阅读说明：本技术 提升机器写作质量的方法、系统、计算机设备及存储介质 (Method, system, computer equipment and storage medium for improving writing quality of machine ) 是由 尤莹 卫海天 于 2020-11-27 设计创作，主要内容包括：本申请公开了一种提升机器写作质量的方法、系统、计算机设备及存储介质。方法包括：BERT模型构建步骤：构建连贯性推理器BERT模型,并对所述连贯性推理器BERT模型进行训练；连贯性分数三元组获得步骤：对文章进行分句处理后输入训练后的所述连贯性推理器BERT模型获得连贯性分数,根据所述连贯性分数构建连贯性分数三元组；处理步骤：根据所述连贯性分数三元组构建分割点列表,根据所述分割点列表对所述文章进行处理。本发明充分利用提升机器写作质量的方法的优势,提供语句连贯性的置信度,语句的删除灵活可调整。(The application discloses a method, a system, computer equipment and a storage medium for improving writing quality of a machine. The method comprises the following steps: building a BERT model: constructing a coherence reasoner BERT model and training the coherence reasoner BERT model; a consistency score triple obtaining step: after sentence dividing processing is carried out on the article, inputting the trained coherence reasoner BERT model to obtain a coherence score, and constructing a coherence score triple according to the coherence score; the processing steps are as follows: and constructing a division point list according to the continuity score triple, and processing the article according to the division point list. The invention fully utilizes the advantages of the method for improving the writing quality of the machine, provides the confidence coefficient of sentence continuity, and flexibly and adjustably deletes the sentences.)

1. A method of improving machine writing quality, comprising:

building a BERT model: constructing a coherence reasoner BERT model and training the coherence reasoner BERT model;

a consistency score triple obtaining step: after sentence dividing processing is carried out on the article, inputting the trained coherence reasoner BERT model to obtain a coherence score, and constructing a coherence score triple according to the coherence score;

the processing steps are as follows: and constructing a division point list according to the continuity score triple, and processing the article according to the division point list.

2. The method of improving machine authoring quality as claimed in claim 1, wherein said BERT model building step comprises: and collecting corpora of related specified fields, and training the coherence reasoner BERT model.

3. The method of improving machine authoring quality as claimed in claim 1, wherein said continuity score triplet obtaining step comprises:

sentence dividing step: the method comprises the steps of dividing an input article into sentences to obtain a sentence list of the article, wherein the number of the sentences is T;

arranging: taking a sentence as a unit, and forming T-1 continuous sentence pairs by all sentences by adopting a sliding window;

a consistency score calculation step: inputting the sentence pair into the trained inference device BERT model, and calculating the continuity score of the current sentence pair;

and (3) outputting the consistency fraction triples: and constructing the consistency score triple according to the consistency score.

4. The method of improving machine writing quality of claim 1, wherein the processing step comprises:

and identifying article segmentation points: setting a segmentation point according to the coherence score of the coherence score triplet;

a step of obtaining a division point list: constructing the division point list according to the division points;

article continuity judging step: and judging the continuity of the article according to the segmentation point list and outputting a judgment result.

Sentence calculation and deletion: and when the judgment result shows that the segmentation point list is not empty, the article is divided into a plurality of paragraphs by the segmentation points, the number of sentences contained in each paragraph is calculated according to the subscript of each segmentation point in the segmentation point list, and if the number of sentences in the paragraph is less than or equal to a threshold value, all sentences in the paragraph are deleted.

5. A system for improving machine writing quality, comprising:

the system comprises a BERT model building module, a coherence reasoner BERT model building module and a coherence reasoner BERT model training module, wherein the BERT model building module builds a coherence reasoner BERT model and trains the coherence reasoner BERT model;

the consistency score triple obtaining module is used for performing sentence dividing processing on the article, inputting the article into the trained coherence inference engine BERT model to obtain a consistency score, and constructing a consistency score triple according to the consistency score;

and the processing module constructs a division point list according to the continuity score triples and processes the article according to the division point list.

6. The system for improving machine authoring quality as claimed in claim 5, wherein said BERT model building module collects corpora of related specified domains and trains said coherence reasoner BERT model.

