CN113378520B - Text editing method and system - Google Patents
Text editing method and system Download PDFInfo
- Publication number
- CN113378520B CN113378520B CN202110421610.9A CN202110421610A CN113378520B CN 113378520 B CN113378520 B CN 113378520B CN 202110421610 A CN202110421610 A CN 202110421610A CN 113378520 B CN113378520 B CN 113378520B
- Authority
- CN
- China
- Prior art keywords
- rule
- editing
- text
- machine learning
- logic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 83
- 230000002452 interceptive effect Effects 0.000 claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 36
- 238000010801 machine learning Methods 0.000 claims description 149
- 238000012423 maintenance Methods 0.000 claims description 44
- 230000006978 adaptation Effects 0.000 claims description 41
- 238000012986 modification Methods 0.000 claims description 8
- 230000004048 modification Effects 0.000 claims description 8
- 238000012217 deletion Methods 0.000 claims description 7
- 230000037430 deletion Effects 0.000 claims description 7
- 238000007792 addition Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 description 20
- 238000011156 evaluation Methods 0.000 description 15
- 238000012549 training Methods 0.000 description 15
- 230000000694 effects Effects 0.000 description 3
- 230000002787 reinforcement Effects 0.000 description 3
- 238000013526 transfer learning Methods 0.000 description 3
- 230000014509 gene expression Effects 0.000 description 2
- 238000003780 insertion Methods 0.000 description 2
- 230000037431 insertion Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- 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
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/451—Execution arrangements for user interfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Machine Translation (AREA)
Abstract
The invention discloses a text editing system, which comprises an editing rule processing device, a text editing program and an interactive display terminal; the editor starts a text editing system and runs a text editing program; the editor performs interactive editing operation on the interactive display terminal through a text editing program; the text editing program obtains the editing rules through the editing rule processing device and assists editing operations of editing personnel by utilizing the editing rules. In addition, the invention also discloses a text editing method. According to the text editing method and system, in the process of processing texts in batches, an editor is effectively assisted to search the text to be edited, and the editing operation execution method is recommended to the text to be edited, so that the text batch processing efficiency is greatly improved, and the occurrence probability of editing errors is reduced.
Description
Technical Field
The invention relates to the technical field of text editing, in particular to a text editing method and a text editing system.
Background
In the prior art, editors often need to perform batch processing of text in order to achieve a specific goal, and the object to be processed is typically a set of text lines. However, the inventor finds that the text editing system in the prior art cannot effectively assist the editors, and the editors need to perform a great deal of repeated labor in the text batch processing process, so that manpower and material resources are wasted greatly, and the working efficiency of the text editing system is extremely low.
Disclosure of Invention
Based on the above, in order to solve the technical problems in the prior art, a text editing method is specifically provided, which includes:
the editor starts a text editing system, and the text editing system runs a text editing program; the editor performs interactive editing operation on the interactive display terminal through the text editing program;
the text editing program obtains editing rules through an editing rule processing device and assists editing operations of the editing personnel by utilizing the editing rules.
In one embodiment, the editing rule processing means comprises an explicit rule sub-means, an implicit rule sub-means; the editing rules comprise explicit rules and implicit rules; the explicit rule is an editing rule described through a logic rule; the implicit rule is an editing rule learned from the editing operation history by using a machine learning algorithm;
the text editing program presents an interactive interface to the editing personnel through the interactive display terminal, wherein the interactive interface comprises an explicit rule maintenance interface, an implicit rule maintenance interface and a text editing interface, and the editing personnel selects to enter the corresponding interface;
when the editor enters an explicit rule maintenance interface, operating and processing the logic rule through an explicit rule sub-device; when the editor enters an implicit rule maintenance interface, operating and processing a machine learning model through an implicit rule sub-device; when the editor enters a text editing interface, the text editing interface presents the text to be edited, and the editor carries out editing operation on the text to be edited in the text editing interface.
In one embodiment, the explicit rules sub-means comprises a logical rules database; the logic rule database is used for storing logic rules; in the explicit rule maintenance interface, the editor selects a logic rule set used in the current text editing operation process from a logic rule database, or performs addition, deletion and modification operations on logic rules stored in the logic rule database;
the implicit rule sub-device comprises a machine learning model library and a model trainer; the machine learning model library is used for storing and managing machine learning models; in the implicit rule maintenance interface, the editor selects a machine learning model used in the current text editing operation process from a machine learning model library, or trains the machine learning model by using a model trainer and stores the machine learning model into the machine learning model library.
