A template export method

By identifying the functional keywords in the template requirement text to match the template database, calculating the overall matching parameters, and selecting the best alternative template or fuzzy similar template, the data filling gap caused by not including target fields in the database is solved, and template recommendations are improved accuracy and flexibility.

CN118586376BActive Publication Date: 2024-11-29北京科杰科技有限公司
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202410688757.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-30
Publication Date
2024-11-29
Estimated Expiration
2044-05-30

AI Technical Summary

Technical Problem

When the prior art directly fills data in the target fields not included in the database, the gap between the feedback results and the data derived from the requirements is too large, and this problem cannot be effectively solved.

Method used

By obtaining the template requirement text entered by the user, identifying functional keywords, and matching with the template name in the template database, calculating the overall matching parameters, determining the matching response category, selecting the optimal alternative template or fuzzy similar template for export, ensuring the recommended results accuracy and flexibility.

Benefits of technology

It realizes accurate judgment of whether the requirements can be directly implemented in the existing database, improves the flexibility and accuracy of template recommendations, saves users time to find suitable templates, and enhances the scope of application of the system and the accuracy of template export.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118586376B_ABST
    Figure CN118586376B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of data processing technology, and in particular to a template export method, comprising: step S1, obtaining a template requirement text input by a user end, and determining a functional keyword based on the part of speech of each keyword in the template requirement text; step S2, matching each keyword in the template requirement text with the keyword in each template name one by one, determining the keyword that matches successfully, and calculating an overall matching parameter based on the matching result; step S3, determining a matching response category for the template requirement text based on the overall matching parameter; step S4, exporting a template in a template database based on the matching response category. The present invention determines a template export method based on the matching result of the template requirement in the database, thereby ensuring that the exported template is obtained quickly and accurately, saving the computing power of the system, reasonably increasing the feedback results of the user end, and facilitating the subsequent updating of the template database.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a template exporting method. Background Art

[0002] The technical tools for customizing PPT templates mainly include PPT design software (such as PowerPoint, Keynote), online template libraries (such as Canva, Slidesgo), image design software (such as Photoshop, Illustrator) and template generation tools. The combination of template automatic generation technology covers template library design algorithms, data-driven generation, intelligent recommendation systems and natural language processing technology. These methods help users quickly generate PPT templates that meet personalized requirements according to their needs, simplifying the creation process and improving efficiency. The continuous development of technology has expanded the range of user choices, making customized PPT templates more creative and professional.

[0003] Chinese Patent Publication No.: CN110059305A discloses a method and device for data export, the method comprising: determining a plurality of data pages to be exported in a system, the plurality of data pages respectively including a plurality of data fields; obtaining a data template to be filled, the data template being pre-configured with a plurality of target fields, the target fields corresponding to the plurality of data fields in the plurality of data pages; extracting target data stored in a plurality of data fields corresponding to the plurality of target fields from the plurality of data pages; filling the target data into the data template to generate a data file; it can be seen that the method and device for data export have the following problems: the target fields not included in the database are not taken into consideration, and the data is directly filled in for feedback, resulting in a large gap between the feedback result and the data required to be exported. Summary of the invention

[0004] To this end, the present invention provides a template export method to overcome the problem in the prior art that, considering the target fields not included in the database, directly filling in data for feedback leads to a large gap between the feedback result and the data to be exported.

[0005] To achieve the above object, the present invention provides a template export method, comprising:

[0006] Step S1, obtaining a template requirement text input by a user terminal, and determining a functional keyword based on the part of speech of each keyword in the template requirement text;

[0007] Step S2, extracting the function keywords in each template name of the template database and matching them with each function keyword of the template requirement text one by one, marking the function keywords that have been matched successfully, and matching each keyword in the template requirement text with each keyword in the template name one by one, marking the keywords that have been matched successfully, and calculating the overall matching parameter based on the matching results;

[0008] Step S3, determining a matching response category for the template requirement text according to the overall matching parameter;

[0009] Step S4, selecting a template in the template database for export according to the matching response category, including:

[0010] Calculate the template matching parameters for a single template based on the marked function keywords, select the best candidate template, determine the supplementary template based on the unmarked keywords in the name of the best candidate template, combine the template page corresponding to the unmarked keywords in the supplementary template with the best candidate template, and export them;

[0011] Or, a number of templates whose names contain the marked function keywords are screened out, a number of optimal fuzzy similar templates are screened out based on the semantic association between the template requirement text and each template name, and each of the optimal fuzzy similar templates is exported.

