CN116579305A - Intelligent word document editing and generating system - Google Patents
Intelligent word document editing and generating system Download PDFInfo
- Publication number
- CN116579305A CN116579305A CN202310528361.2A CN202310528361A CN116579305A CN 116579305 A CN116579305 A CN 116579305A CN 202310528361 A CN202310528361 A CN 202310528361A CN 116579305 A CN116579305 A CN 116579305A
- Authority
- CN
- China
- Prior art keywords
- document
- generation
- generation mode
- picture
- paragraph
- 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
- 238000011156 evaluation Methods 0.000 claims abstract description 32
- 238000007689 inspection Methods 0.000 claims abstract description 20
- 238000000034 method Methods 0.000 claims abstract description 20
- 238000004458 analytical method Methods 0.000 claims abstract description 14
- 238000012790 confirmation Methods 0.000 claims abstract description 9
- 238000000605 extraction Methods 0.000 claims abstract description 9
- 238000012937 correction Methods 0.000 claims description 24
- 238000004364 calculation method Methods 0.000 claims description 22
- 238000010586 diagram Methods 0.000 claims description 8
- 239000013598 vector Substances 0.000 claims description 7
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 8
- 238000003780 insertion Methods 0.000 description 7
- 230000037431 insertion Effects 0.000 description 7
- 238000005457 optimization Methods 0.000 description 5
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/53—Querying
- G06F16/532—Query formulation, e.g. graphical querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/189—Automatic justification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Library & Information Science (AREA)
- Mathematical Physics (AREA)
- Machine Translation (AREA)
Abstract
The invention relates to the technical field of word document editing and generating, and particularly discloses an intelligent word document editing and generating system, which comprises an editing data acquisition module, a generating mode confirmation module, an initial document information extraction module, an initial document arrangement module, a final document inspection module, a platform document generation and evaluation module, an early warning terminal and a cloud database; according to the invention, through confirming the generation mode, the documents in different generation modes are subjected to arrangement analysis and processing, so that a final document is generated, and the analysis is performed according to the inspection information of each generated document, so that a document generation deviation trend evaluation index is obtained, the problems of complex process, low efficiency and the like in the current manual editing are effectively solved, the pertinence and the flexibility editing of the document are realized, the efficiency of document editing generation is improved, the different document generation requirements of a user are met, the practicability is high, and the viscosity of the user and a document generation platform is further maintained.
Description
Technical Field
The invention relates to the technical field of word document editing and generating, in particular to an intelligent word document editing and generating system.
Background
With the advent of the information age, document editing generation has become an unavoidable part in daily work and study of people, through document editing generation, people can arrange ideas into texts with clear logic and complete structure, and information transmission and sharing are convenient, so that in order to improve document editing generation efficiency, automatic document editing generation needs to be more urgent.
The existing document editing generation mode is manual editing, and obviously, the manual editing mode has the following problems: 1. the problems of complicated process, low efficiency and the like are solved, the quality and the accuracy of the document are difficult to ensure, and meanwhile, the editing generation precision and the suitability of the document are low.
2. In the image-text form document editing mode, the insertion basis of the image only depends on the matching degree of the image elements and the keywords in the text paragraphs, and when the keyword repetition rate of each paragraph of the document is high, the insertion position of the image cannot be determined, so that the possibility of deviation of the insertion position of the image is high.
3. The document automatic generation platform is not subjected to deep analysis on the accurate generation condition, so that the error correction effect and the optimization effect of the document automatic generation platform are not obvious, further, the document generation requirement of a user cannot be met, and meanwhile, the viscosity of the platform and the user cannot be improved.
Disclosure of Invention
In view of this, in order to solve the problems set forth in the above background art, an intelligent word document editing and generating system is now proposed.
The aim of the invention can be achieved by the following technical scheme: the invention provides an intelligent word document editing and generating system, which comprises the following steps: and the editing data acquisition module is used for acquiring the data to be edited uploaded by the current user in the document generation platform.
The generation mode confirming module is used for confirming the current document generation mode according to the data to be edited uploaded by the current user, generating a final document of the single-text generation mode if the current document generation mode is the single-text generation mode, starting the final document checking module at the same time, confirming an initial document of the single-text generation mode or the image-text combination generation mode if the current document generation mode is the single-text generation mode or the image-text combination generation mode, and starting the initial document information extracting module.
And the initial document information extraction module is used for extracting key information of the initial document corresponding to the current single-image generation mode or the image-text combination generation mode.
And the initial document arrangement module is used for extracting the number of paragraphs from the key information, calculating the matching degree of the pictures and the paragraphs, and arranging the initial document to obtain a final document of the current single-picture generation mode or the picture-text combination generation mode.
And the final document checking module is used for extracting document checking information of the user and calculating final document generation accuracy.
And the platform document generation evaluation module is used for carrying out generation deviation evaluation on the document generation platform to obtain a document generation deviation trend evaluation index.
And the early warning terminal is used for early warning when the document generation deviation trend evaluation index is lower than a set value.
And the cloud database is used for storing the keywords of each associated picture corresponding to each document template and storing the checking information of each historical generated document in the single-text generation mode, the single-picture generation mode and the picture-text combination generation mode.