In one embodiment, in the explicit rule maintenance interface, the editor selects a logic rule set used in a current text editing operation from a logic rule database, and specifically includes:
the explicit rule sub-device comprises a logic rule evaluator, wherein the logic rule evaluator evaluates the adaptation degree of the logic rule between the logic rule and the text, and the editor selects the corresponding logic rule according to the adaptation degree of the logic rule;
or the editor selects the corresponding logic rule according to the rule generator, the rule source, the rule generation time and the rule label which are stored in the logic rule database and correspond to the logic rule;
the editor selects one or more logic rules to form a logic rule set;
in the implicit rule maintenance interface, the editor selects a machine learning model used in the current text editing operation process from a machine learning model library, and specifically includes:
the implicit rule sub-device comprises a model evaluator, wherein the model evaluator evaluates the model adaptation degree between a machine learning model and a text, and an editor selects the machine learning model according to the model adaptation degree;
or the editor selects a corresponding machine learning model according to the type of the machine learning model.
In one embodiment, the editing personnel performs editing operation on the text to be edited in the text editing interface, which specifically includes:
the editor selects to edit in an explicit rule auxiliary mode, an implicit rule auxiliary mode and an independent editing mode;
when the editor performs editing operation in the explicit rule auxiliary mode, a logic rule executor acquires a logic rule from the logic rule database, processes a text to be edited by using the logic rule to obtain a text position to be edited, an editing operation execution method recommended by the text applicable logic rule and an expected editing operation execution result, and displays the text position and the text applicable logic rule recommended by the text applicable logic rule through the text editing interface; the editor judges whether to execute the editing operation execution method recommended by the text applicable logic rule according to the display information;
when the editor performs editing operation in the implicit rule auxiliary mode, a model reasoner processes a text to be edited by using a machine learning model to obtain a text position to be edited, a model recommended editing operation execution method and an expected editing operation execution result, the text is displayed through the text editing interface, and the editor judges whether to execute the model recommended editing operation execution method at the text position to be edited according to display information;
when the editor performs editing operation in the independent editing mode, the editor independently edits the text without depending on logic rules or assistance of a machine learning model.
In addition, in order to solve the technical problems in the prior art, a text editing system is specially provided, which comprises an editing rule processing device, a text editing program running on the text editing system and an interactive display terminal;
the editor starts a text editing system, and the text editing system runs a text editing program; the editor performs interactive editing operation on the interactive display terminal through the text editing program;
the text editing program obtains editing rules through the editing rule processing device, and assists editing operations of the editing personnel by utilizing the editing rules.
In one embodiment, the editing rule processing means comprises an explicit rule sub-means, an implicit rule sub-means; the editing rules comprise explicit rules and implicit rules; the explicit rule is an editing rule described through a logic rule; the implicit rule is an editing rule learned from the editing operation history by using a machine learning algorithm;
the text editing program presents an interactive interface to the editing personnel through the interactive display terminal, wherein the interactive interface comprises an explicit rule maintenance interface, an implicit rule maintenance interface and a text editing interface, and the editing personnel selects to enter the corresponding interface;
when the editor enters an explicit rule maintenance interface, operating and processing the logic rule through an explicit rule sub-device; when the editor enters an implicit rule maintenance interface, operating and processing a machine learning model through an implicit rule sub-device; when the editor enters a text editing interface, the text editing interface presents the text to be edited, and the editor carries out editing operation on the text to be edited in the text editing interface.
In one embodiment, the explicit rules sub-means comprises a logical rules database; the logic rule database is used for storing logic rules; in the explicit rule maintenance interface, the editor selects a logic rule set used in the current text editing operation process from a logic rule database, or performs addition, deletion and modification operations on logic rules stored in the logic rule database;
wherein the implicit rule sub-means comprises a machine learning model library and a model trainer; the machine learning model library is used for storing and managing machine learning models; in the implicit rule maintenance interface, the editor selects a machine learning model used in the current text editing operation process from a machine learning model library, or trains the machine learning model by using a model trainer and stores the machine learning model into the machine learning model library.
In one embodiment, the explicit rule sub-device comprises a logic rule evaluator, the logic rule evaluator evaluates the logic rule adaptation degree between the logic rule and the text, and the editor selects a corresponding logic rule according to the logic rule adaptation degree; or the editor selects the corresponding logic rule according to the rule generator, the rule source, the rule generation time and the rule label which are stored in the logic rule database and correspond to the logic rule; the editor selects one or more logic rules to form a logic rule set;
the implicit rule sub-device comprises a model evaluator, wherein the model evaluator evaluates the model adaptation degree between a machine learning model and a text, and an editor selects the machine learning model according to the model adaptation degree; or the editor selects a corresponding machine learning model according to the type of the machine learning model.
