Standard processing method and device for academic resources

Abstract

The invention discloses a standard processing method, device and equipment for academic resources, relates to the technical field of artificial intelligence, and can solve the standard problem of resource elements in the creation process so as to avoid the standard problem in the academic creation process in advance and improve the academic creation level while ensuring the academic creation efficiency. The method comprises the following steps: acquiring a target academic resource in a creation process, performing element identification in the target academic resource according to a resource type of the target academic resource to obtain a resource element of the target academic resource, performing at least one specification identification on the resource element in the target academic resource to obtain an identification result of the resource element, the identification result of the resource elements comprises whether the application of the resource elements in the target academic resources accords with the specification constraint, and performing specification prompting on the resource elements which do not accord with the specification constraint in the target academic resources according to the identification result of the resource elements.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method and device for standardizing the processing of academic resources. Background Art

[0002] As academic research becomes increasingly international and complex, the importance of academic norms becomes more prominent. Therefore, a deep understanding and strict implementation of academic norms is not only the responsibility of every researcher, but also a need for the common development of the entire academic community. Academic norms are the embodiment of scientific research integrity, the guarantee of academic freedom and creation, the cornerstone of maintaining and enhancing the trust of the academic community, and an important driving force for promoting knowledge discovery and dissemination. Only by fully complying with academic norms can the healthy and sustainable development of academic research be ensured.

[0003] In related technologies, standardized processing of academic resources can be achieved through training programs and paper duplication checks. However, both of these academic standardization methods standardize the creation of completed academic resources. It is difficult to integrate standardization issues into the resource creation process, which affects the improvement of academic creation levels. Summary of the invention

[0004] In view of this, the present application provides a method and device for standardizing the processing of academic resources, the main purpose of which is to solve the problem that it is difficult for existing technologies to integrate standardization issues into the resource creation process, affecting the improvement of academic creation level.

[0005] According to a first aspect of the present application, a method for standardizing processing of academic resources is provided, comprising:

[0006] Acquire targeted academic resources in the creative process;

[0007] Identify elements in the target academic resource according to the resource type of the target academic resource to obtain resource elements of the target academic resource;

[0008] Perform at least one specification identification on the resource elements in the target academic resource to obtain an identification result of the resource element, wherein the identification result of the resource element includes whether the application of the resource element in the target academic resource complies with the specification constraint;

[0009] According to the identification results of the resource elements, standardized prompts are given to the resource elements in the target academic resources that do not meet the standardized constraints.

[0010] Furthermore, the step of performing element identification in the target academic resource according to the resource type of the target academic resource to obtain the resource element of the target academic resource includes:

[0011] According to the resource type of the target academic resource, a target academic resource sample of the resource type corresponding to the target academic resource is obtained from a pre-precipitated resource set, wherein the resource set includes academic resource samples of different resource types, and the academic resource sample has resource elements of the corresponding resource type;

[0012] The target academic resource sample is used as a comparison sample to perform element identification in the target academic resource to obtain the resource elements of the target academic resource.

[0013] Furthermore, before performing at least one specification identification on the resource elements in the target academic resources to obtain the identification results of the resource elements, the method further includes:

[0014] Creating a normative knowledge base of academic resources, wherein the normative knowledge base includes normative constraints formed by processing different academic normative resources;

[0015] Inputting the specification knowledge base and academic resource samples of different resource types into a machine learning model, so as to train the mapping relationship of whether the application of resource elements in the academic resource samples meets the format requirements of the specification constraints through the machine learning model according to the specification constraints;

[0016] When the training process meets the iteration stop condition, a format specification recognition model is generated according to the mapping relationship obtained through training;

[0017] Accordingly, performing at least one standard identification on the resource elements in the target academic resources to obtain the identification result of the resource elements includes:

[0018] The pre-trained format specification recognition model is used to perform format specification recognition on the resource elements in the target academic resources to obtain the format recognition results of the resource elements.

[0019] Furthermore, the pre-trained format specification recognition model is used to perform format specification recognition on resource elements in the target academic resources to obtain format recognition results of the resource elements, including:

[0020] Obtaining text content of resource elements in the target academic resource;

[0021] After converting the text content of the resource element into a format text vector using a pre-trained format specification recognition model, judging whether the format text vector meets the format requirements of the specification constraint according to the constraint specification;

[0022] If so, it is determined that the application of resource elements in the target academic resources complies with the format specification constraints; otherwise, it is determined that the application of resource elements in the target academic resources does not comply with the format specification constraints.

[0023] Furthermore, providing a standardized prompt for the resource elements in the target academic resource that do not meet the standardized constraints according to the identification result of the resource elements includes:

[0024] When the application of the resource element in the target academic resource does not comply with the format specification constraint, locating the first text content that does not comply with the format specification constraint in the target academic resource;

[0025] While providing format specification prompts for the first text content, text content examples with corresponding format requirements for the first text content are displayed. The text content examples are text content extracted from academic resources that conform to format specification constraints.

[0026] Furthermore, before performing at least one specification identification on the resource elements in the target academic resources to obtain the identification results of the resource elements, the method further includes:

[0027] Acquire auxiliary reference resources added historically, wherein the auxiliary reference resources at least include document resources with citation associations to target academic resources in the creation process;

[0028] Inputting the auxiliary reference resources and academic resource samples of different resource types into a machine learning model, so as to train the mapping relationship of whether the application of resource elements in the academic resource samples meets the reference requirements of the specification constraints according to the reference association through the machine learning model;

[0029] When the training process meets the iteration stop condition, a reference specification recognition model is generated according to the mapping relationship obtained through training;

[0030] Accordingly, performing at least one standard identification on the resource elements in the target academic resources to obtain the identification result of the resource elements includes:

[0031] The pre-trained citation specification recognition model is used to perform citation specification recognition on the resource elements in the target academic resources to obtain citation recognition results of the resource elements.

[0032] Furthermore, the citation norm recognition model trained in advance is used to perform citation norm recognition on resource elements in the target academic resources to obtain citation recognition results of the resource elements, including:

[0033] According to the text content of the resource element in the target academic resource, obtain the reference document resource marked by the text content, wherein the reference document resource is a document resource having a reference association with the text content;

[0034] After converting the text content of the resource element into a content text vector using a pre-trained citation specification recognition model, judging whether the content text vector meets the citation requirements of the specification constraints according to the similarity result between the content text vector and the text vector of the cited document resource;

[0035] If so, it is determined that the application of resource elements in the target academic resources complies with the citation standard constraints; otherwise, it is determined that the application of resource elements in the target academic resources does not comply with the citation standard constraints.