[0012] Furthermore, in the step S1, determining the functional keywords based on the part of speech of each keyword in the template requirement text includes:

[0013] Determining the parts of speech of all words in the template requirement text;

[0014] Identify nouns, adjectives, and quantifiers as keywords;

[0015] The nouns in the keywords are determined as the functional keywords.

[0016] Furthermore, in step S2, the overall matching parameter is determined according to formula (1):

[0017]

[0018] In formula (1), G is the overall matching parameter, a1 is the total number of nouns of the marked keywords in the template requirement text, A1 is the total number of nouns of the keywords in the template requirement text, b1 is the total number of adjectives of the marked keywords in the template requirement text, B1 is the total number of adjectives of the keywords in the template requirement text, c1 is the total number of quantifiers of the marked keywords in the template requirement text, C1 is the total number of quantifiers of the keywords in the template requirement text, k1 is the noun weight, k2 is the adjective weight, and k3 is the quantifier weight.

[0019] Furthermore, in the step S3, determining the matching response category for the template requirement text according to the overall matching parameter includes:

[0020] If the overall matching parameter is greater than or equal to the preset overall matching parameter, the matching response category of the template requirement text is a strong matching response category;

[0021] If the overall matching parameter is less than the preset overall matching parameter, the matching response category of the template requirement text is a weak matching response category.

[0022] Furthermore, in the step S4, a template in a template database is selected for export according to the matching response category, including:

[0023] If the matching response category of the template requirement text is a strong matching response category, the export process is to calculate the template matching parameters for a single template based on the marked function keywords, screen out the best candidate template, determine the supplementary template based on the unmarked keywords in the name of the best candidate template, combine the template page corresponding to the unmarked keywords in the supplementary template with the best candidate template, and then export;

[0024] If the matching response category of the template requirement text is a weak matching response category, the export process is to screen out several templates whose names contain the marked function keywords, screen out several optimal fuzzy similar templates based on the semantic association between the template requirement text and each template name, and export each of the optimal fuzzy similar templates.

[0025] Furthermore, in step S4, the process of selecting the best candidate template includes:

[0026] Determine the template matching parameters between the marked function keyword and each of the templates in the template database;

[0027] Comparing the template matching parameters of each of the templates with the standard template matching parameters to select a candidate template;

[0028] Among all the candidate templates, the candidate template with the highest template matching parameter is selected as the optimal candidate template.

[0029] Furthermore, in step S4, the template matching parameter is determined according to formula (2):

[0030]

[0031] In formula (2), g is the template matching parameter, a1' is the number of marked functional keywords contained in the noun of the template, and A1' is the total number of functional keywords in the template requirement text.

[0032] Furthermore, the selecting of the candidate template includes:

[0033] If the template matching parameter is greater than or equal to the standard template matching parameter, the current template is selected as the candidate template;

[0034] If the template matching parameter is less than the standard template matching parameter, the current template is selected as a non-alternative template.

[0035] Furthermore, the plurality of optimal fuzzy similar templates selected based on the semantic association between the template requirement text and each template name include:

[0036] Calculating the semantic relevance between the template requirement text and each template name;

[0037] Determining whether the current template is a fuzzy similar template according to the semantic association;

[0038] If the semantic relevance is greater than or equal to the standard semantic relevance, the current template is a fuzzy similarity template;

[0039] Select several fuzzy similar templates from all fuzzy similar templates as the optimal fuzzy similar templates.

[0040] Furthermore, the method further includes step S5, determining whether to add the export template to the template database according to feedback information from the user terminal regarding the export template.

[0041] Compared with the prior art, the beneficial effect of the present invention lies in that the system will intelligently analyze the template requirement text input by the user, determine the functional keywords therein, and match them with the template names in the template database, and determine the overall matching parameters through the matching results. The overall matching parameters can accurately judge whether the content that needs to be implemented by the requirement text can be directly obtained in the existing database; the system selects the most appropriate method to produce and export templates according to the specific situation, making the recommendation process more flexible; the system uses automated recommendations to ensure the accuracy of the recommendation results while saving the time consumed by the user in finding a suitable template.

[0042] Furthermore, the larger the overall matching parameter, the greater the number of functions required to be embodied in the current template requirement text that can be directly realized in the template database. This type of template requirement can be directly obtained in the template database through appropriate screening and combination methods. The template obtained in this way is more in line with user-side requirements. On the contrary, for template requirement texts with smaller overall matching parameters, the number of functions that need to be embodied in this type of template requirement that can be realized in the template database is smaller, and a template that meets the user-side requirements cannot be directly given, and other methods are required to provide a template that relatively meets the requirements. Therefore, the present invention considers the adaptive selection of template export methods for template requirement texts of different matching response categories, thereby improving the scope of application of the system and the accuracy of template export.

[0043] Furthermore, by determining the optimal alternative template that can realize most of the template requirements, and combining it with other parts that have not been realized in the optimal alternative template, one template is used as the template body of the feedback, and other parts are added and supplemented to determine the final feedback template. While ensuring the template requirements of the user end, the system's computing power is saved and the system's template export efficiency and template export accuracy are improved.

[0044] Furthermore, in the present invention, under the condition that a suitable export template cannot be directly obtained by screening and combining weakly matching response categories in the template database, several templates that are closer to the template requirements in the existing templates are selected as export templates by obtaining the semantic association between the template requirement text and each template name. While ensuring the template requirements to the greatest extent, the feedback results on the user side are increased, which facilitates the subsequent updating of the template database. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flow chart of a template export method according to an embodiment of the present invention;

[0046] Figure 2 A logic diagram for determining matching response categories for an embodiment of the present invention;

[0047] Figure 3 A logic diagram for determining candidate templates for an embodiment of the present invention;

[0048] Figure 4 A logical diagram for determining fuzzy similarity templates according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0050] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0051] See also Figure 1 As shown, it is a flow chart of a template export method according to an embodiment of the present invention; the present invention provides a template export method, comprising:

[0052] Step S1, obtaining a template requirement text input by a user terminal, and determining a functional keyword based on the part of speech of each keyword in the template requirement text;

[0053] Step S2, extracting the function keywords in each template name of the template database and matching them with each function keyword of the template requirement text one by one, marking the function keywords that have been matched successfully, and matching each keyword in the template requirement text with each keyword in the template name one by one, marking the keywords that have been matched successfully, and calculating the overall matching parameter based on the matching results;

[0054] Step S3, determining a matching response category for the template requirement text according to the overall matching parameter;

[0055] Step S4, selecting a template in the template database for export according to the matching response category, including:

[0056] Calculate the template matching parameters for a single template based on the marked function keywords, select the best candidate template, determine the supplementary template based on the unmarked keywords in the name of the best candidate template, combine the template page corresponding to the unmarked keywords in the supplementary template with the best candidate template, and export them;

[0057] Or, a number of templates whose names contain the marked function keywords are screened out, a number of optimal fuzzy similar templates are screened out based on the semantic association between the template requirement text and each template name, and each of the optimal fuzzy similar templates is exported.

[0058] In implementation, there is no limitation on the method of identifying words in a text and determining the corresponding part of speech of the words. For example, an existing natural language model can be used to implement the corresponding function, which will not be described in detail.

[0059] It is understandable that the template requirement text is a text description of the requirement input by the user, which will not be described in detail here.

[0060] Specifically, there is no limitation on the construction method of the template database, which can be constructed by pre-collecting a number of PPT templates. It can be understood that each template contains a template name, and the template contains a different number of template pages, which will not be repeated here.

[0061] Specifically, in implementation, the step of matching keywords includes comparing the keywords. If the keywords are the same, it is determined that the match is successful, which will not be described in detail.

[0062] In step S4, the template page corresponding to the unmarked keyword in the supplementary template is combined with the optimal candidate template and then exported, including:

[0063] Retrieving template pages containing unmarked keywords in the supplementary template;

[0064] The template page is combined with the optimal candidate template and then exported.

[0065] In the present invention, the system intelligently analyzes the template requirement text input by the user, determines the functional keywords therein, and matches them with the template names in the template database, determines the overall matching parameters through the matching results, and can accurately judge whether the content that needs to be implemented by the requirement text can be directly obtained in the existing database based on the overall matching parameters; the system selects the most appropriate method to produce and export templates according to the specific situation, making the recommendation process more flexible; the system saves the time consumed by the user in finding a suitable template while ensuring the accuracy of the recommendation results through automated recommendations.

[0066] Specifically, in step S1, determining the functional keywords based on the part of speech of each keyword in the template requirement text includes:

[0067] Determining the parts of speech of all words in the template requirement text;

[0068] Identify nouns, adjectives, and quantifiers as keywords;

[0069] The nouns in the keywords are determined as the functional keywords.

[0070] It can be understood that for template requirements, the noun of the requirement is the main part of the requirement, and the adjectives and quantifiers are the limitations of the main part. Therefore, the noun part of all keywords is defined as functional keywords, and the exported template with nouns as functional keywords is more representative for realizing the template requirements.

[0071] In implementation, for the convenience of analysis, the number of words included in the template requirement text should not be too small. Preferably, the number of words included in the template requirement text should be greater than a predetermined number threshold, and the number threshold is in the range [10,20].

[0072] See also Figure 2 As shown, it is a logic diagram for determining the matching response category in an embodiment of the present invention. In step S2, the overall matching parameter is determined according to formula (1):

[0073]

[0074] In formula (1), G is the overall matching parameter, a1 is the total number of nouns of the marked keywords in the template requirement text, A1 is the total number of nouns of the keywords in the template requirement text, b1 is the total number of adjectives of the marked keywords in the template requirement text, B1 is the total number of adjectives of the keywords in the template requirement text, c1 is the total number of quantifiers of the marked keywords in the template requirement text, C1 is the total number of quantifiers of the keywords in the template requirement text, k1 is the noun weight, k2 is the adjective weight, and k3 is the quantifier weight.

[0075] In implementation, k1=0.7, k2=0.15, k3=0.15.

[0076] If b1 or c1 is 0, k1=0.85; if b1=c1=0, k1=1.

[0077] The total number of keywords of each part of speech can be directly counted during the matching process.

[0078] It is understandable that the noun weight only increases and does not decrease in order to ensure the accuracy of the main body of the derived template.

[0079] Specifically, in step S3, determining the matching response category for the template requirement text according to the overall matching parameter includes:

[0080] If the overall matching parameter is greater than or equal to the preset overall matching parameter, the matching response category of the template requirement text is a strong matching response category;

[0081] If the overall matching parameter is less than the preset overall matching parameter, the matching response category of the template requirement text is a weak matching response category.

[0082] In implementation, the preset overall matching parameter is selected in the interval [0.8, 0.9]. Generally, the fewer the total number of keywords in the template requirement text, the larger the value of the preset overall matching parameter.

[0083] Specifically, in step S4, a template in a template database is selected for export according to the matching response category, including:

[0084] If the matching response category of the template requirement text is a strong matching response category, the export process is to calculate the template matching parameters for a single template based on the marked function keywords, screen out the best candidate template, determine the supplementary template based on the unmarked keywords in the name of the best candidate template, combine the template page corresponding to the unmarked keywords in the supplementary template with the best candidate template, and then export;

[0085] If the matching response category of the template requirement text is a weak matching response category, the export process is to screen out several templates whose names contain the marked function keywords, screen out several optimal fuzzy similar templates based on the semantic association between the template requirement text and each template name, and export each of the optimal fuzzy similar templates.

[0086] In the present invention, the larger the overall matching parameter, the greater the number of functions required to be embodied in the current template requirement text that can be directly realized in the template database. This type of template requirement can be directly obtained in the template database through appropriate screening and combination methods. The template obtained in this way is more in line with the user-side needs. On the contrary, for template requirement texts with smaller overall matching parameters, the number of functions that need to be embodied in this type of template requirement that can be realized in the template database is smaller, and a template that meets the user-side needs cannot be directly given, and other methods are required to provide a template that relatively meets the needs. Therefore, the present invention considers the adaptive selection of template export methods for template requirement texts of different matching response categories, thereby improving the scope of application of the system and the accuracy of template export.

[0087] See also Figure 3 As shown, it is a logic diagram for determining candidate templates in an embodiment of the present invention. In the step S4, the screening of the optimal candidate template includes:

[0088] Determine the template matching parameters between the marked function keyword and each of the templates in the template database;

[0089] Comparing the template matching parameters of each of the templates with the standard template matching parameters to select a candidate template;

[0090] Among all the candidate templates, the candidate template with the highest template matching parameter is selected as the optimal candidate template.

[0091] In implementation, if the number of candidate templates with the highest template matching parameters is not unique, these candidate templates are deemed to be able to embody the functions required in the template requirement text with the same effect, and any one of them is selected when determining the optimal candidate template.

[0092] Specifically, in step S4, the template matching parameter is determined according to formula (2):

[0093]

[0094] In formula (2), g is the template matching parameter, a1' is the number of marked functional keywords contained in the noun of the template, and A1' is the total number of functional keywords in the template requirement text.

[0095] Specifically, the selecting of the candidate template includes:

[0096] If the template matching parameter is greater than or equal to the standard template matching parameter, the current template is selected as the candidate template;

[0097] If the template matching parameter is less than the standard template matching parameter, the current template is selected as a non-alternative template.

[0098] In the implementation, the standard template matching parameters are selected in the interval [0.65, 0.75].

[0099] In the present invention, the optimal alternative template that can realize most of the template requirements is determined, and it is combined with other parts that cannot be realized in the optimal alternative template. One template is used as the template body of the feedback, and other parts are added and supplemented to determine the final feedback template. While ensuring the template requirements of the user end, the computing power of the system is saved, and the template export efficiency and accuracy of the system are improved.

[0100] See also Figure 4 As shown, it is a logic diagram for determining fuzzy similar templates in an embodiment of the present invention, wherein the optimal fuzzy similar templates are selected based on the semantic association between the template requirement text and each template name, including:

[0101] Calculating the semantic relevance between the template requirement text and each template name;

[0102] Determining whether the current template is a fuzzy similar template according to the semantic association;

[0103] If the semantic relevance is greater than or equal to the standard semantic relevance, the current template is a fuzzy similarity template;

[0104] If the semantic relevance is less than the standard semantic relevance, the current template is a non-fuzzy similar template;

[0105] Select several fuzzy similar templates from all fuzzy similar templates as the optimal fuzzy similar templates.

[0106] In implementation, determining the semantic correlation between the template requirement text and each template name is an existing technology related to language information processing, which can be obtained by any semantic correlation calculation method such as cosine similarity, Jaccard similarity, Euclidean distance, etc. This is an existing technology and will not be repeated here.

[0107] The standard semantic relevance is selected in the interval [0.55, 0.75]. Generally, the longer the template requirement text is, the larger the value of the standard semantic relevance is.

[0108] In the present invention, under the condition that a suitable export template cannot be directly obtained by screening and combining weakly matching response categories in the template database, several templates that are closer to the template requirements in the existing templates are selected as export templates by obtaining the semantic association between the template requirement text and each template name. While ensuring the template requirements to the greatest extent, the feedback results on the user side are increased, which facilitates the subsequent updating of the template database.

[0109] Specifically, the method further includes step S5, determining whether to add the export template to the template database according to feedback information from the user terminal regarding the export template.

[0110] In implementation, the feedback information from the user side includes whether the best candidate template combination needs to be added as a new existing template.

[0111] If the feedback information indicates that the optimal candidate template combination needs to be added as a new existing template, the optimal candidate template combination is added as a new existing template.

[0112] If the template export method of the present invention is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0113] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0114] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A template export method, characterized in that: include: Step S1, obtaining a template requirement text input by a user terminal, and determining a functional keyword based on the part of speech of each keyword in the template requirement text; Step S2, extracting the function keywords in each template name of the template database and matching them with each function keyword of the template requirement text one by one, marking the function keywords that have been matched successfully, and matching each keyword in the template requirement text with each keyword in the template name one by one, marking the keywords that have been matched successfully, and calculating the overall matching parameter based on the matching results; Step S3, determining a matching response category for the template requirement text according to the overall matching parameter; Step S4, selecting a template in the template database for export according to the matching response category, including: Calculate the template matching parameters for a single template based on the marked function keywords, select the best candidate template, determine the supplementary template based on the unmarked keywords in the name of the best candidate template, combine the template page corresponding to the unmarked keywords in the supplementary template with the best candidate template, and export them; Or, a number of templates whose names contain the marked function keywords are screened out, a number of optimal fuzzy similar templates are screened out based on the semantic association between the template requirement text and each template name, and each of the optimal fuzzy similar templates is exported; In step S2, the overall matching parameter is determined according to formula (1): In formula (1), G is the overall matching parameter, a1 is the total number of nouns of the marked keywords in the template demand text, A1 is the total number of nouns of the keywords in the template demand text, b1 is the total number of adjectives of the marked keywords in the template demand text, B1 is the total number of adjectives of the keywords in the template demand text, c1 is the total number of quantifiers of the marked keywords in the template demand text, C1 is the total number of quantifiers of the keywords in the template demand text, k1 is the noun weight, k2 is the adjective weight, and k3 is the quantifier weight; In the step S3, determining the matching response category for the template requirement text according to the overall matching parameter includes: If the overall matching parameter is greater than or equal to the preset overall matching parameter, the matching response category of the template requirement text is a strong matching response category; If the overall matching parameter is less than the preset overall matching parameter, the matching response category of the template requirement text is a weak matching response category; In the step S4, a template in a template database is selected for export according to the matching response category, including: If the matching response category of the template requirement text is a strong matching response category, the export process is to calculate the template matching parameters for a single template based on the marked function keywords, screen out the best candidate template, determine the supplementary template based on the unmarked keywords in the name of the best candidate template, combine the template page corresponding to the unmarked keywords in the supplementary template with the best candidate template, and then export; If the matching response category of the template requirement text is a weak matching response category, the export process is to screen out several templates whose names contain the marked function keywords, screen out several optimal fuzzy similar templates based on the semantic association between the template requirement text and each template name, and export each of the optimal fuzzy similar templates.

2. The template export method according to claim 1, characterized in that: In the step S1, determining the functional keywords based on the part of speech of each keyword in the template requirement text includes: Determining the parts of speech of all words in the template requirement text; Identify nouns, adjectives, and quantifiers as keywords; The nouns in the keywords are determined as the functional keywords.

3. The template export method according to claim 1, characterized in that: In step S4, the process of selecting the best candidate template includes: Determine the template matching parameters between the marked function keyword and each of the templates in the template database; Comparing the template matching parameters of each of the templates with the standard template matching parameters to select a candidate template; Among all the candidate templates, the candidate template with the highest template matching parameter is selected as the optimal candidate template.

4. The template export method according to claim 3, characterized in that: In step S4, the template matching parameter is determined according to formula (2): In formula (2), g is the template matching parameter, a1' is the number of marked functional keywords contained in the noun of the template, and A1' is the total number of functional keywords in the template requirement text.

5. The template export method according to claim 4, characterized in that: The selecting of the candidate template comprises: If the template matching parameter is greater than or equal to the standard template matching parameter, the current template is selected as the candidate template; If the template matching parameter is less than the standard template matching parameter, the current template is selected as a non-alternative template.

6. The template export method according to claim 5, characterized in that: The method of selecting a number of optimal fuzzy similar templates based on the semantic association between the template requirement text and each template name includes: Calculating the semantic relevance between the template requirement text and each template name; Determining whether the current template is a fuzzy similar template according to the semantic association; If the semantic relevance is greater than or equal to the standard semantic relevance, the current template is a fuzzy similarity template; Select several fuzzy similar templates from all fuzzy similar templates as the optimal fuzzy similar templates.

7. The template export method according to claim 1, characterized in that: The method further includes step S5, determining whether to add the export template to the template database according to feedback information from the user terminal regarding the export template.

Citation Information

Patent Citations

  • Data exporting method and device

    CN110059305A

  • Keyword extracting method and device

    CN105426360A

  • Search word processing method and device and computer readable storage medium

    CN110019738A