7. The system for improving machine authoring quality of claim 5 wherein said continuity score triplet harvesting module comprises:

the sentence dividing unit is used for dividing the input article to obtain a sentence list of the article, wherein the number of sentences is T;

the arrangement unit takes one sentence as a unit and adopts a sliding window to combine all sentences into T-1 continuous sentence pairs;

a coherence score calculating unit, which inputs the sentence pair into the trained inference engine coherence inference engine BERT model and calculates the coherence score of the current sentence pair;

and the consistency fraction triple output unit constructs the consistency fraction triple according to the consistency fraction.

8. The system for improving machine authoring quality as claimed in claim 5, wherein said processing module comprises:

an article segmentation point identification unit which sets segmentation points according to the coherence score of the coherence score triplets;

a division point list obtaining unit that constructs the division point list according to the division points;

and the article consistency judging unit is used for judging the consistency of the articles according to the segmentation point list and outputting a judgment result.

And a sentence calculation and deletion unit, wherein when the judgment result shows that the division point list is not empty, the article is divided into a plurality of paragraphs by the division points, the number of sentences contained in each paragraph is calculated through subscripts of each division point in the division point list, and if the number of sentences in the paragraph is less than or equal to a threshold value, all sentences in the paragraph are deleted.

9. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor when executing the computer program implements a method of improving machine writing quality as claimed in any one of claims 1 to 4.

10. A storage medium on which a computer program is stored, which program, when executed by a processor, carries out a method of improving machine writing quality as claimed in any one of claims 1 to 4.

Technical Field

The invention belongs to the field of improving machine writing quality, and particularly relates to a method and a system for improving machine writing quality, computer equipment and a storage medium.

Background

Natural language generation is an important research branch in the field of natural language processing, and is widely applied to scenes such as machine translation and text generation (or machine writing). Meanwhile, due to the limitation of the prior art means on natural language semantic mining, the natural language generation model often cannot achieve a satisfactory effect, and typical problems include repeated texts, meaningless texts, incoherent texts and the like, wherein the problem of text incoherence is particularly prominent. Therefore, the incoherent sentences in the generated text need to be identified and removed, so that the machine writing system can output semantically coherent articles under the condition of not wasting machine resources, and the quality of machine writing is improved.

The invention patent of CN2019101477256 discloses a method and a device for training a text recognition model and text consistency recognition, wherein the method for training the text recognition model comprises the following steps: acquiring a first training text and a second training text, wherein the second training text is a reference training text corresponding to the first training text; extracting first training characteristic information from a first training text, and extracting second training characteristic information from a second training text, wherein the first training characteristic information is a text characteristic with a disordered word sequence, and the second training characteristic information is a text characteristic with a coherent word sequence; and training the support vector machine model by using the first training characteristic information and the second training characteristic information to obtain a text recognition model. The text recognition model is formed by extracting the characteristic information of the training text, so that the consistency of the text to be recognized can be recognized quickly, the recognition efficiency of the text consistency is obviously improved, the text consistency can be recognized manually instead, and further, a great deal of manual energy is reduced.

The technology has the defects that training data which are labeled manually needs to be input for training the support vector machine model, and the effect of the text recognition model depends on the standard and quality of manual labeling to a great extent. Due to the reasons of labor cost, labeling standards and the like, the method generally has the following problems: 1. compared with a BERT model pre-trained based on mass data, the accuracy of text consistency identification is low; 2. due to the limited set of annotation data, the text recognition model is less versatile and can normally work only on a specific range of text sets.

Disclosure of Invention

The embodiment of the application provides a method, a system, computer equipment and a storage medium for improving the writing quality of a machine, so as to at least solve the problem of subjective factor influence in the related technology.

The invention provides a method for improving the writing quality of a machine, which comprises the following steps:

building a BERT model: constructing a coherence reasoner BERT model and training the coherence reasoner BERT model;

a consistency score triple obtaining step: after sentence dividing processing is carried out on the article, inputting the trained coherence reasoner BERT model to obtain a coherence score, and constructing a coherence score triple according to the coherence score;

the processing steps are as follows: and constructing a division point list according to the continuity score triple, and processing the article according to the division point list.