[0036] Furthermore, after obtaining the auxiliary reference resources added historically, the method further includes:

[0037] The auxiliary reference resources and academic resource samples of different resource types are input into a machine learning model, so as to train the mapping relationship between the text content of the resource elements in the academic resource samples and the auxiliary reference resources according to the citation association through the machine learning model;

[0038] When the training process meets the iteration stop condition, a reference association model is generated according to the mapping relationship obtained through training;

[0039] Accordingly, providing a standardized prompt for the resource elements in the target academic resource that do not meet the standardized constraints according to the identification result of the resource elements includes:

[0040] When the application of the resource element in the target academic resource does not comply with the citation standard constraint, locating the second text content that does not comply with the citation standard constraint in the target academic resource;

[0041] While providing a reference standard prompt for the second text content, a pre-trained reference association model is used to query a first document resource that has a reference association with the second text content;

[0042] If a citation adjustment instruction of an academic resource is received, adjusting the citation mark of the second text content in the target academic resource according to the first document resource;

[0043] Furthermore, the method further comprises:

[0044] Obtain the third text content of unmarked reference literature resources in the target academic resources;

[0045] While generating a citation prompt for the third text content, a pre-trained citation association model is used to query a second document resource that has a citation association with the third text content;

[0046] If a citation generation instruction for an academic resource is received, a citation mark for the third text content in the target academic resource is generated based on the second document resource.

[0047] According to a second aspect of the present application, a standardized processing device for academic resources is provided, comprising:

[0048] The first acquisition unit is used to obtain the target academic resources in the creative process;

[0049] An element identification unit, used for identifying elements in the target academic resource according to the resource type of the target academic resource, to obtain resource elements of the target academic resource;

[0050] A specification identification unit, used for performing at least one specification identification on the resource elements in the target academic resources to obtain an identification result of the resource elements, wherein the identification result of the resource elements includes whether the application of the resource elements in the target academic resources complies with the specification constraints;

[0051] A prompting unit is used to provide standard prompts for resource elements in the target academic resources that do not meet standard constraints based on the identification results of the resource elements.

[0052] Furthermore, the element identification unit is specifically used to:

[0053] According to the resource type of the target academic resource, a target academic resource sample of the resource type corresponding to the target academic resource is obtained from a pre-precipitated resource set, wherein the resource set includes academic resource samples of different resource types, and the academic resource sample has resource elements of the corresponding resource type;

[0054] The target academic resource sample is used as a comparison sample to perform element identification in the target academic resource to obtain the resource elements of the target academic resource.

[0055] Furthermore, the device also includes:

[0056] A creation unit, used for creating a normative knowledge base of academic resources before performing at least one normative identification on resource elements in the target academic resources to obtain identification results of resource elements, wherein the normative knowledge base includes normative constraints formed by processing different academic normative resources;

[0057] A first training unit is used to input the standard knowledge base and academic resource samples of different resource types into a machine learning model, so as to train the mapping relationship of whether the application of resource elements in the academic resource samples meets the format requirements of the standard constraints through the machine learning model according to the standard constraints;

[0058] A first generating unit is used to generate a format specification recognition model according to the mapping relationship obtained through training when the training process meets the iteration stopping condition;

[0059] Correspondingly, the specification recognition unit is specifically used to use a pre-trained format specification recognition model to perform format specification recognition on the resource elements in the target academic resources to obtain format recognition results of the resource elements.

[0060] Furthermore, the specification identification unit is further configured to:

[0061] Obtaining text content of resource elements in the target academic resource;

[0062] After converting the text content of the resource element into a format text vector using a pre-trained format specification recognition model, judging whether the format text vector meets the format requirements of the specification constraint according to the constraint specification;

[0063] If so, it is determined that the application of resource elements in the target academic resources complies with the format specification constraints; otherwise, it is determined that the application of resource elements in the target academic resources does not comply with the format specification constraints.

[0064] Furthermore, the prompt unit is specifically used for:

[0065] When the application of the resource element in the target academic resource does not comply with the format specification constraint, locating the first text content that does not comply with the format specification constraint in the target academic resource;

[0066] While providing format specification prompts for the first text content, text content examples with corresponding format requirements for the first text content are displayed. The text content examples are text content extracted from academic resources that conform to format specification constraints.

[0067] Furthermore, the device also includes:

[0068] A second acquisition unit is used to acquire historically added auxiliary reference resources before performing at least one standard recognition on the resource elements in the target academic resource to obtain the recognition result of the resource elements, wherein the auxiliary reference resources at least include the literature resources with which the target academic resource in the creation process has a citation association;

[0069] A second training unit is used to input the auxiliary reference resources and academic resource samples of different resource types into a machine learning model, so as to train the mapping relationship of whether the application of resource elements in the academic resource samples meets the reference requirements of the specification constraints according to the reference association through the machine learning model;

[0070] A second generating unit is used to generate a reference specification recognition model according to the mapping relationship obtained through training when the training process meets the iteration stopping condition;

[0071] Correspondingly, the specification identification unit is specifically used to use a pre-trained citation specification identification model to perform citation specification identification on resource elements in the target academic resources to obtain citation identification results of the resource elements.

[0072] Furthermore, the specification identification unit is further configured to:

[0073] According to the text content of the resource element in the target academic resource, obtain the reference document resource marked by the text content, wherein the reference document resource is a document resource having a reference association with the text content;

[0074] After converting the text content of the resource element into a content text vector using a pre-trained citation specification recognition model, judging whether the content text vector meets the citation requirements of the specification constraints according to the similarity result between the content text vector and the text vector of the cited document resource;

[0075] If so, it is determined that the application of resource elements in the target academic resources complies with the citation standard constraints; otherwise, it is determined that the application of resource elements in the target academic resources does not comply with the citation standard constraints.

[0076] Furthermore, the device also includes:

[0077] A third training unit is used for inputting the auxiliary reference resources and academic resource samples of different resource types into the machine learning model after the auxiliary reference resources added by the acquisition history are obtained, so as to train the mapping relationship between the text content of the resource elements in the academic resource samples and the auxiliary reference resources according to the citation association through the machine learning model;

[0078] A third generating unit is used to generate a reference association model according to the mapping relationship obtained through training when the training process meets the iteration stopping condition;

[0079] Accordingly, the prompt unit is specifically used for:

[0080] When the application of the resource element in the target academic resource does not comply with the citation standard constraint, locating the second text content that does not comply with the citation standard constraint in the target academic resource;

[0081] While providing a reference standard prompt for the second text content, a pre-trained reference association model is used to query a first document resource that has a reference association with the second text content;

[0082] If a citation adjustment instruction for an academic resource is received, the citation mark of the second text content in the target academic resource is adjusted according to the first document resource.