Specifically, the specific confirmation process of the initial document confirming the single-image generation mode or the image-text combination generation mode is as follows: and extracting uploaded pictures from the data to be edited uploaded by the current user, extracting picture elements from the pictures, and taking the picture elements as picture keywords.
And matching and comparing each picture keyword with each associated picture keyword corresponding to each document template stored in the cloud database, and taking the document template as an initial document template of a single-picture generation mode if each picture keyword is successfully matched with each associated picture keyword corresponding to a certain document template stored in the cloud database.
And extracting the uploaded characters or voices from the data to be edited uploaded by the current user, and generating an initial document of the image-text combination generation mode in the same way according to the generation mode of the final document of the single-text generation mode.
Specifically, the key information includes the number of paragraphs, the total number of fonts for each paragraph, paragraph keywords, the number of fonts for paragraph keywords, the number of occurrences of paragraph keywords, and the number of occurrences of each keyword.
Specifically, the calculating the matching degree of the picture and each paragraph includes calculating the matching degree of the picture and each paragraph in the initial document in a single picture generation mode and the matching degree of the picture and each paragraph in the initial document in a picture-text combination generation mode, wherein the two calculation modes are the same mode, and the specific calculation process of the matching degree of the picture and each paragraph in the initial document in the single picture generation mode is as follows: a1, extracting paragraph keywords corresponding to all paragraphs from the key information of the initial document.
A2, comparing the paragraph keywords corresponding to the paragraphs, and if the paragraph keywords corresponding to the paragraphs are different from each other, extracting each picture element corresponding to the picture in the data to be edited and taking the picture element as each picture keyword.
A3, constructing paragraph keyword vectors corresponding to all paragraphs and word vectors corresponding to all picture keywords, and respectively marking as beta i and χj Where i denotes the number of each paragraph, i=1, 2,..n, j denotes the number of each picture, j=1, 2,..m.
A4, calculating the matching degree phi of the picture under the key words of different paragraphs and each paragraph in the initial document i ，
A5, if the paragraph keywords corresponding to the paragraphs have the same paragraph keywords, calculating the matching degree phi 'of the picture under the same paragraph keywords and the paragraphs in the initial document' i Thereby obtaining the matching degree delta of each paragraph in the picture and the initial document in the single-picture generation mode, wherein the delta takes the value phi i or φ′i 。
Specifically, the matching degree of the picture under the same paragraph keyword and each paragraph in the initial document is calculated by the following steps: b1, extracting the total number of fonts of each paragraph, the number of fonts of the paragraph keywords, the occurrence number of the paragraph keywords and the occurrence number of each keyword from the key information of the initial document.
B2, counting the number of times of occurrence of the keywords of the corresponding paragraphs of each paragraph and the number of times of occurrence of the keywords of the reference, and respectively marking as epsilon i and ε′i 。
B3, counting the word number ratio of the keywords of the corresponding paragraphs of each paragraph, and marking as eta i 。
B4, calculating the matching degree phi 'of the picture under the same paragraph keyword and each paragraph in the initial document' i ，
wherein ,and eta' respectively represent the keyword matching degree and word number ratio of the set reference, a 1 、a 2 and a3 Respectively representing the set matching degree of the keywords, the number of occurrences of the keywords and the duty ratio weight of the word number duty ratio corresponding to the matching degree of each paragraph in the initial document, lambda 1 And representing the matching degree correction factors of the set pictures under the same paragraph keywords and all paragraphs in the initial document.
Specifically, the calculation process of the final document generation accuracy is as follows: and C1, counting the number of wrong words, wrongly written characters and wrong symbols in the final document from the document inspection information of the user when the final document is in a single-text generation mode.
C2, calculating the final document generation precision psi of the current single document generation mode 1 。
And C3, counting the number of miswords, wrongly written characters, wrong symbols and picture position non-coincidence positions in the final document from the document inspection information of the user when the final document is in the single-picture generation mode.
C4, calculating the final document generation precision psi of the current single-image generation mode 2 。
And C5, when the final document is in the image-text combination generation mode, the final document generation accuracy in the image-text combination generation mode is obtained by the same calculation according to the calculation mode of the final document generation accuracy in the single image generation mode.
Specifically, the final document generation accuracy of the current document generation mode is calculated by the following steps: d1, respectively recording the number of the wrong words, the wrongly written words and the wrong symbols in the final document asθ 0 and ρ0 。
D2, calculating the final document generation precision psi of the current single document generation mode 1 ，
wherein ,/>θ′ 0 and ρ′0 Respectively representing the number of error words, written words and error symbols of the set reference, a 4 、a 5 and a6 Respectively representing the set error words, the number of the individual words and the number of the error symbols corresponding to the final document generation precision duty ratio weight lambda 2 The document corresponding to the set single text generation mode generates the accurate evaluation correction factor.
Specifically, the document generates a bias trend evaluation index, and the specific analysis process is as follows: and E1, extracting the inspection information of each generated document in the single-text generation mode from the cloud database.
E2, performing the same calculation according to the calculation mode of the final document generation accuracy of the current single document generation mode to obtain the accuracy of each historical generated document in the single document generation mode;
e3, sequencing the final document of the current single-document generation mode and each historical generation document in the single-document generation mode according to time sequence to obtain the sequence of each generation document corresponding to the single-document generation mode, taking the sequence as the generation sequence of each generation document, and extracting the generation accuracy of each generation document;
e4, constructing a document generation precision growth curve corresponding to the single text generation mode by taking the generation sequence as an abscissa and the precision as an ordinate, positioning a slope value from the curve, and marking the slope value as a precision growth rate corresponding to the single text generation mode as K 1 ；
E5, obtaining the precision increment rate corresponding to the single-image generation mode and the image-text combination generation mode by performing similar analysis according to the analysis mode of the precision increment rate corresponding to the single-text generation mode, and marking the precision increment rate as K respectively 2 and K3 ；
E6, calculating a document generation deviation trend evaluation index zeta,
wherein ,ω1 、ω 2 and ω3 Respectively represent the corresponding precision increment rate correction factors, K 'under Shan Wen, single-image and image-text combination generation modes' 1 、K′ 2 and K′3 Respectively representing the corresponding precision increment rate under the set reference Shan Wen, single diagram and image-text combination generation mode, b 1 、b 2 and b3 The set Shan Wen, single diagram and image-text combination generation modes are respectively represented, and the corresponding document generation deviation trend evaluation duty ratio weight is generated.
Specifically, the Shan Wen, single-image and image-text combination generation mode correspond to the same setting mode of the precision growth rate correction factors, wherein the Shan Wen generation mode corresponds to the precision growth rate correction factor in the setting mode that: f1, using a starting point of a precision growth curve corresponding to a single-text generation mode as a base point, using a set reference precision growth rate as a slope, constructing a reference datum line in a document generation precision growth curve corresponding to the single-text generation mode, positioning the generation times below the reference datum line from the document generation precision growth curve corresponding to the single-text generation mode, and marking the generation times as deviation times as M.
And F2, positioning the amplitude of the precision from the file generation precision growth curve corresponding to the Shan Wen generation mode, and marking as H'.
And F3, taking a line segment of a slope in a document generation precision growth curve corresponding to the single text generation mode as a reference line, extracting the precision corresponding to a starting point and an end point in the reference line, and obtaining a precision deviation by difference, and marking as H'.
F4, calculating a correction factor omega of the precision increment rate corresponding to the single text generation mode 1 ，
Deviation representing deviation times of setting reference and amplitude of accuracy and accuracy deviation, b 3 and b4 The set number of deviation times and the duty weight of the precision deviation corresponding to the precision growth rate correction factor are respectively represented.
Compared with the prior art, the embodiment of the invention has at least the following advantages or beneficial effects: (1) According to the invention, through confirming the generation mode and carrying out arrangement analysis and processing on the documents under different generation modes, the final document is generated, the problems of complex process, low efficiency and the like in the current manual editing are effectively solved, the pertinence and the flexibility editing of the document are realized, the accuracy and the suitability of the document editing are improved, the efficiency of the document editing generation is obviously improved, the different document generation requirements of a user are met, the practicability is strong, and the viscosity of the user and a document generation platform is further maintained.
(2) According to the invention, when the matching degree of the picture and each paragraph in the initial document is calculated in the single picture generation mode and the picture-text combination generation mode, the accuracy of the picture insertion position is ensured, the deviation of picture position insertion is reduced, and the effect and quality of document arrangement are improved by analyzing and confirming different paragraph keywords and the same paragraph keywords.
(3) According to the invention, the document generation deviation trend evaluation index is obtained by analyzing the test information of each generated document according to the current and different generation modes combined with histories, the deviation situation of document generation is intuitively displayed, and a reliable decision basis is provided for the subsequent optimization of the document automatic generation platform, so that the error correction effect and the optimization effect of the document automatic generation platform are improved, and the detection efficiency of the document automatic generation platform on the generated deviation is also improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram showing the connection of the system modules according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the present invention provides an intelligent word document editing and generating system, which includes: the system comprises an editing data acquisition module, a generation mode confirmation module, an initial document information extraction module, an initial document arrangement module, a final document inspection module, a platform document generation evaluation module, an early warning terminal and a cloud database.
The editing data acquisition module is connected with the generation mode confirmation module, the generation mode confirmation module is connected with the initial document information extraction module, the final document inspection module and the cloud database, the initial document information extraction module is connected with the initial document arrangement module, the initial document arrangement module is connected with the final document inspection module, the final document inspection module is connected with the platform document generation evaluation module and the cloud database, and the platform document generation evaluation module is connected with the early warning terminal.
The editing data acquisition module is used for acquiring data to be edited uploaded by a current user in the document generation platform.
It should be noted that, the data to be edited is text, voice and picture uploaded to the document generation platform by the current user.
The generation mode confirmation module is used for confirming the current document generation mode according to the data to be edited uploaded by the current user, generating a final document of the single-text generation mode if the current document generation mode is the single-text generation mode, simultaneously starting the final document inspection module, confirming an initial document of the single-text generation mode or the image-text combination generation mode if the current document generation mode is the single-text generation mode or the image-text combination generation mode, and starting the initial document information extraction module.
Note that, the Shan Wen generation mode refers to a single text generation mode when only text or voice exists in the data to be edited uploaded by the current user.
The single-image generation mode refers to a mode that only pictures exist in the data to be edited uploaded by the current user, and the current document generation mode is the single-image generation mode.
The image-text combination generating mode refers to that characters and pictures, voice and pictures or characters, voice and pictures exist in the data to be edited uploaded by the current user, and the current document generating mode is the image-text combination generating mode.
The final document for generating the single text generation mode is a well-established technology, and is not described in detail herein.
In a specific embodiment of the present invention, the specific confirmation process of the initial document confirming the single-image generating mode or the image-text combination generating mode is as follows: and extracting uploaded pictures from the data to be edited uploaded by the current user, extracting picture elements from the pictures, and taking the picture elements as picture keywords.
And matching and comparing each picture keyword with each associated picture keyword corresponding to each document template stored in the cloud database, and taking the document template as an initial document template of a single-picture generation mode if each picture keyword is successfully matched with each associated picture keyword corresponding to a certain document template stored in the cloud database.
And extracting the uploaded characters or voices from the data to be edited uploaded by the current user, and generating an initial document of the image-text combination generation mode in the same way according to the generation mode of the final document of the single-text generation mode.
It should be noted that the picture elements include, but are not limited to, characters, animals, buildings, and landscapes.
And the initial document information extraction module is used for extracting key information of the initial document corresponding to the current single-image generation mode or the image-text combination generation mode.
In a specific embodiment of the present invention, the key information includes a number of paragraphs, a total number of fonts of each paragraph, a paragraph keyword, a number of fonts of a paragraph keyword, a number of occurrences of a paragraph keyword, and a number of occurrences of each keyword.
Note that, the paragraph keywords refer to keywords having the largest occurrence number in each paragraph.
The initial document arranging module is used for extracting the number of paragraphs from the key information, calculating the matching degree of the pictures and the paragraphs, and arranging the initial document to obtain a final document of a current single-picture generation mode or a picture-text combination generation mode.
In a specific embodiment of the present invention, the calculating the matching degree between the picture and each paragraph includes calculating the matching degree between the picture and each paragraph in the initial document in the single-picture generating mode and the matching degree between the picture and each paragraph in the initial document in the image-text combining generating mode, where the calculating modes are the same, and the specific calculating process of the matching degree between the picture and each paragraph in the initial document in the single-picture generating mode includes: a1, extracting paragraph keywords corresponding to all paragraphs from the key information of the initial document.
A2, comparing the paragraph keywords corresponding to the paragraphs, and if the paragraph keywords corresponding to the paragraphs are different from each other, extracting each picture element corresponding to the picture in the data to be edited and taking the picture element as each picture keyword.
A3, constructing paragraph keyword vectors and picture keywords corresponding to all the paragraphsCorresponding word vectors and are respectively marked as beta i and χj Where i denotes the number of each paragraph, i=1, 2,..n, j denotes the number of each picture, j=1, 2,..m.
The beta is that i and χj The word frequency and the inverse document frequency are calculated, and word vectors are constructed according to the word frequency and the inverse document frequency, wherein the calculation mode of the word frequency and the inverse document frequency is the existing mature technology, and the description is omitted here.
A4, calculating the matching degree phi of the picture under the key words of different paragraphs and each paragraph in the initial document i ，
A5, if the paragraph keywords corresponding to the paragraphs have the same paragraph keywords, calculating the matching degree phi 'of the picture under the same paragraph keywords and the paragraphs in the initial document' i Thereby obtaining the matching degree delta of each paragraph in the picture and the initial document in the single-picture generation mode, wherein the delta takes the value phi i or φ′i 。
In a specific embodiment of the present invention, the matching degree between the picture under the same paragraph keyword and each paragraph in the initial document is specifically calculated as follows: b1, extracting the total number of fonts of each paragraph, the number of fonts of the paragraph keywords, the occurrence number of the paragraph keywords and the occurrence number of each keyword from the key information of the initial document.
B2, counting the number of times of occurrence of the keywords of the corresponding paragraphs of each paragraph and the number of times of occurrence of the keywords of the reference, and respectively marking as epsilon i and ε′i 。
It should be noted that, the statistics process of the reference occurrence frequency ratio of the paragraph keywords corresponding to each paragraph is as follows: according to the occurrence times of the keywords, the occurrence times of the keywords of each paragraph in other paragraphs are screened out from the occurrence times, so that the corresponding occurrence times of the keywords of each paragraph in other paragraphs are counted, the maximum occurrence times are screened out from the occurrence times, and the maximum occurrence times are used as the reference occurrence times of the keywords of the corresponding paragraphs of each paragraph.
The statistical formula of the occurrence frequency of the paragraph keywords corresponding to each paragraph is as follows:
b3, counting the word number ratio of the keywords of the corresponding paragraphs of each paragraph, and marking as eta i ；
It should be noted that the word number of the paragraph keyword corresponding to each paragraph is calculated as follows
B4, calculating the matching degree phi 'of the picture under the same paragraph keyword and each paragraph in the initial document' i ，
wherein ,and eta' respectively represent the keyword matching degree and word number ratio of the set reference, a 1 、a 2 and a3 Respectively representing the set matching degree of the keywords, the number of occurrences of the keywords and the duty ratio weight of the word number duty ratio corresponding to the matching degree of each paragraph in the initial document, lambda 1 And representing the matching degree correction factors of the set pictures under the same paragraph keywords and all paragraphs in the initial document.
It should be noted that, the editing of the initial document refers to taking the paragraph with the highest matching degree with the picture as the insertion paragraph of the target picture, and inserting the picture into the end position of the target insertion paragraph.
According to the embodiment of the invention, the final document is generated by confirming the generation mode and carrying out arrangement analysis and processing on the documents in different generation modes, so that the problems of complex process, low efficiency and the like in the current manual editing are effectively solved, the pertinence and the flexibility of the document editing are realized, the accuracy and the suitability of the document editing are improved, the efficiency of the document editing generation is obviously improved, the different document generation requirements of a user are met, the practicability is high, and the viscosity of the user and a document generation platform is further maintained.
The final document inspection module is used for extracting document inspection information of a user and calculating final document generation accuracy.
In a specific embodiment of the present invention, the calculation process of the final document generation accuracy is: and C1, counting the number of wrong words, wrongly written characters and wrong symbols in the final document from the document inspection information of the user when the final document is in a single-text generation mode.
C2, calculating the final document generation precision psi of the current single document generation mode 1 。
In a specific embodiment of the present invention, the final document generation accuracy of the current document generation mode is specifically calculated as follows: d1, respectively recording the number of the wrong words, the wrongly written words and the wrong symbols in the final document asθ 0 and ρ0 。
D2, calculating the final document generation precision psi of the current single document generation mode 1 ， wherein ,/>θ′ 0 and ρ′0 Respectively representing the number of error words, written words and error symbols of the set reference, a 4 、a 5 and a6 Respectively representing the set error words, the number of the individual words and the number of the error symbols corresponding to the final document generation precision duty ratio weight lambda 2 The document corresponding to the set single text generation mode generates the accurate evaluation correction factor.
When the final document is in the single-image generation mode, counting the number of miswords, wrongly written characters, wrong symbols and non-conforming positions of the image in the final document from document inspection information of a user;
c4, calculating the final document generation precision psi of the current single-image generation mode 2 ；
The final document generation accuracy of the current single-image generation mode is described, and the specific calculation process is as follows: g1, respectively marking the number of miswords, wrongly written characters, wrong symbols and positions of pictures in the final document as the number of non-conforming positionsθ 1 、ρ 1 and τ1 。
G2, calculating the final document generation precision psi of the current single-image generation mode 2 ， wherein ,/>θ′ 1 、ρ′ 1 and τ′1 Respectively representing the number of places where the positions of the error word, the written word, the error symbol and the picture of the set reference are not matched, a 7 、a 8 、a 9 and a10 Respectively representing the set error word, the set character, the set error symbol and the set final document generation precision duty ratio weight lambda corresponding to the number of the positions of the picture which are not matched 3 The set single-graph generation mode is represented to correspond to a document generation precision evaluation correction factor, and e represents a natural constant.
And C5, when the final document is in the image-text combination generation mode, the final document generation accuracy in the image-text combination generation mode is obtained by the same calculation according to the calculation mode of the final document generation accuracy in the single image generation mode.
According to the embodiment of the invention, when the matching degree of the picture and each paragraph in the initial document is calculated in the single picture generation mode and the picture-text combination generation mode, the keywords of different paragraphs and the keywords of the same paragraph are analyzed and confirmed, so that the accuracy of the picture inserting position is ensured, the deviation of picture position inserting is reduced, and the effect and quality of document arrangement are improved.
The platform document generation evaluation module is used for carrying out generation deviation evaluation on the document generation platform to obtain a document generation deviation trend evaluation index.
In a specific embodiment of the invention, the document generates a bias trend evaluation index, and the specific analysis process is as follows: and E1, extracting the inspection information of each generated document in the single-text generation mode from the cloud database.
And E2, performing the same calculation according to the calculation mode of the final document generation accuracy of the current single document generation mode to obtain the accuracy of each generated document in the historical single document generation mode.
And E3, performing the same calculation according to the calculation mode of the final document generation accuracy of the current single document generation mode to obtain the accuracy of each historical generated document in the single document generation mode.
And E4, sequencing the final document of the current single-document generation mode and each historical generation document in the single-document generation mode according to time sequence, obtaining the sequence of each generation document corresponding to the single-document generation mode, taking the sequence as the generation sequence of each generation document, and simultaneously extracting the generation accuracy of each generation document.
E5, constructing a document generation precision growth curve corresponding to the single text generation mode by taking the generation sequence as an abscissa and the precision as an ordinate, positioning a slope value from the curve, and marking the slope value as a precision growth rate corresponding to the single text generation mode as K 1 。
E6, obtaining the precision increment rate corresponding to the single-image generation mode and the image-text combination generation mode by performing similar analysis according to the analysis mode of the precision increment rate corresponding to the single-text generation mode, and marking the precision increment rate as K respectively 2 and K3 。
E7, calculating a document generation deviation trend evaluation index zeta,
wherein ,ω1 、ω 2 and ω3 Respectively represent the corresponding precision increment rate correction factors, K 'under Shan Wen, single-image and image-text combination generation modes' 1 、K′ 2 and K′3 Respectively represent corresponding precision increasing rate under Shan Wen, single diagram and image-text combination generating mode of setting reference，b 1 、b 2 and b3 The set Shan Wen, single diagram and image-text combination generation modes are respectively represented, and the corresponding document generation deviation trend evaluation duty ratio weight is generated.
In a specific embodiment of the present invention, the corresponding precision growth rate correction factors in the Shan Wen, single-image and image-text combination generating mode are the same setting mode, wherein the setting mode of the precision growth rate correction factor corresponding to the Shan Wen generating mode is as follows: f1, using a starting point of a precision growth curve corresponding to a single-text generation mode as a base point, using a set reference precision growth rate as a slope, constructing a reference datum line in a document generation precision growth curve corresponding to the single-text generation mode, positioning the generation times below the reference datum line from the document generation precision growth curve corresponding to the single-text generation mode, and marking the generation times as deviation times as M.
And F2, positioning the amplitude of the precision from the file generation precision growth curve corresponding to the Shan Wen generation mode, and marking as H'.
And F3, taking a line segment of a slope in a document generation precision growth curve corresponding to the single text generation mode as a reference line, extracting the precision corresponding to a starting point and an end point in the reference line, and obtaining a precision deviation by difference, and marking as H'.
F4, calculating a correction factor omega of the precision increment rate corresponding to the single text generation mode 1 ，Wherein M 'and ΔH' represent the deviation of the set reference deviation times and the deviation of the amplitude and the accuracy of the accuracy, b 3 and b4 The set number of deviation times and the duty weight of the precision deviation corresponding to the precision growth rate correction factor are respectively represented.
According to the embodiment of the invention, the document generation deviation trend evaluation index is obtained by analyzing the test information of each generated document according to the current and different generation modes combined with histories, the deviation situation of document generation is intuitively displayed, and a reliable decision basis is provided for the subsequent optimization of the document automatic generation platform, so that the error correction effect and the optimization effect of the document automatic generation platform are improved, and the detection efficiency of the document automatic generation platform on the generated deviation is also improved.
And the early warning terminal is used for early warning when the document generation deviation trend evaluation index is lower than a set value.
And the cloud database is used for storing the keywords of each associated picture corresponding to each document template and storing the checking information of each historical generated document in the single-text generation mode, the single-picture generation mode and the picture-text combination generation mode.
The foregoing is merely illustrative and explanatory of the principles of this invention, as various modifications and additions may be made to the specific embodiments described, or similar arrangements may be substituted by those skilled in the art, without departing from the principles of this invention or beyond the scope of this invention as defined in the claims.
Claims (9)
1. An intelligent word document editing and generating system, comprising:
the editing data acquisition module is used for acquiring data to be edited uploaded by a current user in the document generation platform;
the generation mode confirmation module is used for confirming a current document generation mode according to the data to be edited uploaded by the current user, generating a final document of the single-text generation mode if the current document generation mode is the single-text generation mode, starting the final document inspection module at the same time, confirming an initial document of the single-text generation mode or the image-text combination generation mode if the current document generation mode is the single-text generation mode or the image-text combination generation mode, and starting the initial document information extraction module;
the initial document information extraction module is used for extracting key information of the initial document corresponding to the current single-image generation mode or the image-text combination generation mode;
the initial document arrangement module is used for extracting the number of paragraphs from the key information, calculating the matching degree of the pictures and the paragraphs, and arranging the initial document to obtain a final document of a current single-picture generation mode or a picture-text combination generation mode;
the final document inspection module is used for extracting document inspection information of a user and calculating final document generation accuracy;
the platform document generation evaluation module is used for carrying out generation deviation evaluation on the document generation platform to obtain a document generation deviation trend evaluation index;
the early warning terminal is used for early warning when the document generation deviation trend evaluation index is lower than a set value;
and the cloud database is used for storing the keywords of each associated picture corresponding to each document template and storing the checking information of each historical generated document in the single-text generation mode, the single-picture generation mode and the picture-text combination generation mode.
2. The intelligent word document edit creation system of claim 1, wherein: the specific confirmation process of the initial document for confirming the single-image generation mode or the image-text combination generation mode comprises the following steps:
extracting uploaded pictures from the data to be edited uploaded by the current user, extracting picture elements from the pictures, and taking the picture elements as picture keywords;
matching and comparing each picture keyword with each associated picture keyword corresponding to each document template stored in the cloud database, and taking the document template as an initial document template of a single-picture generation mode if each picture keyword is successfully matched with each associated picture keyword corresponding to a certain document template stored in the cloud database;
and extracting the uploaded characters or voices from the data to be edited uploaded by the current user, and generating an initial document of the image-text combination generation mode in the same way according to the generation mode of the final document of the single-text generation mode.
3. The intelligent word document edit creation system of claim 1, wherein: the key information includes the number of paragraphs, the total number of fonts of each paragraph, the paragraph keywords, the number of fonts of the paragraph keywords, the number of occurrences of the paragraph keywords, and the number of occurrences of each keyword.
4. The intelligent word document edit creation system according to claim 3, wherein: the calculating the matching degree of the picture and each paragraph comprises calculating the matching degree of the picture and each paragraph in the initial document in a single picture generation mode and the matching degree of the picture and each paragraph in the initial document in a picture-text combination generation mode, wherein the calculating modes are the same, and the specific calculating process of the matching degree of the picture and each paragraph in the initial document in the single picture generation mode comprises the following steps:
a1, extracting paragraph keywords corresponding to all paragraphs from the key information of the initial document;
a2, comparing the paragraph keywords corresponding to the paragraphs, and if the paragraph keywords corresponding to the paragraphs are different from each other, extracting each picture element corresponding to the picture in the data to be edited and taking the picture element as each picture keyword;
a3, constructing paragraph keyword vectors corresponding to all paragraphs and word vectors corresponding to all picture keywords, and respectively marking as beta i and χj Where i represents the number of each paragraph, i=1, 2,..n, j represents the number of each picture, j=1, 2,..m;
a4, calculating the matching degree phi of the picture under the key words of different paragraphs and each paragraph in the initial document i ，
A5, if the paragraph keywords corresponding to the paragraphs have the same paragraph keywords, calculating the matching degree phi of the picture under the same paragraph keywords and the paragraphs in the initial document i ' obtaining the matching degree delta of each paragraph in the picture and the initial document in the single-picture generation mode, wherein the delta takes the value phi i or φi ′。
5. The intelligent word document editing and generation system of claim 4, wherein: the matching degree of the picture under the same paragraph keyword and each paragraph in the initial document is specifically calculated as follows:
b1, extracting the total number of fonts of each paragraph, the number of fonts of the paragraph keywords, the occurrence number of the paragraph keywords and the occurrence number of each keyword from the key information of the initial document;
b2, counting the number of times of occurrence of the keywords of the corresponding paragraphs of each paragraph and the number of times of occurrence of the keywords of the reference, and respectively marking as epsilon i and εi ′；
B3, counting the word number ratio of the keywords of the corresponding paragraphs of each paragraph, and marking as eta i ；
B4, calculating the matching degree phi of the picture under the same paragraph keyword and each paragraph in the initial document i ′，
wherein ,and eta' respectively represent the keyword matching degree and word number ratio of the set reference, a 1 、a 2 and a3 Respectively representing the set matching degree of the keywords, the number of occurrences of the keywords and the duty ratio weight of the word number duty ratio corresponding to the matching degree of each paragraph in the initial document, lambda 1 And representing the matching degree correction factors of the set pictures under the same paragraph keywords and all paragraphs in the initial document.
6. The intelligent word document edit creation system of claim 1, wherein: the calculation process of the final document generation accuracy comprises the following steps:
c1, counting the number of wrong words, written words and wrong symbols in the final document from document inspection information of a user when the final document is in a single-text generation mode;
c2, calculating the final document generation precision psi of the current single document generation mode 1 ；
When the final document is in the single-image generation mode, counting the number of miswords, wrongly written characters, wrong symbols and non-conforming positions of the image in the final document from document inspection information of a user;
c4, calculating the final document generation precision psi of the current single-image generation mode 2 ；
And C5, when the final document is in the image-text combination generation mode, the final document generation accuracy in the image-text combination generation mode is obtained by the same calculation according to the calculation mode of the final document generation accuracy in the single image generation mode.
7. The intelligent word document edit creation system according to claim 6, wherein: the final document generation accuracy of the current single document generation mode comprises the following specific calculation processes:
d1, respectively recording the number of the wrong words, the wrongly written words and the wrong symbols in the final document asθ 0 and ρ0 ；
D2, calculating the final document generation precision psi of the current single document generation mode 1 ， wherein ,/>θ′ 0 and ρ′0 Respectively representing the number of error words, written words and error symbols of the set reference, a 4 、a 5 and a6 Respectively representing the set error words, the number of the individual words and the number of the error symbols corresponding to the final document generation precision duty ratio weight lambda 2 The document corresponding to the set single text generation mode generates the accurate evaluation correction factor.
8. The intelligent word document edit creation system of claim 1, wherein: the document generates a deviation trend evaluation index, and the specific analysis process is as follows:
e1, extracting test information of each generated document in a single-text generation mode from a cloud database;
e2, performing the same calculation according to the calculation mode of the final document generation accuracy of the current single document generation mode to obtain the accuracy of each historical generated document in the single document generation mode;
e3, sequencing the final document of the current single-document generation mode and each historical generation document in the single-document generation mode according to time sequence to obtain the sequence of each generation document corresponding to the single-document generation mode, taking the sequence as the generation sequence of each generation document, and extracting the generation accuracy of each generation document;
e4, constructing a document generation precision growth curve corresponding to the single text generation mode by taking the generation sequence as an abscissa and the precision as an ordinate, positioning a slope value from the curve, and marking the slope value as a precision growth rate corresponding to the single text generation mode as K 1 ；
E5, obtaining the precision increment rate corresponding to the single-image generation mode and the image-text combination generation mode by performing similar analysis according to the analysis mode of the precision increment rate corresponding to the single-text generation mode, and marking the precision increment rate as K respectively 2 and K3 ；
E6, calculating a document generation deviation trend evaluation index zeta,
wherein ,ω1 、ω 2 and ω3 Respectively represent the corresponding precision increment rate correction factors, K 'under Shan Wen, single-image and image-text combination generation modes' 1 、K′ 2 and K′3 Respectively representing the corresponding precision increment rate under the set reference Shan Wen, single diagram and image-text combination generation mode, b 1 、b 2 and b3 The set Shan Wen, single diagram and image-text combination generation modes are respectively represented, and the corresponding document generation deviation trend evaluation duty ratio weight is generated.
9. The intelligent word document edit creation system of claim 8, wherein: the Shan Wen, single diagram and image-text combination generation mode are all of the same setting mode, wherein the Shan Wen generation mode corresponds to the setting process of the precision growth rate correction factor:
f1, constructing a reference datum line in a document generation precision growth curve corresponding to a single generation mode by taking a starting point of the precision growth curve corresponding to the single generation mode as a base point and taking a set reference precision growth rate as a slope, positioning the generation times below the reference datum line from the document generation precision growth curve corresponding to the single generation mode, and recording the generation times as deviation times as M;
f2, positioning the amplitude of the precision from the document generation precision growth curve corresponding to the Shan Wen generation mode, and marking as H';
f3, taking a line segment of a slope in a document generation precision growth curve corresponding to the single text generation mode as a reference line, extracting precision corresponding to a starting point and a finishing point in the reference line, and taking difference to obtain precision deviation, wherein the precision deviation is marked as H';
f4, calculating a correction factor omega of the precision increment rate corresponding to the single text generation mode 1 ，Wherein M 'and ΔH' represent the deviation of the set reference deviation times and the deviation of the amplitude and the accuracy of the accuracy, b 3 and b4 The set number of deviation times and the duty weight of the precision deviation corresponding to the precision growth rate correction factor are respectively represented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310528361.2A CN116579305A (en) | 2023-05-11 | 2023-05-11 | Intelligent word document editing and generating system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310528361.2A CN116579305A (en) | 2023-05-11 | 2023-05-11 | Intelligent word document editing and generating system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116579305A true CN116579305A (en) | 2023-08-11 |
Family
ID=87539057
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310528361.2A Pending CN116579305A (en) | 2023-05-11 | 2023-05-11 | Intelligent word document editing and generating system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116579305A (en) |
-
2023
- 2023-05-11 CN CN202310528361.2A patent/CN116579305A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109190722B (en) | Font style migration transformation method based on Manchu character picture | |
| CN106847288B (en) | Error correction method and device for voice recognition text | |
| CN1187257A (en) | Method and apparatus for character recognition of handwriting input | |
| CN105654135A (en) | Image character sequence recognition system based on recurrent neural network | |
| CN112528041B (en) | Scheduling term specification verification method based on knowledge graph | |
| CN105654129A (en) | Optical character sequence recognition method | |
| CN110443864A (en) | A kind of characters in a fancy style body automatic generation method based on single phase a small amount of sample learning | |
| US11587275B2 (en) | Method for generating a handwriting vector | |
| CN107993636B (en) | Recursive neural network-based music score modeling and generating method | |
| CN110941720A (en) | Knowledge base-based specific personnel information error correction method | |
| CN112182156B (en) | Aspect-level interpretable deep network scoring prediction recommendation method based on text processing | |
| CN113673228A (en) | Text error correction method, text error correction device, computer storage medium and computer program product | |
| CN110414524B (en) | Character recognition result reasoning and error correction method for aviation cable coding | |
| CN114791958A (en) | Zero sample cross-modal retrieval method based on variational self-encoder | |
| CN115759119A (en) | Financial text emotion analysis method, system, medium and equipment | |
| Lund et al. | Improving optical character recognition through efficient multiple system alignment | |
| CN114547232A (en) | Nested entity identification method and system with low labeling cost | |
| CN111950267B (en) | Text triplet extraction method and device, electronic equipment and storage medium | |
| CN116579305A (en) | Intelligent word document editing and generating system | |
| CN115393875B (en) | MobileNet V3-based staff identification and numbered musical notation conversion method and system | |
| CN113420564B (en) | Hybrid matching-based electric power nameplate semantic structuring method and system | |
| CN112560849B (en) | Neural network algorithm-based grammar segmentation method and system | |
| CN107609006A (en) | A kind of chess game optimization method based on local chronicle research | |
| CN117216228B (en) | Psychological accompanying robot and interaction method thereof | |
| CN111985519B (en) | Text similarity quantification method, equipment and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |