subsequent reference content according to the linguistic style and grammatical specifications of the report; generating the report according to the existing report content and the revised subsequent reference content.
Processor 602 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data Processing chip in some embodiments. The processor 602 generally operates to control the overall operation of the computer device 600. Such as performing control and processing related to data interaction or communication with the computer device 600. In this embodiment, the processor 602 is configured to execute the program code of the report generation method stored in the memory 601.
In this embodiment, the report generation method stored in the memory 601 may be further divided into one or more program modules and executed by one or more processors (in this embodiment, the processor 602) to complete the present invention.
The network interface 603 may comprise a wireless network interface or a wired network interface, and the network interface 603 is typically used to establish communication links between the computer device 600 and other computer devices. For example, the network interface 603 is used to connect the computer apparatus 600 to an external terminal via a network, establish a data transmission channel and a communication link between the computer apparatus 600 and the external terminal, and the like. The network may be an Intranet (Internet), the Internet (Internet), a Global System of Mobile communication (GSM), wideband Code Division Multiple Access (WCDMA), a 4G network, a 5G network, bluetooth (Bluetooth), wi-Fi, or other wireless or wired network.
Example four
The present embodiments also provide a computer-readable storage medium including a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, a server, an App application store, etc., having stored thereon a computer program that, when executed by a processor, implements the steps of the report generating method. The report generation method comprises the following steps: receiving an instruction for generating a report by using a learning model, and reading the existing report content in the report; inputting the existing report content into a preset learning model to obtain subsequent reference content for generating the report; determining the language style and grammar specification of the report according to the existing report content; revising the subsequent reference content according to the language style and the grammar specification of the report; generating the report according to the existing report content and the revised subsequent reference content.
It will be apparent to those skilled in the art that the modules or steps of the embodiments of the invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, embodiments of the invention are not limited to any specific combination of hardware and software.
It should be noted that the numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the description of the foregoing embodiments, it is clear to those skilled in the art that the method of the foregoing embodiments may be implemented by software plus a necessary general hardware platform, and certainly may also be implemented by hardware, but in many cases, the former is a better implementation.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A method of report generation, the method comprising:
receiving an instruction for representing the generation of a report by using a learning model, and reading the existing report content in the report;
inputting the existing report content into a preset learning model to obtain subsequent reference content for generating the report;
determining the language style and grammar specification of the report according to the existing report content;
revising the subsequent reference content according to the linguistic style and grammatical specifications of the report;
and generating the report according to the existing report content and the corrected subsequent reference content.
2. The method of claim 1, wherein said determining a linguistic style and grammatical specification of the report based on the existing report content comprises:
analyzing the existing report content by using a natural language generation technology to determine a language style and a grammar specification of the existing report content;
and taking the language style of the existing report content as the language style of the report, and taking the grammar specification of the existing report content as the grammar specification of the report.
3. The method of claim 1, wherein said revising said subsequent reference content according to said reported linguistic and grammatical specifications comprises:
generating a directory structure of the subsequent reference content according to the existing report content; the directory structure includes a plurality of paragraph titles;
dividing the subsequent reference content under corresponding paragraph titles of the directory structure;
and correcting the content divided under each paragraph title according to the language style and the grammar specification of the report.
4. The method of claim 3, wherein generating the directory structure of the report based on the existing report content comprises:
acquiring paragraph titles of the existing report content;
and generating a directory structure of the subsequent reference content according to the paragraph titles of the existing report content and the subsequent reference content by utilizing a natural language generation technology.
5. The method of claim 1, wherein prior to the reading report content already present in the report in response to the instructions to generate a report using a learning model, the method further comprises:
detecting whether a specific symbol for starting a cloud word stock exists in the content input to the report in real time in the process of editing the report; the cloud word stock is used for indexing a corresponding characteristic value according to the input characteristic words;
when the specific symbol is detected, extracting the feature words defined by the specific symbol;
inputting the characteristic words into the cloud word stock to obtain characteristic values of the characteristic words;
replacing the feature words with the feature values in the report.