[0083] Furthermore, the device also includes:

[0084] A third acquisition unit is used to acquire the third text content of the unmarked reference document resources in the target academic resource;

[0085] A query unit, configured to query a second document resource having a citation association with the third text content by using a pre-trained citation association model while generating a prompt for citing the third text content;

[0086] The fourth generating unit is used to generate a reference mark for the third text content in the target academic resource according to the second document resource if a reference generation instruction for the academic resource is received.

[0087] According to a third aspect of the present application, a computer device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in the first aspect when executing the computer program.

[0088] According to a fourth aspect of the present application, a readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0089] By means of the above technical scheme, the present application provides a method and device for standardized processing of academic resources. Compared with the current existing technology that realizes standardized processing of academic resources through training programs and paper duplication checking, the present application obtains the target academic resources in the creation process, identifies elements in the target academic resources according to the resource types of the target academic resources, obtains the resource elements of the target academic resources, performs at least one standardized identification on the resource elements in the target academic resources, obtains the identification results of the resource elements, and the identification results of the resource elements include whether the application of the resource elements in the target academic resources complies with the standardized constraints. According to the identification results of the resource elements, standardized prompts are given to the resource elements in the target academic resources that do not comply with the standardized constraints. The whole process solves the standardization problems of resource elements in the creation process by integrating standardized processing into the creation process of academic resources, so as to avoid the standardization problems in the academic creation process in advance, and improve the level of academic creation while ensuring the efficiency of academic creation.

[0090] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0092] Figure 1 It is a flowchart of a standardized processing method of academic resources provided in one embodiment of the present application;

[0093] Figure 2 yes Figure 1 A schematic flow chart of a specific implementation of step 102;

[0094] Figure 3 yes Figure 1 A schematic diagram of a specific implementation process before step 103;

[0095] Figure 4 yes Figure 1 Another specific implementation flow chart before step 103;

[0096] Figure 5 yes Figure 4 A schematic diagram of a specific implementation process after step 401;

[0097] Figure 6 This is a front-end framework structure diagram for standardizing the processing of academic resources provided by an embodiment of the present application;

[0098] Figure 7 This is a backend framework structure diagram for standardizing the processing of academic resources provided by an embodiment of the present application;

[0099] Figure 8 It is a structural diagram of a standard processing device for academic resources in one embodiment of the present application;

[0100] Fig. 9 It is a schematic diagram of the device structure of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0101] Now the content of the present invention will be discussed with reference to several exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and thus implement the content of the present invention, rather than implying any limitation on the scope of the present invention.

[0102] As used herein, the term "including" and variations thereof are to be interpreted as open-ended terms meaning "including but not limited to." The term "based on" is to be interpreted as "based, at least in part, on." The terms "one embodiment" and "an embodiment" are to be interpreted as "at least one embodiment." The term "another embodiment" is to be interpreted as "at least one other embodiment."

[0103] In related technologies, standardized processing of academic resources can be achieved through training programs and paper duplication checks. However, both of these academic standardization methods standardize the creation of completed academic resources. It is difficult to integrate standardization issues into the resource creation process, which affects the improvement of academic creation levels.

[0104] In order to solve this problem, this embodiment provides a standardized processing method for academic resources, such as Figure 1 As shown, the method is applied to the server side of academic resource creation, including the following steps:

[0105] 101. Obtain target academic resources during the creative process.

[0106] Generally speaking, the creation process of academic resources is inseparable from academic norms. Academic norms provide guidance and norms for scholars' academic work, ensure the rigor, reliability and fairness of academic research, and play a vital role in the academic community. In order to ensure that the academic resources in the creation process follow academic norms, target academic resources can be selected for normative processing during the creation process. This normative processing can identify whether the resource elements in the target academic resources meet the academic norms, so as to make normative recommendations, normative prompts, and normative corrections for the target academic resources.

[0107] In this embodiment, the academic resources in the creative process can be paper resources, journal resources, database resources, etc. Here, the target academic resources can be all the content in the academic resources in the creative process, or it can be part of the content selected in the creative process, for example, the content of the abstract part of the creative process, or the content of a certain paragraph in the creative process.

[0108] The executor of this embodiment can be a standardized processing device or equipment for academic resources, which can be configured on the server side corresponding to the standardized processing of academic resources. By acquiring the target academic resources in the creation process and performing standardized processing on the set academic resources, the standardized issues can be deeply integrated into the creation process of academic resources, thereby playing a role in improving academic standardization capabilities.

[0109] 102. Identify elements in the target academic resource according to the resource type of the target academic resource to obtain resource elements of the target academic resource.

[0110] It is understandable that the resource types of different academic resources have different meanings for academic exchanges and scientific research activities. The richness or diversification of resource types is of great significance to academic exchanges and scientific research. The resource types of the target academic resources here include papers, journals, conference minutes, academic achievements, academic conference papers, etc. Usually before creating academic resources, the author can select the resource type of academic resources. Considering that different resource types have different structural characteristics and corresponding element compositions, the resource elements of the target academic resources need to follow the constraints of the corresponding structural characteristics during the creation process.

[0111] For example, academic resources of the paper type include resource elements such as title, abstract, introduction, references and conclusion. Among them, the resource element of the title needs to be concise, specific and accurate, and be able to summarize the specific content of the paper. The resource element of the abstract is a brief statement of the content of the paper without annotations and comments, and it is required to briefly explain the purpose, research methods and final conclusions of the research work.

[0112] Specifically, the target academic resources can be identified by using a pre-trained element recognition model, so that the target academic resources can be structurally analyzed according to the structural characteristics of the target academic papers of the corresponding type through the element recognition model, so as to locate resource elements with different structural characteristics in the target academic resources. For example, the target academic resource in the creation process is a journal type, and the element recognition model can be used to identify the resource elements of the target academic resource including the introduction and the method. It is understandable that the target academic resource in the creation process is not necessarily the entire academic resource, that is, the target academic resource may only contain some resource elements in the academic resource. Accordingly, the resource elements identified by the element recognition model are not the resource elements of the entire academic resource, but the resource elements of the target academic resource.

[0113] 103. Perform at least one standardized identification on the resource elements in the target academic resources to obtain identification results of the resource elements.

[0114] Among them, the identification result of resource elements includes whether the application of resource elements in the target academic resources complies with the normative constraints. Here, the normative constraints may include but are not limited to structural normative constraints, content normative constraints, and reference normative constraints, etc., which can be obtained by processing and sorting out academic norm-related standards, rules, manuals, video materials, etc. Correspondingly, in the process of performing at least one normative identification on the resource elements in the target academic resources, the format normative identification of the resource elements in the target academic resources can be performed according to the format normative constraints, the content normative identification of the resource elements in the target academic resources can be performed according to the content normative constraints, and the reference normative identification of the resource elements in the target academic resources can be performed according to the reference normative constraints.

[0115] Specifically, in the process of format specification identification, for each resource element in the target academic resource, the format of the resource element is obtained. The format here mainly considers the location specification and font specification of the resource element. The format of the resource element is compared with the standard format to determine whether the format of the resource element meets the format specification requirements. The application of the resource element that meets the requirements in the target academic resource complies with the format specification constraints. For example, the title format of the abstract in the academic resource of the paper will appear in a prominent position in the header or footer with words such as "Abstract" or "Research Abstract", and the font of the abstract name is one font size larger than the abstract text. At this time, the text content in the resource element meets the format specification requirements.

[0116] Specifically, in the process of identifying content standards, for each resource element in the target academic resource, the content of the resource element is obtained. The content here mainly considers the background, purpose, method and results contained in the resource element. The content of the resource element is compared with the standard content to determine whether the content of the resource element meets the requirements of the content standards. The application of the resource elements that meet the requirements in the target academic resources complies with the content standard constraints. For example, the content of the literature review in the academic resources of the paper should review previous related research and viewpoints, outline the development trend of previous research, and point out the research gaps in this field. For example, the conclusion section should summarize the main performance of the paper and restate the research purpose and questions, rather than answering the research hypothesis or goal. At this time, the text content in the resource element meets the requirements of the content standard.

[0117] Specifically, in the process of identifying citation standards, for each resource element in the target academic resource, the citation of the resource element is obtained. The citation here mainly considers the literature cited in the text content of the resource element. The content of the resource element is compared with the content of the cited literature to determine whether the citation of the resource element meets the requirements of the citation standards. The application of resource elements that meet the requirements in the target academic resource complies with the citation standard constraints. For example, the similarity between the text content in the resource element in the paper academic resource and the cited literature should be greater than the set value. At this time, the text content in the resource element meets the requirements of the citation standard.

[0118] 104. Based on the identification results of the resource elements, provide standardized prompts for the resource elements in the target academic resources that do not meet the standardized constraints.

[0119] It is understandable that for the resource elements in the target academic resources that do not meet the standard constraints, it means that the resource elements have standard problems, and further standard prompts are given to the resource elements with standard problems. The standard prompts here can be marking prompts, such as yellow, bold or underlined, etc., which are not limited in this embodiment. For example, if the font size of the abstract name in the target academic resource does not meet the format specifications, a standard prompt can be given by marking the position of the abstract name in yellow. The specific prompt content can be that the font size of the abstract name should be No. 4 Songti. Of course, in order to facilitate the creator to carry out standard processing, an automatic correction function can also be set for the resource elements that do not meet the standard constraints. The automatic correction function can be used to correct the resource elements that do not meet the standard constraints according to the requirements of the standard constraints on academic resources.

[0120] Considering the limitations of creators’ understanding of standard issues in academic resources, we can further obtain the resource template corresponding to the target academic resource during the creation process, and show the creators the standard resource elements covered in the academic resource through the resource template, and provide the related standard explanations of the standard resource elements, so that creators can avoid various standard issues in advance, and can also play a role in guiding creators in the creation of academic resources. For example, in the process of creating an abstract, the creator is shown the abstract template and the related standard explanations of the abstract, including but not limited to the abstract concept, abstract content, abstract format, etc.

[0121] The standardized processing method of academic resources provided in the embodiment of the present application is compared with the current prior art method of achieving standardized processing of academic resources through training programs and paper duplication checking. The present application obtains the target academic resources in the creation process, identifies elements in the target academic resources according to the resource type of the target academic resources, obtains the resource elements of the target academic resources, performs at least one standardized identification on the resource elements in the target academic resources, obtains the identification results of the resource elements, and the identification results of the resource elements include whether the application of the resource elements in the target academic resources complies with the standardized constraints. According to the identification results of the resource elements, standardized prompts are given to the resource elements in the target academic resources that do not comply with the standardized constraints. The entire process solves the standardization problems of resource elements in the creation process by integrating standardized processing into the creation process of academic resources, so as to avoid the standardization problems in the academic creation process in advance, and improve the level of academic creation while ensuring the efficiency of academic creation.

[0122] In the above embodiment, it is considered that academic resources of different resource types contain resource elements of different resource types, specifically, Figure 2 As shown, step 102 includes the following steps:

[0123] 201. According to the resource type of the target academic resource, a target academic resource sample of the resource type corresponding to the target academic resource is obtained from a pre-precipitated resource set.

[0124] 202. Using the target academic resource sample as a comparison sample, perform element identification in the target academic resource to obtain resource elements of the target academic resource.

[0125] Among them, the resource collection includes academic resource samples of different resource types. The academic resource samples have resource elements of the corresponding resource types. For example, academic resources of the paper resource type include title, abstract, table of contents, introduction, text, conclusion and other resource elements. Academic resource samples of different resource types can be used as comparison samples of target academic resources of the corresponding resource type. The resource elements of the target academic resources can be determined by comparing the samples.

[0126] Of course, in order to accurately identify the resource elements of the target academic resources, different academic resource samples can also be used as training samples to input into the machine learning model, so that the mapping relationship between academic resources of different resource types and resource elements can be trained through the machine learning model according to the resource elements in the academic resource samples, and the element recognition model can be constructed according to the mapping relationship obtained through training. Furthermore, the element recognition model can be used to identify elements in the target academic resources to obtain the resource elements of the target academic resources.

[0127] In the above embodiment, at least one standard recognition includes format standard recognition of resource elements in the target academic resource. In order to accurately recognize the format standard of resource elements in the target academic resource, further, Figure 3 As shown, before step 103, the following steps are included:

[0128] 301. Create a canonical knowledge base of academic resources.

[0129] 302. Input the standard knowledge base and academic resource samples of different resource types into the machine learning model, so as to train the mapping relationship through the machine learning model based on the standard constraints to determine whether the application of resource elements in the academic resource samples meets the format requirements of the standard constraints.

[0130] 303. When the training process meets the iteration stop condition, a format specification recognition model is generated according to the mapping relationship obtained through training.

[0131] Correspondingly, in step 103, a pre-trained format specification recognition model is used to perform format specification recognition on resource elements in the target academic resources to obtain format recognition results of the resource elements.

[0132] In this embodiment, the normative knowledge base includes normative constraints formed by processing different academic normative resources. Specifically, the normative constraints can be formed by collecting and organizing academic normative resources such as standards, guidelines, manuals, and videos related to academic norms, and processing the academic normative resources. The normative constraints include but are not limited to format constraints, content constraints, and citation constraints, etc.

[0133] Specifically, in the process of format specification identification of resource elements in the target academic resources, the text content of the resource elements in the target academic resources can be obtained, and then the text content of the resource elements can be converted into a format text vector using a pre-trained format specification identification model. Then, according to the constraint specification, it is determined whether the format requirement of the specification constraint is met. If so, it is determined that the application of the resource elements in the target academic resources complies with the format specification constraint. Otherwise, it is determined that the application of the resource elements in the target academic resources does not comply with the format specification constraint.

[0134] Correspondingly, step 104 specifically includes the following contents: when the application of resource elements in the target academic resources does not comply with the format specification constraints, the first text content that does not comply with the format specification constraints is located in the target academic resource, and while providing format specification prompts for the first text content, a text content example corresponding to the format requirements of the first text content is displayed, where the text content example is text content that complies with the format specification constraints and is extracted from the academic resource.

[0135] For example, the formatting constraints for the title page in a paper are that each line of text is centered, using size 4 font and bold. For text content in the title page of a paper that does not comply with the formatting constraints, text content examples with the corresponding formatting constraints can be displayed for the creator's reference.

[0136] In the above embodiment, at least one standard identification includes citing standard identification of resource elements in the target academic resource. In order to accurately cite the resource elements in the target academic resource, further, Figure 4 As shown, before step 103, the method further includes the following steps:

[0137] 401. Obtain auxiliary reference resources added historically, wherein the auxiliary reference resources at least include literature resources with citation associations to target academic resources in the creation process.

[0138] 402. Input the auxiliary reference resources and academic resource samples of different resource types into the machine learning model, so as to train the mapping relationship through the machine learning model based on the citation association to determine whether the application of resource elements in the academic resource samples meets the citation requirements of the specification constraints.

[0139] 403. When the training process meets the iteration stop condition, a reference specification recognition model is generated according to the mapping relationship obtained through training.

[0140] Correspondingly, in step 103, a pre-trained citation specification recognition model is used to perform citation specification recognition on resource elements in the target academic resources to obtain citation recognition results of the resource elements.

[0141] In this embodiment, the auxiliary reference resources at least include document resources that have citation associations with the target academic resources in the creation process. For example, the first sentence of the abstract quotes document resource 1 in the auxiliary reference resource, and the second paragraph of the main text quotes document resource 2 in the auxiliary reference resource. Here, auxiliary reference resources can be continuously added during the creation process of academic resources, which is equivalent to the accumulation process of document resources.

[0142] Specifically, according to the text content of the resource elements in the target academic resources, the cited literature resources marked with the text content are obtained. Here, the cited literature resources are literature resources that have a citation association with the text content. After the text content of the resource elements is converted into a content text vector using a pre-trained citation specification recognition model, it is judged whether the content text vector meets the citation requirements of the specification constraints based on the similarity results between the content text vector and the text vector of the cited literature resource. If so, it is determined that the application of the resource elements in the target academic resources complies with the citation specification constraints. Otherwise, it is determined that the application of the resource elements in the target academic resources does not comply with the citation specification constraints.

[0143] It can be understood that the similarity between the content text vector and the text vector of the cited literature resource can reflect the similarity between the text content and the cited references in the target academic resource. If the similarity between the two is high, it means that the text part does use the cited references, and the application of the cited references is in compliance with the citation norms. If the similarity between the two is low, it means that the text part does not use the cited references, and the application of the cited references is not in compliance with the citation norms. Here, the similarity threshold can be used to determine whether the cited references of the text content meet the citation norms.

[0144] Furthermore, considering that the reference mark correction is performed on the text in the target academic resource that does not conform to the reference standard constraint, in the above embodiment, Figure 5 As shown, after step 401, the method further includes the following steps:

[0145] 501. Input the auxiliary reference resources and academic resource samples of different resource types into a machine learning model, so as to train the mapping relationship between the text content of the resource elements in the academic resource samples and the auxiliary reference resources based on the citation association through the machine learning model.

[0146] 502. When the training process meets the iteration stop condition, a reference association model is generated according to the mapping relationship obtained through training.

[0147] Correspondingly, in the above embodiment, step 104 specifically includes the following steps: when the application of resource elements in the target academic resource does not comply with the citation standard constraints, the second text content that does not comply with the citation standard constraints is located in the target academic resource, and while providing a citation standard prompt for the second text content, a pre-trained citation association model is used to query the first document resource that has a citation association with the second text content. If a citation adjustment instruction for the academic resource is received, the citation mark of the second text content in the target academic resource is adjusted according to the first document resource.

[0148] For example, the citation of the second paragraph of text in the main body of the target academic resource does not comply with the citation standard constraints. The citation association model is used to query that the document resource 5 has a citation association with the second paragraph of text, and the document resource 5 is further used as a citation marker for the second paragraph of text.

[0149] In this embodiment, based on the prompt of citation norms for academic resources, the citation adjustment of academic resources can be triggered by the citation correction function of academic resources, and the citation norms can also be automatically adjusted according to the citation association model.

[0150] It should be noted that when the first document resource having a reference association with the second text content is not found through the reference association model, the reference mark of the second text content in the target academic resource can be deleted, and document resources can also be queried from other academic resource platforms.

[0151] Furthermore, in order to facilitate the standardized processing of academic freedom and enable automatic association of cited document resources, the above method also includes the following steps based on the training of the citation association model: obtaining a third text content of the target academic resource that is not marked with a cited document resource; while providing a citation generation prompt for the third text content, using a pre-trained citation association model to query a second document resource that has a citation association with the third text content; if a citation generation instruction for an academic resource is received, generating a citation mark for the third text content in the target academic resource based on the second document resource.

[0152] For example, the fourth paragraph of text in the main body of the target academic resource is not marked with a cited document resource. Through the citation association model query, it is found that document resource 7 has a citation association with the fourth paragraph of text, and document resource 7 is further used as a citation mark for the fourth paragraph of text.

[0153] In this embodiment, based on the prompt of the citation specification of the academic resources, the citation association of the academic resources can be triggered by the citation generation function of the academic resources, and the citation specification can also be automatically generated according to the citation association model.

[0154] Specifically in the actual reference scenario, Figure 6 The front-end framework structure diagram for standardized processing of academic resources. The front-end framework for standardized processing of academic resources mainly includes application layer, incremental tasks, interfaces, task queues, databases, task scheduling, components and other modules. In the application layer, it can provide creators with functions such as initiating detection, result query, history record, title assembly, original text annotation, content interaction and user verification. In incremental tasks, it can provide scheduled tasks, monitoring data dynamics, file / kns full text and call increments. In the interface, it can provide JWT, detection interface, result query, incremental interface, comparison data update and latest time functions. In the task queue, it provides detection, incremental vectorization, data collection and other functions. In the database, it can provide JWT, detection result library, incremental original text library, LOG, task library, all relationship library and other functions. In the component, it can provide PDF full-text extraction, image full-text extraction, data collection & increment, vectorization service, word full-text extraction, HTML full-text extraction and comparison service and other functions.

[0155] Based on the above-mentioned framework structure for the standardized processing of academic resources, the process of standardized processing of academic resources includes detection flow, query flow and detection task flow. For the detection flow, after the detection is initiated, the tag and full text are obtained, written into the detection library, the detection task is generated, written into the queue, and the detection id is returned. For the query flow, after the query is initiated, the id / result type is detected. If the detection is completed, the result is returned through the JSON result. If the detection is not completed, the result is returned directly. For the detection task flow, after the task is discovered, the tag and parameters are obtained, and the detection data is obtained through the detection task. It is determined whether there is a comparison library with the specified tag. If there is a comparison library with the specified tag, the data is updated and collected. Then it is determined whether there is text. If so, the text is extracted. When the text is extracted successfully, the comparison service is called. When the text extraction fails, the task status is modified, and then the task is terminated. If there is no comparison library with the specified tag, the data collection task is initiated and written into the queue. Then the detection task is re-initiated and written into the queue. Then the task is terminated. If an error occurs in the detection task, the task is terminated. The process of standardizing the processing of incremental academic resources includes incremental flow and incremental task flow. For incremental flow, after the incremental task is initiated, the tag, file, and business system file key are obtained, and then MDS calculation is performed and written into the incremental original document library file storage. The incremental task is generated and then ended. For incremental task flow, after the task is discovered, the task type and parameters are obtained, and the incremental data is obtained by detecting the task. If the incremental data is text, the vectorization service is called, the task status is modified and the task is ended. If the incremental data is not text, the text is extracted. If the text is extracted, the vectorization service is called, the task status is modified and the task is ended. Otherwise, the task status is modified and the task is ended. If an error occurs in the detection task, the task is ended.

[0156] Specifically in the actual reference scenario, Figure 7The back-end framework structure diagram for standardized processing of academic resources. The back-end framework for standardized processing of the entire academic resources includes modules such as comparison result processing, similarity calculation, vectorization processing, data storage and text processing. Based on the original basic data, data access is initiated through two channels: concurrent processing and exception processing. The academic resources to be processed and the original basic data are passed to the text processing through the data access module. The text processing performs text cleaning, text segmentation, sentence segmentation and word segmentation on the academic resources to be processed and the original basic data. Then, the text-processed data is vectorized through the training data set, including data extraction, vectorization algorithm library, model training, model saving and loading, text vectorization, model update and model optimization, etc. Then, the vectorized data is passed to the data storage module, and the vector storage format conversion is implemented in the data storage module to obtain the vector model file, including the vocabulary, sentence library, text block, etc. Then, the similarity calculation module is used to implement the similarity calculation, threshold determination and sorting algorithm. Finally, the comparison processing module is used to implement the result evaluation, and then the result format conversion - json and API access.

[0157] Further, as Figure 1-5 The specific implementation of the method, the embodiment of the present application provides a standardized processing device for academic resources, such as Figure 8 As shown, the device includes: a first acquisition unit 61, an element recognition unit 62, a standard recognition unit 63, and a prompt unit 64.

[0158] A first acquisition unit 61 is used to acquire target academic resources in the creation process;

[0159] An element identification unit 62 is used to identify elements in the target academic resource according to the resource type of the target academic resource to obtain resource elements of the target academic resource;

[0160] A specification identification unit 63 is used to perform at least one specification identification on the resource elements in the target academic resource to obtain an identification result of the resource element, wherein the identification result of the resource element includes whether the application of the resource element in the target academic resource complies with the specification constraint;

[0161] The prompting unit 64 is used to provide standard prompts for the resource elements in the target academic resources that do not meet the standard constraints according to the identification results of the resource elements.

[0162] The standardized processing device for academic resources provided by the embodiment of the present invention is compared with the current prior art method of realizing standardized processing of academic resources through training programs and paper duplication checking. The present application obtains the target academic resources in the creation process, identifies elements in the target academic resources according to the resource types of the target academic resources, obtains the resource elements of the target academic resources, performs at least one standardized identification on the resource elements in the target academic resources, obtains the identification results of the resource elements, and the identification results of the resource elements include whether the application of the resource elements in the target academic resources complies with the standardized constraints. According to the identification results of the resource elements, standardized prompts are given to the resource elements in the target academic resources that do not comply with the standardized constraints. The whole process solves the standardization problems of resource elements in the creation process by integrating standardized processing into the creation process of academic resources, so as to avoid the standardization problems in the academic creation process in advance, and improve the level of academic creation while ensuring the efficiency of academic creation.

[0163] In a specific application scenario, the element identification unit is specifically used to:

[0164] According to the resource type of the target academic resource, a target academic resource sample of the resource type corresponding to the target academic resource is obtained from a pre-precipitated resource set, wherein the resource set includes academic resource samples of different resource types, and the academic resource sample has resource elements of the corresponding resource type;

[0165] The target academic resource sample is used as a comparison sample to perform element identification in the target academic resource to obtain the resource elements of the target academic resource.

[0166] In a specific application scenario, the device further includes:

[0167] A creation unit, used for creating a normative knowledge base of academic resources before performing at least one normative identification on resource elements in the target academic resources to obtain identification results of resource elements, wherein the normative knowledge base includes normative constraints formed by processing different academic normative resources;

[0168] A first training unit is used to input the standard knowledge base and academic resource samples of different resource types into a machine learning model, so as to train the mapping relationship of whether the application of resource elements in the academic resource samples meets the format requirements of the standard constraints through the machine learning model according to the standard constraints;

[0169] A first generating unit is used to generate a format specification recognition model according to the mapping relationship obtained through training when the training process meets the iteration stopping condition;

[0170] Correspondingly, the specification recognition unit is specifically used to use a pre-trained format specification recognition model to perform format specification recognition on the resource elements in the target academic resources to obtain format recognition results of the resource elements.

[0171] In a specific application scenario, the specification identification unit is further used to:

[0172] Obtaining text content of resource elements in the target academic resource;

[0173] After converting the text content of the resource element into a format text vector using a pre-trained format specification recognition model, judging whether the format text vector meets the format requirements of the specification constraint according to the constraint specification;

[0174] If so, it is determined that the application of resource elements in the target academic resources complies with the format specification constraints; otherwise, it is determined that the application of resource elements in the target academic resources does not comply with the format specification constraints.

[0175] In a specific application scenario, the prompt unit is specifically used to:

[0176] When the application of the resource element in the target academic resource does not comply with the format specification constraint, locating the first text content that does not comply with the format specification constraint in the target academic resource;

[0177] While providing format specification prompts for the first text content, text content examples with corresponding format requirements for the first text content are displayed. The text content examples are text content extracted from academic resources that conform to format specification constraints.

[0178] In a specific application scenario, the device further includes:

[0179] A second acquisition unit is used to acquire historically added auxiliary reference resources before performing at least one standard recognition on the resource elements in the target academic resource to obtain the recognition result of the resource elements, wherein the auxiliary reference resources at least include the literature resources with which the target academic resource in the creation process has a citation association;

[0180] A second training unit is used to input the auxiliary reference resources and academic resource samples of different resource types into a machine learning model, so as to train the mapping relationship of whether the application of resource elements in the academic resource samples meets the reference requirements of the specification constraints according to the reference association through the machine learning model;

[0181] A second generating unit is used to generate a reference specification recognition model according to the mapping relationship obtained through training when the training process meets the iteration stopping condition;

[0182] Correspondingly, the specification identification unit is specifically used to use a pre-trained citation specification identification model to perform citation specification identification on resource elements in the target academic resources to obtain citation identification results of the resource elements.

[0183] In a specific application scenario, the specification identification unit is further used to:

[0184] According to the text content of the resource element in the target academic resource, obtain the reference document resource marked by the text content, wherein the reference document resource is a document resource having a reference association with the text content;

[0185] After converting the text content of the resource element into a content text vector using a pre-trained citation specification recognition model, judging whether the content text vector meets the citation requirements of the specification constraints according to the similarity result between the content text vector and the text vector of the cited document resource;

[0186] If so, it is determined that the application of resource elements in the target academic resources complies with the citation standard constraints; otherwise, it is determined that the application of resource elements in the target academic resources does not comply with the citation standard constraints.

[0187] In a specific application scenario, the device further includes:

[0188] A third training unit is used for inputting the auxiliary reference resources and academic resource samples of different resource types into the machine learning model after the auxiliary reference resources added by the acquisition history are obtained, so as to train the mapping relationship between the text content of the resource elements in the academic resource samples and the auxiliary reference resources according to the citation association through the machine learning model;

[0189] A third generating unit is used to generate a reference association model according to the mapping relationship obtained through training when the training process meets the iteration stopping condition;

[0190] Accordingly, the prompt unit is specifically used for:

[0191] When the application of the resource element in the target academic resource does not comply with the citation standard constraint, locating the second text content that does not comply with the citation standard constraint in the target academic resource;

[0192] While providing a reference standard prompt for the second text content, a pre-trained reference association model is used to query a first document resource that has a reference association with the second text content;

[0193] If a citation adjustment instruction for an academic resource is received, the citation mark of the second text content in the target academic resource is adjusted according to the first document resource.

[0194] In a specific application scenario, the device further includes:

[0195] A third acquisition unit is used to acquire the third text content of the unmarked reference document resources in the target academic resource;

[0196] A query unit, configured to query a second document resource having a citation association with the third text content by using a pre-trained citation association model while generating a prompt for citing the third text content;

[0197] The fourth generating unit is used to generate a reference mark for the third text content in the target academic resource according to the second document resource if a reference generation instruction for the academic resource is received.

[0198] It should be noted that for other corresponding descriptions of the functional units involved in the standard processing device for academic resources provided in this embodiment, please refer to Figure 1-Figure 5 The corresponding description in will not be repeated here.

[0199] Based on the above Figure 1-Figure 5 The method shown in the embodiment of the present application accordingly provides a storage medium on which a computer program is stored, and when the program is executed by a processor, the above-mentioned Figure 1-Figure 5 The canonical treatment of scholarly resources shown.

[0200] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0201] Based on the above Figure 1-Figure 5 The method shown, and Figure 8 In order to achieve the above-mentioned purpose, the embodiment of the present application also provides a physical device for the standardized processing of academic resources, which can be a computer, a smart phone, a tablet computer, a smart watch, a server, or a network device, etc. The physical device includes a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to achieve the above-mentioned Figure 1-Figure 5 The canonical treatment of scholarly resources shown.

[0202] Optionally, the physical device may also include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a WI-FI module, etc. The user interface may include a display, an input unit such as a keyboard, etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface), etc.

[0203] In an exemplary embodiment, see Fig. 9 The physical device includes a communication bus, a processor, a memory and a communication interface, and may also include an input / output interface and a display device, wherein each functional unit can communicate with each other through the bus. The memory stores a computer program, and the processor is used to execute the program stored in the memory and execute the standard processing method of academic resources in the above embodiment.

[0204] Those skilled in the art will understand that the physical device structure for the standardized processing of academic resources provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or a combination of certain components, or different arrangements of components.

[0205] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the physical device for the standardized processing of the above-mentioned academic resources, and supports the operation of information processing programs and other software and / or programs. The network communication module is used to realize the communication between the components inside the storage medium, and the communication with other hardware and software in the information processing physical device.

[0206] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented by means of software plus the necessary general hardware platform, or by hardware. By applying the technical solution of the present application, compared with the current existing methods, the present application solves the standardization problem of resource elements in the creation process by integrating standardization processing into the creation process of academic resources, so as to avoid standardization problems in the academic creation process in advance, and improve the level of academic creation while ensuring the efficiency of academic creation.

[0207] Those skilled in the art will appreciate that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily necessary for implementing the present application. Those skilled in the art will appreciate that the modules in the devices in the implementation scenario can be distributed in the devices of the implementation scenario according to the description of the implementation scenario, or can be changed accordingly and located in one or more devices different from the present implementation scenario. The modules of the above-mentioned implementation scenario can be combined into one module, or can be further split into multiple submodules.

[0208] The above serial numbers of this application are only for description and do not represent the advantages and disadvantages of the implementation scenarios. The above disclosure is only a few specific implementation scenarios of this application, but this application is not limited to them, and any changes that can be thought of by technicians in this field should fall within the scope of protection of this application.

Claims

1. A standardized processing method for academic resources, characterized in that: include: Acquire targeted academic resources in the creative process; Identify elements in the target academic resource according to the resource type of the target academic resource to obtain resource elements of the target academic resource; Perform at least one specification identification on the resource elements in the target academic resource to obtain an identification result of the resource element, wherein the identification result of the resource element includes whether the application of the resource element in the target academic resource complies with the specification constraint; According to the identification results of the resource elements, standardized prompts are given to the resource elements in the target academic resources that do not meet the standardized constraints.

2. The method according to claim 1, characterized in that The step of identifying elements in the target academic resource according to the resource type of the target academic resource to obtain resource elements of the target academic resource includes: According to the resource type of the target academic resource, a target academic resource sample of the resource type corresponding to the target academic resource is obtained from a pre-precipitated resource set, wherein the resource set includes academic resource samples of different resource types, and the academic resource sample has resource elements of the corresponding resource type; The target academic resource sample is used as a comparison sample to perform element identification in the target academic resource to obtain the resource elements of the target academic resource.

3. The method according to claim 1, characterized in that Before performing at least one specification identification on the resource elements in the target academic resources to obtain the identification results of the resource elements, the method further includes: Creating a normative knowledge base of academic resources, wherein the normative knowledge base includes normative constraints formed by processing different academic normative resources; Inputting the specification knowledge base and academic resource samples of different resource types into a machine learning model, so as to train the mapping relationship of whether the application of resource elements in the academic resource samples meets the format requirements of the specification constraints through the machine learning model according to the specification constraints; When the training process meets the iteration stop condition, a format specification recognition model is generated according to the mapping relationship obtained through training; Accordingly, performing at least one standard identification on the resource elements in the target academic resources to obtain the identification result of the resource elements includes: The pre-trained format specification recognition model is used to perform format specification recognition on the resource elements in the target academic resources to obtain the format recognition results of the resource elements.

4. The method according to claim 3, characterized in that The method of using the pre-trained format specification recognition model to perform format specification recognition on resource elements in the target academic resources to obtain format recognition results of the resource elements includes: Obtaining text content of resource elements in the target academic resource; After converting the text content of the resource element into a format text vector using a pre-trained format specification recognition model, judging whether the format text vector meets the format requirements of the specification constraint according to the constraint specification; If so, it is determined that the application of resource elements in the target academic resources complies with the format specification constraints; otherwise, it is determined that the application of resource elements in the target academic resources does not comply with the format specification constraints.

5. The method according to claim 4, characterized in that The step of providing a standardized prompt for the resource elements in the target academic resource that do not conform to the standardized constraints according to the identification result of the resource elements includes: When the application of the resource element in the target academic resource does not comply with the format specification constraint, locating the first text content that does not comply with the format specification constraint in the target academic resource; While providing format specification prompts for the first text content, text content examples with corresponding format requirements for the first text content are displayed. The text content examples are text content extracted from academic resources that conform to format specification constraints.

6. The method according to any one of claims 1 to 5, characterized in that Before performing at least one specification identification on the resource elements in the target academic resources to obtain the identification results of the resource elements, the method further includes: Acquire auxiliary reference resources added historically, wherein the auxiliary reference resources at least include document resources with citation associations to target academic resources in the creation process; Inputting the auxiliary reference resources and academic resource samples of different resource types into a machine learning model, so as to train the mapping relationship of whether the application of resource elements in the academic resource samples meets the reference requirements of the specification constraints according to the reference association through the machine learning model; When the training process meets the iteration stop condition, a reference specification recognition model is generated according to the mapping relationship obtained through training; Accordingly, performing at least one standard identification on the resource elements in the target academic resources to obtain the identification result of the resource elements includes: The pre-trained citation specification recognition model is used to perform citation specification recognition on the resource elements in the target academic resources to obtain citation recognition results of the resource elements.

7. The method according to claim 6, characterized in that The method of using the pre-trained citation norm recognition model to perform citation norm recognition on resource elements in the target academic resource to obtain citation recognition results of the resource elements includes: According to the text content of the resource element in the target academic resource, obtain the reference document resource marked by the text content, wherein the reference document resource is a document resource having a reference association with the text content; After converting the text content of the resource element into a content text vector using a pre-trained citation specification recognition model, judging whether the content text vector meets the citation requirements of the specification constraints according to the similarity result between the content text vector and the text vector of the cited document resource; If so, it is determined that the application of resource elements in the target academic resources complies with the citation standard constraints; otherwise, it is determined that the application of resource elements in the target academic resources does not comply with the citation standard constraints.

8. The method according to claim 6, characterized in that After acquiring the auxiliary reference resource added historically, the method further includes: The auxiliary reference resources and academic resource samples of different resource types are input into a machine learning model, so as to train the mapping relationship between the text content of the resource elements in the academic resource samples and the auxiliary reference resources according to the citation association through the machine learning model; When the training process meets the iteration stop condition, a reference association model is generated according to the mapping relationship obtained through training; Accordingly, providing a standardized prompt for the resource elements in the target academic resource that do not meet the standardized constraints according to the identification result of the resource elements includes: When the application of the resource element in the target academic resource does not comply with the citation standard constraint, locating the second text content that does not comply with the citation standard constraint in the target academic resource; While providing a reference standard prompt for the second text content, a pre-trained reference association model is used to query a first document resource that has a reference association with the second text content; If a citation adjustment instruction for an academic resource is received, the citation mark of the second text content in the target academic resource is adjusted according to the first document resource.

9. The method according to claim 8, characterized in that The method further comprises: Obtain the third text content of unmarked reference literature resources in the target academic resources; While generating a citation prompt for the third text content, a pre-trained citation association model is used to query a second document resource that has a citation association with the third text content; If a citation generation instruction for an academic resource is received, a citation mark for the third text content in the target academic resource is generated based on the second document resource.

10. A standardized processing device for academic resources, characterized in that: include: The first acquisition unit is used to obtain the target academic resources in the creative process; An element identification unit, used for identifying elements in the target academic resource according to the resource type of the target academic resource, to obtain resource elements of the target academic resource; A specification identification unit, used for performing at least one specification identification on the resource elements in the target academic resources to obtain an identification result of the resource elements, wherein the identification result of the resource elements includes whether the application of the resource elements in the target academic resources complies with the specification constraints; A prompting unit is used to provide standard prompts for resource elements in the target academic resources that do not meet standard constraints based on the identification results of the resource elements.

Images