The method, wherein the step of building the BERT model comprises: and collecting corpora of related specified fields, and training the coherence reasoner BERT model.

The method, wherein the consistency score triplet obtaining step includes:

sentence dividing step: the method comprises the steps of dividing an input article into sentences to obtain a sentence list of the article, wherein the number of the sentences is T;

arranging: taking a sentence as a unit, and forming T-1 continuous sentence pairs by all sentences by adopting a sliding window;

a consistency score calculation step: inputting the sentence pair into the trained inference device BERT model, and calculating the continuity score of the current sentence pair;

and (3) outputting the consistency fraction triples: and constructing the consistency score triple according to the consistency score.

The method, wherein the processing step comprises:

and identifying article segmentation points: setting a segmentation point according to the coherence score of the coherence score triplet;

a step of obtaining a division point list: constructing the division point list according to the division points;

article continuity judging step: and judging the continuity of the article according to the segmentation point list and outputting a judgment result.

Sentence calculation and deletion: and when the judgment result shows that the segmentation point list is not empty, the article is divided into a plurality of paragraphs by the segmentation points, the number of sentences contained in each paragraph is calculated according to the subscript of each segmentation point in the segmentation point list, and if the number of sentences in the paragraph is less than or equal to a threshold value, all sentences in the paragraph are deleted.

The invention also provides a system for improving the writing quality of a machine, wherein the system comprises:

the system comprises a BERT model building module, a coherence reasoner BERT model building module and a coherence reasoner BERT model training module, wherein the BERT model building module builds a coherence reasoner BERT model and trains the coherence reasoner BERT model;

the consistency score triple obtaining module is used for performing sentence dividing processing on the article, inputting the article into the trained coherence inference engine BERT model to obtain a consistency score, and constructing a consistency score triple according to the consistency score;

and the processing module constructs a division point list according to the continuity score triples and processes the article according to the division point list.

In the system, the BERT model building module collects corpora of the related specified field, and trains the coherence reasoner BERT model.

The system, wherein the consistency score triple obtaining module includes:

the sentence dividing unit is used for dividing the input article to obtain a sentence list of the article, wherein the number of sentences is T;

the arrangement unit takes one sentence as a unit and adopts a sliding window to combine all sentences into T-1 continuous sentence pairs;

a coherence score calculating unit, which inputs the sentence pair into the trained inference engine coherence inference engine BERT model and calculates the coherence score of the current sentence pair;

and the consistency fraction triple output unit constructs the consistency fraction triple according to the consistency fraction.

The system, wherein the processing module comprises:

an article segmentation point identification unit which sets segmentation points according to the coherence score of the coherence score triplets;

a division point list obtaining unit that constructs the division point list according to the division points;

and the article consistency judging unit is used for judging the consistency of the articles according to the segmentation point list and outputting a judgment result.

And a sentence calculation and deletion unit, wherein when the judgment result shows that the division point list is not empty, the article is divided into a plurality of paragraphs by the division points, the number of sentences contained in each paragraph is calculated through subscripts of each division point in the division point list, and if the number of sentences in the paragraph is less than or equal to a threshold value, all sentences in the paragraph are deleted.

The invention also provides a computer device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the method for improving the writing quality of the machine.

The invention also provides a storage medium on which a computer program is stored, which program, when executed by a processor, implements a method of improving machine writing quality as described in any one of the above.

Compared with the prior art, the invention has the following beneficial effects:

1. the method can be used for modifying the defective article generated by machine writing, identifying and deleting the incoherent sentences, thereby not only improving the readability of the machine writing, but also avoiding the waste of computing resources caused by giving up the defective article.

2. The confidence of sentence continuity is provided, and the deletion of the sentence is flexible and adjustable.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:

FIG. 1 is a flow diagram of a method of improving machine authoring quality;

FIG. 2 is a flow chart illustrating the substeps of step S2 in FIG. 1;

FIG. 3 is a flow chart illustrating the substeps of step S3 in FIG. 1;

FIG. 4 is a schematic diagram of a system for improving the writing quality of a machine;

fig. 5 is a block diagram of a computer device according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments provided in the present application without any inventive step are within the scope of protection of the present application.

It is obvious that the drawings in the following description are only examples or embodiments of the present application, and that it is also possible for a person skilled in the art to apply the present application to other similar contexts on the basis of these drawings without inventive effort. Moreover, it should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another.

Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the specification. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of ordinary skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments without conflict.

Unless defined otherwise, technical or scientific terms referred to herein shall have the ordinary meaning as understood by those of ordinary skill in the art to which this application belongs. Reference to "a," "an," "the," and similar words throughout this application are not to be construed as limiting in number, and may refer to the singular or the plural. The present application is directed to the use of the terms "including," "comprising," "having," and any variations thereof, which are intended to cover non-exclusive inclusions; for example, a process, method, system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to the listed steps or elements, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. Reference to "connected," "coupled," and the like in this application is not intended to be limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The term "plurality" as referred to herein means two or more. "and/or" describes an association relationship of associated objects, meaning that three relationships may exist, for example, "A and/or B" may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. Reference herein to the terms "first," "second," "third," and the like, are merely to distinguish similar objects and do not denote a particular ordering for the objects.

The present invention is described in detail with reference to the embodiments shown in the drawings, but it should be understood that these embodiments are not intended to limit the present invention, and those skilled in the art should understand that functional, methodological, or structural equivalents or substitutions made by these embodiments are within the scope of the present invention.

Before describing in detail the various embodiments of the present invention, the core inventive concepts of the present invention are summarized and described in detail by the following several embodiments.

Referring to fig. 1, fig. 1 is a flowchart of a method for improving writing quality of a machine. As shown in fig. 1, a method of improving writing quality of a machine includes:

BERT model construction step S1: constructing a coherence reasoner BERT model and training the coherence reasoner BERT model;

a continuity score triple obtaining step S2: after sentence dividing processing is carried out on the article, inputting the trained coherence reasoner BERT model to obtain a coherence score, and constructing a coherence score triple according to the coherence score;

processing step S3: and constructing a division point list according to the continuity score triple, and processing the article according to the division point list.

Referring to fig. 2, fig. 2 is a flowchart illustrating a sub-step of step S2 in fig. 1. As shown in fig. 2, the continuity score triple obtaining step S2 includes:

sentence dividing step S21: the method comprises the steps of dividing an input article into sentences to obtain a sentence list of the article, wherein the number of the sentences is T;

arranging step S22: taking a sentence as a unit, and forming T-1 continuous sentence pairs by all sentences by adopting a sliding window;

continuity score calculating step S23: inputting the sentence pair into the trained inference device BERT model, and calculating the continuity score of the current sentence pair;

consistency score triplet output step S24: and constructing the consistency score triple according to the consistency score.

Referring to fig. 3, fig. 3 is a flowchart illustrating a sub-step of step S3 in fig. 1. As shown in fig. 3, the processing step S3 includes:

article segmentation point identification step S31: setting a segmentation point according to the coherence score of the coherence score triplet;

dividing point list obtaining step S32: constructing the division point list according to the division points;

article continuity judging step S33: and judging the continuity of the article according to the segmentation point list and outputting a judgment result.

Sentence calculation deletion step S34: and when the judgment result shows that the segmentation point list is not empty, the article is divided into a plurality of paragraphs by the segmentation points, the number of sentences contained in each paragraph is calculated according to the subscript of each segmentation point in the segmentation point list, and if the number of sentences in the paragraph is less than or equal to a threshold value, all sentences in the paragraph are deleted.

The method for improving the writing quality of the machine according to the present invention will be described in detail below with reference to examples.

The first embodiment is as follows:

the invention is a post-step after machine writing, which is used for deleting incoherent sentences in the text and making the readability of the output article stronger. The input of the module is an article generated by machine writing, and the output is an optimized article.

The BERT model is a pre-training model based on a Transformer architecture, and achieves good performance in a plurality of tasks. In order to better learn the association between the semantic information of the text and the context in the article, the BERT model is pre-trained through an MLM task and an NSP task; the latter (i.e., NSP task, Next sequence Prediction) can help us identify the continuity between sentences in the article well. The learning objects for this task are a series of Sentence pairs, half of which are immediately adjacent in the original corpus, i.e., the second Sentence is the Next sequence of the first Sentence, and the other half are randomly drawn pairs of non-immediately adjacent sentences. Through the learning of sentence pairs, for two sentences that are arbitrarily input, the BERT model can infer the probability that the second sentence is the immediate sentence of the first sentence.

The specific execution steps are as follows:

preparing a coherence reasoner BERT model: the BERT model provided by the official is a model file obtained by pre-training on a large amount of general corpora (such as wikipedia and the like), in order to enable the model to be more suitable for specific field scenes, a batch of corpora related to a specific field are collected, the BERT model is trained again (called Fine-Tune or Fine tuning), and the trimmed BERT model is marked as M. Therefore, the inference engine M has the understanding capability of the general natural language and the information of the linguistic data characteristics in the specific field.

The input article is divided into sentences to obtain a sentence list SennceList of the current article, and the number of the sentences is T.

Taking one sentence as a unit, and adopting a sliding window with the size of 2 and the step size of 1, all sentences are formed into T-1 continuous sentence pairs { (S1a, S1b), (S2a, S2b),. }, (ST-1a, ST-1 b).

Inputting each sentence pair (Sia, Sib) into the reasoner M, and calculating the consistency score wi of the current sentence pair, wherein the score is a real number between 0 and 1. The calculation mode of wi is as follows:

wi＝BERT_NSP(Sia,Sib).i＝{0,1,...,T-1}

after this step is completed, we get a series of triplets: (Sia, Sib, wi), i ═ 0,1

Identifying the segmentation points of the article: if the consistency score of a certain sentence pair is less than 0.1 score, the position between two sentences of the sentence pair is a segmentation point; and judging scores of all sentence pairs in sequence to obtain a segmentation point list. If a certain article does not have sentence pairs smaller than 0.1 score, the article is better in consistency, and the list of the segmentation points is empty.

The segmentation point divides an article into a plurality of paragraphs, the number of sentences contained in each paragraph is calculated through subscripts of each segmentation point in a segmentation point list, and if the number of sentences in a certain paragraph is less than 3, all sentences in the paragraph are deleted.

Example two:

referring to fig. 4, fig. 4 is a schematic diagram of a system for improving writing quality of a machine. As shown in fig. 4, the system of the present invention comprises:

the system comprises a BERT model building module, a coherence reasoner BERT model building module and a coherence reasoner BERT model training module, wherein the BERT model building module builds a coherence reasoner BERT model and trains the coherence reasoner BERT model;

the consistency score triple obtaining module is used for performing sentence dividing processing on the article, inputting the article into the trained coherence inference engine BERT model to obtain a consistency score, and constructing a consistency score triple according to the consistency score;

and the processing module constructs a division point list according to the continuity score triples and processes the article according to the division point list.

And the BERT model building module collects corpora of related specified fields and trains the coherence reasoner BERT model.

The consistency score triple obtaining module comprises:

the sentence dividing unit is used for dividing the input article to obtain a sentence list of the article, wherein the number of sentences is T;

the arrangement unit takes one sentence as a unit and adopts a sliding window to combine all sentences into T-1 continuous sentence pairs;

a coherence score calculating unit, which inputs the sentence pair into the trained inference engine coherence inference engine BERT model and calculates the coherence score of the current sentence pair;

and the consistency fraction triple output unit constructs the consistency fraction triple according to the consistency fraction.

The processing module comprises:

an article segmentation point identification unit which sets segmentation points according to the coherence score of the coherence score triplets;

a division point list obtaining unit that constructs the division point list according to the division points;

and the article consistency judging unit is used for judging the consistency of the articles according to the segmentation point list and outputting a judgment result.

And a sentence calculation and deletion unit, wherein when the judgment result shows that the division point list is not empty, the article is divided into a plurality of paragraphs by the division points, the number of sentences contained in each paragraph is calculated through subscripts of each division point in the division point list, and if the number of sentences in the paragraph is less than or equal to a threshold value, all sentences in the paragraph are deleted.

Example three:

referring to fig. 1-3, the present embodiment discloses an embodiment of a computer device. The computer device may comprise a processor 81 and a memory 82 in which computer program instructions are stored.

Specifically, the processor 81 may include a Central Processing Unit (CPU), or A Specific Integrated Circuit (ASIC), or may be configured to implement one or more Integrated circuits of the embodiments of the present Application.

Memory 82 may include, among other things, mass storage for data or instructions. By way of example, and not limitation, memory 82 may include a Hard Disk Drive (Hard Disk Drive, abbreviated to HDD), a floppy Disk Drive, a Solid State Drive (SSD), flash memory, an optical Disk, a magneto-optical Disk, tape, or a Universal Serial Bus (USB) Drive or a combination of two or more of these. Memory 82 may include removable or non-removable (or fixed) media, where appropriate. The memory 82 may be internal or external to the data processing apparatus, where appropriate. In a particular embodiment, the memory 82 is a Non-Volatile (Non-Volatile) memory. In particular embodiments, Memory 82 includes Read-Only Memory (ROM) and Random Access Memory (RAM). The ROM may be mask-programmed ROM, Programmable ROM (PROM), Erasable PROM (EPROM), Electrically Erasable PROM (EEPROM), Electrically rewritable ROM (EAROM), or FLASH Memory (FLASH), or a combination of two or more of these, where appropriate. The RAM may be a Static Random-Access Memory (SRAM) or a Dynamic Random-Access Memory (DRAM), where the DRAM may be a Fast Page Mode Dynamic Random-Access Memory (FPMDRAM), an Extended data output Dynamic Random-Access Memory (eddram), a Synchronous Dynamic Random-Access Memory (SDRAM), and the like.

The memory 82 may be used to store or cache various data files for processing and/or communication use, as well as possible computer program instructions executed by the processor 81.

The processor 81 implements any of the above described embodiments of a method for improving machine writing quality by reading and executing computer program instructions stored in the memory 82.

In some of these embodiments, the computer device may also include a communication interface 83 and a bus 80. As shown in fig. 5, the processor 81, the memory 82, and the communication interface 83 are connected via the bus 80 to complete communication therebetween.

The communication interface 83 is used for implementing communication between modules, devices, units and/or equipment in the embodiment of the present application. The communication port 83 may also be implemented with other components such as: the data communication is carried out among external equipment, image/data acquisition equipment, a database, external storage, an image/data processing workstation and the like.

Bus 80 includes hardware, software, or both to couple the components of the computer device to each other. Bus 80 includes, but is not limited to, at least one of the following: data Bus (Data Bus), Address Bus (Address Bus), Control Bus (Control Bus), Expansion Bus (Expansion Bus), and Local Bus (Local Bus). By way of example, and not limitation, Bus 80 may include an Accelerated Graphics Port (AGP) or other Graphics Bus, an Enhanced Industry Standard Architecture (EISA) Bus, a Front-Side Bus (Front Side Bus), an FSB (FSB), a Hyper Transport (HT) Interconnect, an ISA (ISA) Bus, an Infini Band Interconnect, a Low Pin Count (LPC) Bus, a memory Bus, a microchannel Architecture (MCA) Bus, a PCI (Peripheral Component Interconnect) Bus, a PCI-Express (PCI-X) Bus, a Serial Advanced Technology Attachment (SATA) Bus, a Video Electronics Bus (audio Electronics Association), abbreviated VLB) bus or other suitable bus or a combination of two or more of these. Bus 80 may include one or more buses, where appropriate. Although specific buses are described and shown in the embodiments of the application, any suitable buses or interconnects are contemplated by the application.

The computer device may perform the deletion of the unnecessary sentences in the paragraphs based on a method of improving the quality of the machine writing, thereby implementing the method described in connection with fig. 1-3.

In addition, in combination with the method for improving the writing quality of the machine in the above embodiments, the embodiments of the present application may be implemented by providing a computer-readable storage medium. The computer readable storage medium having stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement any of the above-described embodiments of a method for improving machine writing quality.

The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

In conclusion, the beneficial effects based on the invention are that

1. The method can be used for modifying the defective article generated by machine writing, identifying and deleting the incoherent sentences, thereby not only improving the readability of the machine writing, but also avoiding the waste of computing resources caused by giving up the defective article.

2. The confidence of sentence continuity is provided, and the deletion of the sentence is flexible and adjustable.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.