6. The method according to claim 5, wherein the specific symbol includes a first label and a second label, and the extracting the feature word defined by the specific symbol includes:
and extracting the content between the first label and the second label as a feature word defined by the specific symbol.
7. An apparatus for generating a report, the apparatus comprising:
the reading module is used for receiving an instruction for generating a report by using a learning model, and reading the report content existing in the report;
an input module for inputting the existing report content to a preset learning model to obtain subsequent reference content for generating the report;
a determining module for determining the language style and grammar specification of the report according to the existing report content;
a correction module for correcting the subsequent reference content according to the linguistic style and grammatical specification of the report;
a generating module for generating the report according to the existing report content and the corrected subsequent reference content.
8. The apparatus of claim 7, wherein the determining module is specifically configured to:
analyzing the existing report content by using a natural language generation technology to determine a language style and a grammar specification of the existing report content;
and taking the language style of the existing report content as the language style of the report, and taking the grammar specification of the existing report content as the grammar specification of the report.
9. A computer device, the computer device comprising: memory, processor and computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 6 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210817665.6A CN115238653A (en) | 2022-07-12 | 2022-07-12 | Report generation method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210817665.6A CN115238653A (en) | 2022-07-12 | 2022-07-12 | Report generation method, device, equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115238653A true CN115238653A (en) | 2022-10-25 |
Family
ID=83672995
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210817665.6A Pending CN115238653A (en) | 2022-07-12 | 2022-07-12 | Report generation method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115238653A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117934229A (en) * | 2024-03-18 | 2024-04-26 | 新励成教育科技股份有限公司 | Originality excitation-based talent training guiding method, system, equipment and medium |
-
2022
- 2022-07-12 CN CN202210817665.6A patent/CN115238653A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117934229A (en) * | 2024-03-18 | 2024-04-26 | 新励成教育科技股份有限公司 | Originality excitation-based talent training guiding method, system, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111259873B (en) | Table data extraction method and device | |
| CN111309912A (en) | Text classification method and device, computer equipment and storage medium | |
| CN110516063A (en) | A kind of update method of service system, electronic equipment and readable storage medium storing program for executing | |
| CN112163424A (en) | Data labeling method, device, equipment and medium | |
| CN111159982B (en) | Document editing method, device, electronic equipment and computer readable storage medium | |
| CN109543153B (en) | Sequence labeling system and method | |
| CN109710933A (en) | Acquisition methods, device, computer equipment and the storage medium of training corpus | |
| CN114238629A (en) | Language processing method and device based on automatic prompt recommendation and terminal | |
| CN115391439B (en) | Document data export method, device, electronic equipment and storage medium | |
| CN112395425A (en) | Data processing method and device, computer equipment and readable storage medium | |
| CN111737577A (en) | Data query method, device, equipment and medium based on service platform | |
| CN115238653A (en) | Report generation method, device, equipment and medium | |
| CN113283231B (en) | Method for acquiring signature bit, setting system, signature system and storage medium | |
| CN109753976B (en) | Corpus labeling device and method | |
| CN102662953B (en) | With the semantic tagger system and method that input method is integrated | |
| CN110737432A (en) | script aided design method and device based on root list | |
| WO2019080419A1 (en) | Method for building standard knowledge base, electronic device, and storage medium | |
| CN115982272A (en) | Data labeling method and device for urban big data management and computer storage medium | |
| CN110597874B (en) | Data analysis model creation method and device, computer equipment and storage medium | |
| CN116306573B (en) | Intelligent analysis method, device and equipment for engineering practice and readable storage medium | |
| CN110727767A (en) | Method and system for expanding text sample | |
| CN117851575B (en) | Large language model question-answer optimization method and device, electronic equipment and storage medium | |
| CN116757203B (en) | Natural language matching method, device, computer equipment and storage medium | |
| CN113076454B (en) | Artificial intelligence-based element number analysis method and server | |
| CN110955433B (en) | Automatic deployment script generation method and device |
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 |