In one embodiment, the explicit rule subset further comprises a logical rule executor;
wherein the logic rule executor is connected with the logic rule database; the logic rule executor acquires a logic rule from the logic rule database, and processes the text to be edited by utilizing the logic rule to obtain the text position to be edited, an editing operation execution method recommended by the text applicable logic rule and an expected editing operation execution result;
the implicit rule sub-device also comprises an operation history database and a model reasoner;
the operation history database is used for storing editing operation history data; in the process of editing the text to be edited by the editor, the text editing program records the editing operation history and stores the editing operation history in an operation history database; the model trainer acquires editing operation historical data from the operation historical database, trains the editing operation historical data to obtain a machine learning model and stores the machine learning model into the machine learning model library;
wherein the model reasoner is connected with the machine learning model library; the model inference device acquires a machine learning model from a machine learning model library, processes the text to be edited by using the machine learning model to obtain the position of the text to be edited, recommends an editing operation execution method and expects an editing operation execution result.
The implementation of the embodiment of the invention has the following beneficial effects:
in the process of processing texts in batches, an editor is effectively assisted to search the text position to be edited, and an editing operation execution method is recommended for the text position to be edited, so that the text batch processing efficiency is greatly improved, and the occurrence probability of editing errors is reduced.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Wherein:
FIG. 1 is a schematic diagram of a text editing system 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.
As shown in fig. 1, the present invention discloses a text editing method, which includes:
the editor starts a text editing system, and the text editing system runs a text editing program; the editor performs interactive editing operation on the interactive display terminal through the text editing program;
the text editing program obtains editing rules through an editing rule processing device and assists editing operation of the editing personnel by utilizing the editing rules;
the editing personnel edits the text to be edited according to the editing rule;
particularly, the editors carry out batch processing on the text to be edited;
the text objects to be edited which are processed in batches are word line sets; the editor performs editing operation on the text lines in the text line set to obtain an editing result; editing operations on the text line set include text line insertion, text line deletion, text line modification, and text line movement;
wherein the object of the character line modification is a character set formed by one or more continuous characters in the character line; editing operations for modifying the character set include character insertion, character deletion, character replacement, and character movement;
specifically, the text editing program assists the editing operation of the editing personnel by using the editing rule, and specifically includes:
searching a text position to be edited which needs to be edited; recommending an editing operation execution method for the text position to be edited;
in particular, the editing rule processing means comprises an explicit rule sub-means, an implicit rule sub-means; the editing rules comprise explicit rules and implicit rules; the explicit rule is an editing rule described through a logic rule; the implicit rule cannot be described through a logic rule, and the implicit rule is an editing rule obtained by learning from an editing operation history by using a machine learning algorithm;
wherein the explicit rules include, but are not limited to, keywords, regular expressions, finite state transducers, etc.;
specifically, the implicit rule is an editing rule learned from an editing operation history by using a machine learning algorithm, and specifically includes:
training the editing operation history data to obtain a machine learning model; in the process of training the machine learning model, the training effect of the machine learning model is improved by using methods such as unsupervised learning, transfer learning, reinforcement learning and the like;
particularly, the interactive editing operation of the editor through the text editing program specifically comprises that the text editing program presents an interactive interface to the editor through the interactive display terminal, wherein the interactive interface comprises an explicit rule maintenance interface, an implicit rule maintenance interface and a text editing interface, and the editor selects to enter the corresponding interface;
when the editor enters an explicit rule maintenance interface, operating and processing the logic rule through an explicit rule sub-device; when the editor enters an implicit rule maintenance interface, operating and processing a machine learning model through an implicit rule sub-device;
the explicit rule sub-device comprises a logic rule database, a logic rule evaluator and a logic rule executor;
wherein the implicit rule sub-means comprises an operation history database, a model trainer, a machine learning model library, a model reasoner, and a model evaluator;
specifically, when the editor enters an explicit rule maintenance interface, the logic rule is operated and processed through an explicit rule sub-device, which specifically comprises:
the explicit rules sub-means comprises a logical rules database; the logic rule database is used for storing logic rules;
in the explicit rule maintenance interface, the editor selects a logic rule set used in the current text editing operation process, or performs operations of adding, deleting and modifying logic rules stored in a logic rule database;
in particular, in the explicit rule maintenance interface, the editor selects a logic rule set used in a current text editing operation process from a logic rule database, and specifically includes:
the explicit rule subset includes a logical rule evaluator; the logic rule evaluator evaluates the logic rule adaptation degree between the logic rule and the text; the editors select corresponding logic rules according to the logic rule adaptation degree;
specifically, the logic rule evaluation device calculates the logic rule adaptation degree between the logic rule and the text based on the logic rule adaptation degree parameter and the logic rule evaluation function;
wherein, parameters for evaluating the adaptation degree of the logic rule between the logic rule and the text include, but are not limited to, the number of rule hits, editing distance, etc.;
the logic rule evaluation function may be a preset function, including but not limited to a weighted average function, etc.; or learning from the first supervised dataset by using a machine learning algorithm to obtain the logic rule evaluation function; wherein the first supervised data set is a data set with the adaptation degree of the logic rule manually marked;
or the editor selects the corresponding logic rule according to the rule generator, the rule source, the rule generation time and the rule label which are stored in the logic rule database and correspond to the logic rule;
the editor selects one or more logic rules to form a logic rule set;
the logic rules are summarized and obtained by the editors in the editing operation process, namely the editors are rule generators, texts of the logic rules generated by the editors are rule sources, one or more rule labels are set for the logic rules by the editors, and meanwhile rule generation time is added for the logic rules;
in particular, the explicit rules sub-means comprise a logical rules database; after the logic rule is generated, the explicit rule sub-device correspondingly stores the logic rule, a rule generator, a rule source text, a rule generation time and a rule label to a logic rule database;
in particular, the explicit rule subset comprises a logical rule executor; the logic rule executor is connected with the logic rule database; the logic rule executor acquires a logic rule from the logic rule database, and processes the text to be edited by utilizing the logic rule to obtain a text position to be edited which needs to be edited, an editing operation execution method recommended by the text applicable logic rule and an expected editing operation execution result;
when the editor enters an implicit rule maintenance interface, the machine learning model is operated and processed through an implicit rule sub-device, and the method specifically comprises the following steps:
the implicit rule sub-device comprises a machine learning model library and a model trainer; in the implicit rule maintenance interface, the editor selects a machine learning model used in the current text editing operation process from a machine learning model library, or trains the machine learning model by using a model trainer and stores the machine learning model into the machine learning model library;
in particular, in the implicit rule maintenance interface, the editor selects a machine learning model used in the current text editing operation process from a machine learning model library, and specifically includes:
the implicit rule sub-means includes a model evaluator; the model evaluator evaluates the model adaptation degree between the machine learning model and the text; the editor selects a machine learning model according to the model adaptation degree between the machine learning model and the text;
in particular, the model evaluator calculates the model adaptation degree between the model and the text based on the model adaptation degree parameter and the model evaluation function;
model adaptation parameters for evaluating adaptation between a machine learning model and text include, but are not limited to, entropy values, edit distances, etc.;
the model evaluation function may be a preset function, including but not limited to a weighted average function, etc.; or learning from the second supervised dataset by using a machine learning algorithm to obtain the model evaluation function; wherein the second supervised data set is a data set with model adaptation manually marked;
or the editor selects a corresponding machine learning model according to the type of the machine learning model;
wherein, the machine learning model type includes, but is not limited to, a general model, a personnel personalized model, a text personalized model, a personnel-text personalized model, and the like, and specifically includes:
the universal model is a model trained by utilizing editing operation historical data of a plurality of editors and a plurality of texts;
the personnel personalized model is a model trained by utilizing the editing operation history data of a specific editing personnel;
the text personalized model is a model obtained by training the editing operation history data of a specific text;
the personnel-text personalized model is a model trained by utilizing specific editing personnel and editing operation historical data of the specific editing personnel on specific texts;
wherein the implicit rule sub-means comprises an operation history database; the operation history database is used for storing editing operation history data; in the process of editing the text to be edited by the editor, the text editing program records the editing operation history and stores the editing operation history in an operation history database;
in particular, the operation history data includes an edit text position, an edit operation execution method;
the model trainer acquires editing operation historical data from the operation historical database, trains the editing operation historical data to obtain a machine learning model and stores the machine learning model into the machine learning model library;
in particular, the model trainer improves the training effect of a machine learning model by using methods such as unsupervised learning, transfer learning, reinforcement learning and the like;
wherein the implicit rule subset includes a machine learning model library; the machine learning model library is used for storing and managing the machine learning model obtained through training;
in particular, the machine learning model library manages model version information of the machine learning model, including, but not limited to, version number, training time, training sponsor, training data description information, machine learning model type, etc.;
the model inference device acquires a machine learning model from a machine learning model library, and processes a text to be edited by using the machine learning model to obtain a text position to be edited, a recommended editing operation execution method and an expected editing operation execution result;
when the editor enters a text editing interface, the text editing interface presents a text to be edited, and the editor edits the text to be edited in the text editing interface;
specifically, the editing personnel performs editing operation on the text to be edited in the text editing interface, and specifically includes:
the editor selects to edit in an explicit rule auxiliary mode, an implicit rule auxiliary mode and an independent editing mode;
when the editor performs editing operation in the explicit rule auxiliary mode, a logic rule executor acquires a logic rule from the logic rule database, processes a text to be edited by using the logic rule to obtain a text position to be edited, an editing operation execution method recommended by the text applicable logic rule and an expected editing operation execution result, and displays the text position and the text applicable logic rule recommended by the text applicable logic rule through the text editing interface; the editor judges whether to execute the editing operation execution method recommended by the text applicable logic rule according to the display information;
when the editor performs editing operation in the implicit rule auxiliary mode, a model reasoner acquires a machine learning model from a machine learning model library, processes a text to be edited by using the machine learning model to obtain a text position to be edited, a model recommended editing operation execution method and an expected editing operation execution result, displays the text position and the expected editing operation execution result through the text editing interface, and judges whether to execute the model recommended editing operation execution method at the text position to be edited according to display information by the editor;
when the editor performs editing operation in the independent editing mode, the editor independently edits the text without depending on logic rules or assistance of a machine learning model.
In addition, as shown in fig. 1, the invention also discloses a text editing system, which comprises an editing rule processing device, a text editing program running on the text editing system and an interactive display terminal; the editor starts a text editing system, and the text editing system runs a text editing program;
the editor starts a text editing system, and the text editing system runs a text editing program; the editor performs interactive editing operation on the interactive display terminal through the text editing program;
the text editing program obtains editing rules through the editing rule processing device and assists editing operations of the editing personnel by utilizing the editing rules;
the editing rule processing device comprises an explicit rule sub-device and an implicit rule sub-device; the editing rules comprise explicit rules and implicit rules; the explicit rule is an editing rule described through a logic rule; the implicit rule is an editing rule learned from the editing operation history by using a machine learning algorithm;
wherein the explicit rules comprise keywords, regular expressions and finite state transducers;
the text editing program presents an interactive interface to the editing personnel through the interactive display terminal, wherein the interactive interface comprises an explicit rule maintenance interface, an implicit rule maintenance interface and a text editing interface, and the editing personnel selects to enter the corresponding interface;
when the editor enters an explicit rule maintenance interface, operating and processing the logic rule through an explicit rule sub-device; when the editor enters an implicit rule maintenance interface, operating and processing a machine learning model through an implicit rule sub-device;
the explicit rule sub-device comprises a logic rule database, a logic rule evaluator and a logic rule executor;
the logic rule database is used for storing logic rules; in the explicit rule maintenance interface, the editor selects a logic rule set used in the current text editing operation process from a logic rule database, or performs addition, deletion and modification operations on logic rules stored in the logic rule database;
specifically, the logic rules are summarized and obtained by the editor in the editing operation process, namely the editor is a rule generator, the text of the logic rules generated by the editor is a rule source, the editor sets one or more rule labels for the logic rules, and meanwhile, rule generation time is added for the logic rules;
the logic rule evaluator is used for evaluating the logic rule adaptation degree between the logic rule and the text; the editors select corresponding logic rules according to the logic rule adaptation degree;
the logic rule evaluation device calculates and obtains the logic rule adaptation degree between the logic rule and the text based on the logic rule adaptation degree parameter and the logic rule evaluation function;
in particular, logical rule fitness parameters for evaluating fitness between a logical rule and text include, but are not limited to, number of rule hits, edit distance, etc.;
in particular, the logic rule evaluation function may be a preset function, including but not limited to a weighted average function, etc.; or learning from the first supervised dataset by using a machine learning algorithm to obtain the logic rule evaluation function; wherein the first supervised data set is a data set with the adaptation degree of the logic rule manually marked;
wherein the logic rule executor is connected with the logic rule database; the logic rule executor acquires a logic rule from the logic rule database, and processes the text to be edited by utilizing the logic rule to obtain the text position to be edited, an editing operation execution method recommended by the text applicable logic rule and an expected editing operation execution result;
wherein the implicit rule sub-means comprises an operation history database, a model trainer, a machine learning model library, a model reasoner, and a model evaluator;
in the implicit rule maintenance interface, the editor selects a machine learning model used in the current text editing operation process from a machine learning model library, or trains the machine learning model by using a model trainer;
wherein the implicit rule sub-means comprises an operation history database; the operation history database is used for storing editing operation history data;
in particular, the operation history data includes an edit text position, an edit operation execution method;
the model trainer is connected with the operation history database and the machine learning model library; the model trainer acquires editing operation historical data from the operation historical database, trains the editing operation historical data to obtain a machine learning model and stores the machine learning model into the machine learning model library;
the model evaluator is used for evaluating the model adaptation degree between the machine learning model and the text; selecting a machine learning model by an editor according to the model adaptation degree;
in particular, model adaptation parameters for evaluating model adaptation between a machine learning model and text include, but are not limited to, entropy values, edit distances, and the like;
the model evaluation device calculates and obtains the model adaptation degree between the machine learning model and the text based on the model adaptation degree parameter and the model evaluation function;
the model evaluation function may be a preset function, including but not limited to a weighted average function, etc.; or learning from the second supervised dataset by using a machine learning algorithm to obtain the model evaluation function; wherein the second supervised data set is a data set with model adaptation manually marked;
or the editor selects a corresponding machine learning model according to the type of the machine learning model;
in particular, the machine learning model types include, but are not limited to, generic models, personnel personalization models, text personalization models, personnel-text personalization models, and the like, including in particular:
the universal model is a model trained by utilizing editing operation historical data of a plurality of editors and a plurality of texts;
the personnel personalized model is a model trained by utilizing the editing operation history data of a specific editing personnel;
the text personalized model is a model obtained by training the editing operation history data of a specific text;
the personnel-text personalized model is a model trained by utilizing specific editing personnel and editing operation historical data of the specific editing personnel on specific texts;
in particular, the model trainer improves the training effect of a machine learning model by using methods such as unsupervised learning, transfer learning, reinforcement learning and the like;
the machine learning model library is used for storing and managing the machine learning model obtained through training;
in particular, the machine learning model library manages model version information of the machine learning model, including, but not limited to, version number, training time, training sponsor, training data description information, machine learning model type, etc.;
wherein the model reasoner is connected with the machine learning model library; the model inference device acquires a machine learning model from a machine learning model library, and processes the text to be edited by using the machine learning model to obtain a model inference result and an expected editing operation execution result; the model reasoning result comprises the text position to be edited and the editing operation execution method of model recommendation.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. A text editing method, comprising:
the editor starts a text editing system, and the text editing system runs a text editing program; the editor performs interactive editing operation on the interactive display terminal through the text editing program;
the text editing program obtains editing rules through an editing rule processing device and assists editing operation of the editing personnel by utilizing the editing rules;
the text editing program assists the editing operation of the editing personnel by utilizing the editing rule, and specifically comprises the following steps: searching a text position to be edited which needs to be edited; recommending an editing operation execution method for the text position to be edited;
the editing rule processing device comprises an explicit rule sub-device and an implicit rule sub-device; the editing rules comprise explicit rules and implicit rules; the explicit rule is an editing rule described through a logic rule; the implicit rule is an editing rule learned from the editing operation history by using a machine learning algorithm;
the text editing program presents an interactive interface to the editing personnel through the interactive display terminal, wherein the interactive interface comprises an explicit rule maintenance interface, an implicit rule maintenance interface and a text editing interface, and the editing personnel selects to enter the corresponding interface;
when the editor enters an explicit rule maintenance interface, operating and processing the logic rule through an explicit rule sub-device; when the editor enters an implicit rule maintenance interface, operating and processing a machine learning model through an implicit rule sub-device; when the editor enters a text editing interface, the text editing interface presents the text to be edited, and the editor carries out editing operation on the text to be edited in the text editing interface.
2. The text editing method of claim 1, wherein,
the explicit rules sub-means comprises a logical rules database; the logic rule database is used for storing logic rules; in the explicit rule maintenance interface, the editor selects a logic rule set used in the current text editing operation process from a logic rule database, or performs addition, deletion and modification operations on logic rules stored in the logic rule database;
the implicit rule sub-device comprises a machine learning model library and a model trainer; the machine learning model library is used for storing and managing machine learning models; in the implicit rule maintenance interface, the editor selects a machine learning model used in the current text editing operation process from a machine learning model library, or trains the machine learning model by using a model trainer and stores the machine learning model into the machine learning model library.
3. The text editing method of claim 2, wherein,
in the explicit rule maintenance interface, the editor selects a logic rule set used in the current text editing operation process from a logic rule database, and specifically includes:
the explicit rule sub-device comprises a logic rule evaluator, wherein the logic rule evaluator evaluates the adaptation degree of the logic rule between the logic rule and the text, and the editor selects the corresponding logic rule according to the adaptation degree of the logic rule;
or the editor selects the corresponding logic rule according to the rule generator, the rule source, the rule generation time and the rule label which are stored in the logic rule database and correspond to the logic rule;
the editor selects one or more logic rules to form a logic rule set;
in the implicit rule maintenance interface, the editor selects a machine learning model used in the current text editing operation process from a machine learning model library, and specifically includes:
the implicit rule sub-device comprises a model evaluator, wherein the model evaluator evaluates the model adaptation degree between a machine learning model and a text, and an editor selects the machine learning model according to the model adaptation degree;
or the editor selects a corresponding machine learning model according to the type of the machine learning model.
4. The text editing method of claim 2, wherein,
the editing personnel edit the text to be edited in the text editing interface, and the method specifically comprises the following steps:
the editor selects to edit in an explicit rule auxiliary mode, an implicit rule auxiliary mode and an independent editing mode;
when the editor performs editing operation in the explicit rule auxiliary mode, a logic rule executor acquires a logic rule from the logic rule database, processes a text to be edited by using the logic rule to obtain a text position to be edited, an editing operation execution method recommended by the text applicable logic rule and an expected editing operation execution result, and displays the text position and the text applicable logic rule recommended by the text applicable logic rule through the text editing interface; the editor judges whether to execute the editing operation execution method recommended by the text applicable logic rule according to the display information;
when the editor performs editing operation in the implicit rule auxiliary mode, a model reasoner processes a text to be edited by using a machine learning model to obtain a text position to be edited, a model recommended editing operation execution method and an expected editing operation execution result, the text is displayed through the text editing interface, and the editor judges whether to execute the model recommended editing operation execution method at the text position to be edited according to display information;
when the editor performs editing operation in the independent editing mode, the editor independently edits the text without depending on logic rules or assistance of a machine learning model.
5. The text editing system is characterized by comprising an editing rule processing device, a text editing program running on the text editing system and an interactive display terminal;
the editor starts a text editing system, and the text editing system runs a text editing program; the editor performs interactive editing operation on the interactive display terminal through the text editing program;
the text editing program obtains editing rules through the editing rule processing device and assists editing operations of the editing personnel by utilizing the editing rules;
the text editing program assists the editing operation of the editing personnel by utilizing the editing rule, and specifically comprises the following steps: searching a text position to be edited which needs to be edited; recommending an editing operation execution method for the text position to be edited;
the editing rule processing device comprises an explicit rule sub-device and an implicit rule sub-device; the editing rules comprise explicit rules and implicit rules; the explicit rule is an editing rule described through a logic rule; the implicit rule is an editing rule learned from the editing operation history by using a machine learning algorithm;
the text editing program presents an interactive interface to the editing personnel through the interactive display terminal, wherein the interactive interface comprises an explicit rule maintenance interface, an implicit rule maintenance interface and a text editing interface, and the editing personnel selects to enter the corresponding interface;
when the editor enters an explicit rule maintenance interface, operating and processing the logic rule through an explicit rule sub-device; when the editor enters an implicit rule maintenance interface, operating and processing a machine learning model through an implicit rule sub-device; when the editor enters a text editing interface, the text editing interface presents the text to be edited, and the editor carries out editing operation on the text to be edited in the text editing interface.
6. The text editing system of claim 5, wherein,
wherein the explicit rules sub-means comprises a logical rules database; the logic rule database is used for storing logic rules; in the explicit rule maintenance interface, the editor selects a logic rule set used in the current text editing operation process from a logic rule database, or performs addition, deletion and modification operations on logic rules stored in the logic rule database;
wherein the implicit rule sub-means comprises a machine learning model library and a model trainer; the machine learning model library is used for storing and managing machine learning models; in the implicit rule maintenance interface, the editor selects a machine learning model used in the current text editing operation process from a machine learning model library, or trains the machine learning model by using a model trainer and stores the machine learning model into the machine learning model library.
7. The text editing system of claim 6, wherein,
the explicit rule sub-device comprises a logic rule evaluator, wherein the logic rule evaluator evaluates the adaptation degree of the logic rule between the logic rule and the text, and the editor selects the corresponding logic rule according to the adaptation degree of the logic rule; or the editor selects the corresponding logic rule according to the rule generator, the rule source, the rule generation time and the rule label which are stored in the logic rule database and correspond to the logic rule; the editor selects one or more logic rules to form a logic rule set;
the implicit rule sub-device comprises a model evaluator, wherein the model evaluator evaluates the model adaptation degree between a machine learning model and a text, and an editor selects the machine learning model according to the model adaptation degree; or the editor selects a corresponding machine learning model according to the type of the machine learning model.
8. The text editing system of claim 7, wherein the text editing system,
the explicit rule sub-means further comprises a logic rule executor;
wherein the logic rule executor is connected with the logic rule database; the logic rule executor acquires a logic rule from the logic rule database, and processes the text to be edited by utilizing the logic rule to obtain the text position to be edited, an editing operation execution method recommended by the text applicable logic rule and an expected editing operation execution result;
the implicit rule sub-device also comprises an operation history database and a model reasoner;
the operation history database is used for storing editing operation history data; in the process of editing the text to be edited by the editor, the text editing program records the editing operation history and stores the editing operation history in an operation history database; the model trainer acquires editing operation historical data from the operation historical database, trains the editing operation historical data to obtain a machine learning model and stores the machine learning model into the machine learning model library;
wherein the model reasoner is connected with the machine learning model library; the model inference device acquires a machine learning model from a machine learning model library, processes the text to be edited by using the machine learning model to obtain the position of the text to be edited, recommends an editing operation execution method and expects an editing operation execution result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110421610.9A CN113378520B (en) | 2021-04-20 | 2021-04-20 | Text editing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110421610.9A CN113378520B (en) | 2021-04-20 | 2021-04-20 | Text editing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113378520A CN113378520A (en) | 2021-09-10 |
| CN113378520B true CN113378520B (en) | 2024-04-09 |
Family
ID=77569796
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110421610.9A Active CN113378520B (en) | 2021-04-20 | 2021-04-20 | Text editing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113378520B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH05298083A (en) * | 1992-04-23 | 1993-11-12 | Hokkaido Nippon Denki Software Kk | Text editor |
| US5752058A (en) * | 1995-07-06 | 1998-05-12 | Sun Microsystems, Inc. | System and method for inter-token whitespace representation and textual editing behavior in a program editor |
| JP2002342090A (en) * | 2001-05-16 | 2002-11-29 | Sharp Corp | Text editing device and text editing method |
| CN107480118A (en) * | 2017-08-16 | 2017-12-15 | 科大讯飞股份有限公司 | Method for editing text and device |
| CN108460068A (en) * | 2017-11-28 | 2018-08-28 | 平安科技（深圳）有限公司 | Method, apparatus, storage medium and the terminal that report imports and exports |
| CN111488403A (en) * | 2020-04-03 | 2020-08-04 | 自然资源部第三地形测量队(黑龙江第二测绘工程院) | Data editing method and device, auxiliary editing equipment and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040229199A1 (en) * | 2003-04-16 | 2004-11-18 | Measured Progress, Inc. | Computer-based standardized test administration, scoring and analysis system |
| US7904418B2 (en) * | 2006-11-14 | 2011-03-08 | Microsoft Corporation | On-demand incremental update of data structures using edit list |
| US10031899B2 (en) * | 2014-07-02 | 2018-07-24 | Gracenote Digital Ventures, Llc | Computing device and corresponding method for generating data representing text |
-
2021
- 2021-04-20 CN CN202110421610.9A patent/CN113378520B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH05298083A (en) * | 1992-04-23 | 1993-11-12 | Hokkaido Nippon Denki Software Kk | Text editor |
| US5752058A (en) * | 1995-07-06 | 1998-05-12 | Sun Microsystems, Inc. | System and method for inter-token whitespace representation and textual editing behavior in a program editor |
| JP2002342090A (en) * | 2001-05-16 | 2002-11-29 | Sharp Corp | Text editing device and text editing method |
| CN107480118A (en) * | 2017-08-16 | 2017-12-15 | 科大讯飞股份有限公司 | Method for editing text and device |
| CN108460068A (en) * | 2017-11-28 | 2018-08-28 | 平安科技（深圳）有限公司 | Method, apparatus, storage medium and the terminal that report imports and exports |
| CN111488403A (en) * | 2020-04-03 | 2020-08-04 | 自然资源部第三地形测量队(黑龙江第二测绘工程院) | Data editing method and device, auxiliary editing equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113378520A (en) | 2021-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020182706A1 (en) | Apparatus for the semantic-based optimization of production facilities with explainability | |
| CN112463424A (en) | End-to-end program repair method based on graph | |
| CN113811869A (en) | Translating natural language queries into standard data queries | |
| CN111738807B (en) | Method, computing device, and computer storage medium for recommending target objects | |
| CN111858878A (en) | Method, system and storage medium for automatically extracting answer from natural language text | |
| CN116822579A (en) | Disease classification ICD automatic coding method and device based on contrast learning | |
| JP7287699B2 (en) | Information provision method and device using learning model through machine learning | |
| CN114971530A (en) | Natural language processing-based collaborative operation and maintenance method and device | |
| CN111563140B (en) | Intention identification method and device | |
| US20080113327A1 (en) | Interactive system for teaching and learning algorithms through discovery | |
| CN113378520B (en) | Text editing method and system | |
| CN117151222B (en) | Domain knowledge guided emergency case entity attribute and relation extraction method thereof, electronic equipment and storage medium | |
| CN116955589B (en) | Intelligent proposition method, system, proposition terminal and storage medium based on teaching material knowledge graph | |
| US20240104391A1 (en) | Reward-model based reinforcement learning for performing reasoning tasks | |
| CN112035629B (en) | Method for implementing question-answer model based on symbolized knowledge and neural network | |
| CN115879450B (en) | Gradual text generation method, system, computer equipment and storage medium | |
| CN116433934A (en) | Multi-mode pre-training method for generating CT image representation and image report | |
| CN112988981B (en) | Automatic labeling method based on genetic algorithm | |
| CN111563371B (en) | Text generation method and device | |
| CN115167965A (en) | Transaction progress bar processing method and device | |
| CN115454423A (en) | Static webpage generation method and device, electronic equipment and storage medium | |
| CN111462893B (en) | Chinese medical record auxiliary diagnosis method and system for providing diagnosis basis | |
| CN111460160B (en) | Event clustering method of stream text data based on reinforcement learning | |
| CN114036288A (en) | Relation extraction method based on reinforcement learning | |
| CN118036750A (en) | Training method and system for intelligent task planner with body based on multi-mode large model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |