CN116521159B - Knowledge service platform zero code construction method and system based on scene driving - Google Patents
Knowledge service platform zero code construction method and system based on scene driving Download PDFInfo
- Publication number
- CN116521159B CN116521159B CN202310813982.5A CN202310813982A CN116521159B CN 116521159 B CN116521159 B CN 116521159B CN 202310813982 A CN202310813982 A CN 202310813982A CN 116521159 B CN116521159 B CN 116521159B
- Authority
- CN
- China
- Prior art keywords
- scene
- feature
- disassembly
- user
- application template
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000010276 construction Methods 0.000 title claims abstract description 25
- 238000012216 screening Methods 0.000 claims abstract description 56
- 238000000034 method Methods 0.000 claims abstract description 21
- 238000004458 analytical method Methods 0.000 claims abstract description 14
- 230000000007 visual effect Effects 0.000 claims abstract description 12
- 239000011159 matrix material Substances 0.000 claims description 67
- 239000013598 vector Substances 0.000 claims description 46
- 238000011156 evaluation Methods 0.000 claims description 19
- 238000012937 correction Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/30—Creation or generation of source code
- G06F8/36—Software reuse
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/10—Requirements analysis; Specification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W90/00—Enabling technologies or technologies with a potential or indirect contribution to greenhouse gas [GHG] emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides a knowledge service platform zero code construction method and system based on scene driving, which relate to the technical field of scene analysis, and the method comprises the following steps: creating a scene application template list according to scene demand information of a first user, screening the scene application template list by carrying out key demand feature disassembly on the scene demand information of the first user, generating component feedback screening features according to first user component feedback information, carrying out secondary screening on the application template list with a key feature disassembly result, accessing a thematic field knowledge base through an external interface, carrying out visual display loading on the first scene application template to generate a target knowledge service platform, solving the technical problems that in the prior art, because of more thematic knowledge data sources, the user consumes a long time when using a knowledge service platform, the efficiency is low, and realizing the aim of aiming at analysis of the user by adopting a zero code platform aiming at different scenes.
Description
Technical Field
The application relates to the technical field of scene analysis, in particular to a knowledge service platform zero code construction method and system based on scene driving.
Background
The development condition of the current subject knowledge service platform is determined by researching the deep application modes of technologies such as big data, artificial intelligence and the like in subject precise service, the method focuses on the application in aspects such as precise identification, mass data resource collection and convergence, deep knowledge mining and organization, precise pushing and personalized recommendation and the like of user demands, a subject knowledge service platform can be quickly and precisely constructed by combining a subject expert knowledge with a visual mode, a zero code development mode, application and management of subject service, automatic selection of a subject data set, manual management of the subject data, configuration and arrangement of a subject website by an integrated calculation and service assembly, release and management of a subject public knowledge service platform and the like.
In the prior art, because of more thematic knowledge data sources, the time consumption of a user when using and constructing a knowledge service platform is more, and the technical problem of low efficiency is caused.
Disclosure of Invention
The application provides a knowledge service platform zero code construction method and system based on scene driving, which are used for solving the technical problems of low efficiency caused by more consumption of time when a user builds a knowledge service platform due to more thematic knowledge data sources in the prior art.
In view of the above problems, the application provides a knowledge service platform zero code construction method and system based on scene driving.
In a first aspect, the present application provides a method for constructing a zero code of a knowledge service platform based on scene driving, the method comprising: acquiring scene demand information of a first user, and creating a scene application template list according to the scene demand information of the first user; the key demand feature disassembly is carried out on the scene demand information of the first user, so that a key feature disassembly result is obtained; screening the scene application template list according to the key feature disassembly result to obtain an associated application template list; the associated application template list is sent to the first user, and component feedback screening characteristics are generated according to the component feedback information of the first user; performing secondary screening on the application template list based on the component feedback screening characteristics and the key characteristic dismantling results, and outputting a first scene application template; and accessing an external interface into a thematic field knowledge base, and carrying out visual display loading on the first scene application template to generate a target knowledge service platform.
In a second aspect, the present application provides a knowledge service platform zero code construction system based on scene driving, the system comprising: the system comprises a list creation module, a scene application template module and a scene application template module, wherein the list creation module is used for acquiring scene demand information of a first user and creating a scene application template list according to the scene demand information of the first user; the first feature disassembly module is used for obtaining a key feature disassembly result by carrying out key requirement feature disassembly on the scene requirement information of the first user; the first screening module is used for screening the scene application template list according to the key feature dismantling result to obtain an associated application template list; the feature generation module is used for sending the associated application template list to the first user and generating component feedback screening features according to the component feedback information of the first user; the second screening module is used for carrying out secondary screening on the application template list based on the component feedback screening characteristics and the key characteristic dismantling results, and outputting a first scene application template; the loading module is used for carrying out visual display loading on the first scene application template through accessing the external interface into the thematic field knowledge base, and generating a target knowledge service platform.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
the application provides a knowledge service platform zero code construction method and system based on scene driving, relates to the technical field of scene analysis, solves the technical problem of low efficiency caused by more time consumption of a user when the knowledge service platform is constructed due to more thematic knowledge data sources in the prior art, realizes targeted analysis of the user by adopting a zero code platform aiming at different scenes, provides high-efficiency construction of a multi-scene application platform, and develops or optimizes platform functions and services accurately according to requirements and application feedback.
Drawings
FIG. 1 is a schematic flow diagram of a knowledge service platform zero code construction method based on scene driving;
FIG. 2 is a schematic flow chart of a key feature disassembly result generated in a knowledge service platform zero code construction method based on scene driving;
FIG. 3 is a schematic flow chart of a related application template list obtained in a knowledge service platform zero code construction method based on scene driving;
FIG. 4 is a schematic diagram of a feature flow for feedback screening of a generating component in a knowledge service platform zero code construction method based on scene driving;
fig. 5 is a schematic diagram of a system architecture for constructing a zero code of a knowledge service platform based on scene driving.
Reference numerals illustrate: the system comprises a list creation module 1, a first feature disassembly module 2, a first screening module 3, a feature generation module 4, a second screening module 5 and a loading module 6.
Detailed Description
The application provides a knowledge service platform zero code construction method and system based on scene driving, which are used for solving the technical problem of low efficiency caused by more time consumption of a user when the knowledge service platform is built due to more thematic knowledge data sources in the prior art.
Example 1
As shown in fig. 1, an embodiment of the present application provides a method for constructing a zero code of a knowledge service platform based on scene driving, which includes:
step S100: acquiring scene demand information of a first user, and creating a scene application template list according to the scene demand information of the first user;
specifically, the method for constructing the zero code of the knowledge service platform based on the scene driving is applied to a system for constructing the zero code of the knowledge service platform based on the scene driving, so that in order to ensure the accuracy of a target knowledge service platform generated in the later stage, the scene demand information of a first user needs to be acquired, the first user selects any one target user from the knowledge platform, further, the demand information of the first user can be uploaded to the system according to an external interface in the system, the demand information of the first user can comprise demand characteristics such as an application field, a function point, a solution problem and the like, in order to establish an application scene of a knowledge system based on a low-code platform, the scene demand information of the first user can comprise the application field of the knowledge scene and a function point for mainly solving the problem, and further, a scene application template list corresponding to the scene demand information of the first user is created based on the scene demand information of the first user.
The scene application template list is based on scene application templates screened out from all scene application templates of the system, the scene application template list can be a table header taking scene demand information of a first user as a horizontal axis and taking an actual application scene of the scene demand information of the first user as a vertical axis, and the scene application template list is built, so that screened scene application targets are filled into the created scene application template list to complete creation of the scene application template list, and a knowledge service platform based on scene driving is built for later realization as an important reference basis.
Step S200: the key demand feature disassembly is carried out on the scene demand information of the first user, so that a key feature disassembly result is obtained;
further, as shown in fig. 2, step S200 of the present application further includes:
step S210: inputting the scene demand information of the first user into a feature disassembly model, wherein the feature disassembly model comprises a first disassembly branch and a second disassembly branch, the first disassembly branch performs feature disassembly according to scene relevance, and the second disassembly branch performs feature disassembly according to user demand custom degree;
step S220: analyzing the scene demand information of the first user according to the first disassembly branch, and outputting a first disassembly feature set comprising scene relevance grade identifiers;
step S230: analyzing scene demand information of the first user according to the second dismantling branch, and outputting a second dismantling feature set comprising a user-defined demand level identifier;
step S240: and generating a key feature disassembly result based on the first disassembly feature set and the second disassembly feature set.
Specifically, for accurately completing the construction of the target knowledge service platform, the key demand feature disassembly is needed to be carried out on the obtained scene demand information of the first user, namely the scene demand information of the first user is firstly input into a feature disassembly model, the feature disassembly model comprises a first disassembly branch and a second disassembly branch, wherein the first disassembly branch carries out feature disassembly on the scene demand information of the first user according to the scene relevance, namely the feature disassembly is carried out on the scene demand information of the first user through the feature correlation grade of the self-combination direction of the user, and the second disassembly branch carries out feature disassembly on the scene demand information of the first user according to the user demand custom degree, namely the feature disassembly is carried out on the scene demand information of the first user through the custom grade of the demand information input.
Further, the first disassembly branch in the feature disassembly model is used for carrying out disassembly analysis on the scene demand information of the first user, namely, the first user is disassembled according to the relevance between the demand scene of the first user and the current scene contained in the scene demand information of the first user, namely, when the relevance is lower than 30%, the first user is disassembled, further, the first disassembly feature set with the scene relevance grade identification is obtained according to the disassembly result, the scene relevance grade is classified into a first grade, a second grade and a third grade, the first grade is the grade with highest scene relevance, the second disassembly branch in the feature disassembly model is used for carrying out disassembly analysis on the scene demand information of the first user, namely, the first user is disassembled according to the self-defined scene demand of the first user, and if the self-defined scene demand of the first user is an electrically-related scene, the first user is extracted and disassembled, further, the second feature set with the self-defined demand grade identification is output, and the self-defined demand grade can be classified into the first grade, the second grade and the self-defined scene demand grade is the highest, and the self-defined scene demand grade is the first user demand grade.
And finally, generating a key feature disassembly result on the basis of the output first disassembly feature set and the second disassembly feature set, wherein the key feature disassembly result refers to that an overlapped disassembly feature part of the first disassembly feature set and the second disassembly feature set is marked as a key disassembly feature to be output, so that the construction of a knowledge service platform based on scene driving is guaranteed.
Step S300: screening the scene application template list according to the key feature disassembly result to obtain an associated application template list;
further, as shown in fig. 3, step S300 of the present application further includes:
step S310: generating a first feature identification matrix according to the first disassembled feature set with the scene relevance grade identification;
step S320: generating a second feature identification matrix according to the second disassembled feature set with the user-defined requirement level identification;
step S330: performing matrix calculation according to the first characteristic identification matrix and the second characteristic identification matrix, multiplying the overlapping vectors in the first characteristic identification matrix and the second characteristic identification matrix, and outputting a third characteristic identification matrix;
step S340: and screening the scene application template list according to the third characteristic identification matrix to obtain an associated application template list.
Specifically, in order to construct a scene application template list attached to the scene demand information of the first user, the first method includes screening the scene application template list according to the key feature dismantling result obtained by dismantling the key demand features, that is, first dismantling feature sets with scene relevance grade identifiers are used as horizontal axes, dismantling feature filling with different scene relevance grade identifiers is performed on the first dismantling feature sets, an n×n matrix is established as a first feature identifier matrix, n is a positive integer, a second dismantling feature set with custom demand grade identifiers is used as a vertical axis, dismantling feature filling with different custom demand grade identifiers is performed on the second dismantling feature sets, an m×m matrix is established as a second feature identifier matrix, m is a positive integer, matrix calculation is performed according to the first feature identification matrix and the second feature identification matrix, namely, matrix fusion is performed on a first disassembled feature set serving as a horizontal axis and a second disassembled feature set serving as a vertical axis, namely, the structures of rectangular matrixes on a finite field under the group action are combined, matrix fusion is completed according to primitive, dual, P polynomial and other basic properties contained in the first feature identification matrix and the second feature identification matrix and from the homogeneous group, further, overlapping vectors in the first feature identification matrix and the second feature identification matrix are multiplied, the overlapping vectors refer to vectors of current feature data existing in the first feature identification matrix and the second feature identification matrix, and meanwhile, data filling is performed after the obtained overlapping vectors are multiplied by a matrix, wherein a is a positive integer, and the matrix with the data filled is marked as a third characteristic identification matrix.
Finally, taking the third feature identification matrix as a basis, screening the scene application templates in the scene application template list, namely performing traversal matching on the scene application templates in the scene application template list and vector data in the third feature identification matrix, screening and extracting traversal matching templates with data similarity higher than 85%, and meanwhile, after all the screened and extracted templates are summarized, arranging in a descending order according to the association degree and marking the templates as an associated application template list for outputting, and constructing a ramming basis for a subsequent implementation on a knowledge service platform based on scene driving.
Further, step S340 of the present application includes:
step S341: initializing key feature combination vectors, wherein the key feature combination vectors at least comprise two key feature vectors;
step S342: generating a search space according to the vector in the third feature identification matrix;
step S343: and carrying out template searching and predicting by taking the initialized key feature combination vector as an initial variable, obtaining a searching and predicting result, and carrying out combination change on the key feature combination vector based on the searching space when the searching and predicting result return value is empty until the searching and predicting result return value is not empty.
Further, step S341 of the present application includes:
step S3411: n key features randomly selected based on the first feature identification matrix, wherein the scene relevance levels of the N key features are all larger than the preset scene relevance, and N is a positive integer larger than or equal to 0;
step S3412: m key features randomly selected based on the second feature identification matrix, wherein the custom demand level of the M key features is larger than the preset demand level, and M is a positive integer larger than or equal to 0;
step S3413: and obtaining an initialization key feature combination vector by using the M key features and the N key features.
Specifically, in order to improve the search accuracy when key feature combination is performed in the constructed knowledge platform, an initialized key feature combination vector corresponding to the key features needs to be acquired, wherein the key feature combination vector at least comprises two key feature vectors, random selection is performed in the constructed first feature identification matrix, N key features are randomly selected, N is a positive integer greater than or equal to 0, wherein the scene relevance levels of the N key features are all greater than the preset scene relevance, the preset scene relevance is preset by relevant technicians according to scene relevance data in big data, the scene relevance levels of the N key features are all greater than the preset scene relevance, namely, key features with small scene relevance in a first feature identification matrix are screened out, random selection is performed in the constructed second feature identification matrix, M key features are randomly selected, M is a positive integer greater than or equal to 0, the self-defined demand levels of the M key features are all greater than the preset demand levels, the preset demand levels are all preset by relevant technicians according to the preset scene relevance data in big data, the scene relevance levels of the N key features are all greater than the preset scene relevance data, the key features with small scene relevance is selected in the first feature identification matrix, the first feature identification matrix is selected randomly, the M key features are selected, the self-defined by the user demand levels are combined according to the preset user preference levels, and the first feature level is further defined, and the key feature is combined with the preset in the first feature level is lower than the preset level is lower than the key feature level is defined by the preset, and is lower than the key feature level is better than the user.
Further, defining a search space range of data according to the data vectors in the third feature identification matrix, namely, arranging overlapping data vectors contained in the third feature identification matrix in a descending order, taking the data vector with the first order as an upper boundary of the search space, taking the data vector with the last order as a lower boundary of the search space, generating the search space, carrying out long template search prediction in the search space by taking the initialized key feature combination vector as an initial variable, obtaining a search prediction result, and when a return value of the search prediction result is empty, considering that the currently initialized key feature combination vector is in an absence state in the search space, so that combination change is required to be carried out on the key feature combination vector in the range of the search space, namely, re-extracting and combining data in the key feature combination vector in M key features and N key features, and then harvesting the re-combined key feature combination vector in the search space again, so that the key feature combination vector is output until the return value of the search prediction result is not empty, and realizing the function of defining a service platform based on a scene.
Further, step S340 of the present application includes:
step S344: building a template recommendation recognition model, wherein the template recommendation recognition model comprises a component recommendation index, an application rate recommendation index and an update recommendation index;
step S345: connecting the template recommendation recognition model with the associated application template list, and outputting recommendation index lists respectively corresponding to the associated application template list according to the template recommendation recognition model;
step S346: and recommending the associated application template list according to the recommendation index list.
In order to ensure the accuracy of the association application template list recommended by the knowledge platform, the template recommendation recognition model is firstly required to be built, namely the requirement information of the first user is taken as input data, wherein the template recommendation recognition model comprises a component recommendation index, an application rate recommendation index and an update recommendation index, the component recommendation index refers to functions of each component, the application rate recommendation index refers to the number of times each association application template is applied, the update recommendation index refers to whether the current association application template is a new updated template or not, the template recommendation recognition model is further connected with the association application template list, the template recommendation recognition model is used for outputting the most relevant association application template according to the relevance, and simultaneously, the recommendation index corresponding to the association application template is obtained, the association application template carries recommendation display information of each association application template, further, the component recommendation index, the application rate recommendation index and the update recommendation index are taken as bases, accordingly, the recommendation index list corresponding to the association application template list is output, screening is carried out on the recommendation index list, the association application index is required to be disassembled based on a first layer of association class identification in the requirement information of the first user, and the association class identification is a new association class identification, association class is used as a recommendation index, and association class is used as a final association class of association class, and association class is used for carrying out association class analysis when the association class is required to be disassembled, and association class-level is used as a final association class index.
Step S400: the associated application template list is sent to the first user, and component feedback screening characteristics are generated according to the component feedback information of the first user;
further, as shown in fig. 4, step S400 of the present application further includes:
step S410: storing the component feedback information of the first user to a feedback storage module, and calling all the component feedback information in the feedback storage module to analyze to obtain negative feedback quantity distribution density, negative feedback frequency distribution density and feedback proportion of positive feedback and negative feedback;
step S420: generating an evaluation index set of each component object according to the negative feedback quantity distribution density, the negative feedback frequency distribution density and the feedback proportion of positive feedback and negative feedback;
step S430: acquiring scene application fields of components corresponding to all evaluation indexes in the evaluation index set;
step S440: regularizing the evaluation index set based on the scene application field, and judging whether to activate correction conditions according to regularized results;
step S450: and if the correction condition is activated, generating a component feedback screening feature according to the regularization processing result.
Specifically, the output relevant application template list is used as sending data, the sending data are sent to a first user, meanwhile, the building feedback screening feature is output according to the building feedback information fed back by the first user after the first user receives the building feedback information, namely, the building feedback information of the first user is stored in a feedback storage module, after the relevant application template list is sent to the first user, the first user requests the relevant application template list, all the building feedback information in the feedback storage module is further called and analyzed, the negative feedback quantity distribution density, the negative feedback frequency distribution density and the feedback proportion of positive feedback and negative feedback of the first user are obtained, and if the first user requests the relevant application template list, the negative feedback quantity is higher as the first user requests the relevant application template list, the negative feedback frequency is higher, the positive feedback and the negative feedback proportion is larger, and further, the negative feedback quantity distribution density and the negative feedback frequency distribution and the negative feedback proportion are larger as the first user requests the relevant application template list, the negative feedback quantity distribution density and the negative feedback frequency are inversely related to the negative feedback quantity and the negative feedback frequency, and the negative feedback index is larger than the evaluation object is generated, and the evaluation object is set.
Further, each evaluation index in the evaluation index set is collected through regularization processing, the regularization processing refers to a processing means for introducing the scene application field of the corresponding component to each evaluation index in the evaluation index set at this time so as to prevent overfitting and improve the generalization performance of the index, the scene field further comprises the rareness degree of the scene, and if the number of times of application of one scene is high, and the scene is exemplarily included in the evaluation index, the judgment is also carried out, meanwhile, according to the regularization processing result, whether the correction condition is activated is judged, when the correction condition is activated, the number of times of application of the scene is high, but the rare degree of the scene is high, finally, after the corresponding correction is carried out on the scene through the activated correction condition, the generation of feedback screening characteristics of the component is finally completed based on the regularization processing result, and the construction accuracy of the knowledge service platform based on scene driving is improved.
Step S500: performing secondary screening on the application template list based on the component feedback screening characteristics and the key characteristic dismantling results, and outputting a first scene application template;
specifically, in order to ensure the accuracy of selecting a scene application template in a knowledge platform, then, in the case of taking a component feedback screening feature generated according to component feedback information of a first user and a key feature dismantling result obtained by carrying out key requirement feature dismantling on the scene requirement information of the first user as screening references, carrying out secondary screening on an associated application template list obtained by screening a scene application template list according to the key feature dismantling result, namely, matching the associated application template through the component positive feedback feature, the component negative feedback feature, the application scene feature and the key requirement feature of the first user contained in the component feedback screening feature, and carrying out template screening on the template with the matching degree lower than 80%, and integrating templates with the matching degree higher than or equal to 80% to be recorded as the first scene application template to output, thereby ensuring that a knowledge service platform based on scene driving is better constructed in the later stage.
Step S600: and accessing an external interface into a thematic field knowledge base, and carrying out visual display loading on the first scene application template to generate a target knowledge service platform.
Specifically, in order to enable the target knowledge service platform to be built with high efficiency, firstly, an external interface and a thematic domain knowledge base are required to be accessed, the external interface is an interface for connecting a peripheral network of the thematic domain knowledge base, the thematic domain knowledge base is an interface for connecting theoretical knowledge and fact data related to the thematic domain, heuristic knowledge obtained by expert experience, such as definition, theorem, algorithm, common sense knowledge and the like related to the thematic domain, on the basis of the heuristic knowledge, the first scene application template is subjected to visual display loading, the carried out visual display loading is to carry out data driving assembly and data rendering, in the loading process, corresponding assembly data setting is carried out according to the data types to be displayed by the assembly, then the assembly after the data setting is rendered to obtain a visual screen interface, and finally, the displayed visual screen interface is used as a service of the target knowledge service platform for carrying out thematic knowledge on the first user, so that the construction of the scene driving knowledge service platform is completed based on the target knowledge service platform.
In summary, the method for constructing the zero code of the knowledge service platform based on the scene driving provided by the embodiment of the application at least comprises the following technical effects that the user is subjected to targeted analysis by adopting the zero code platform aiming at different scenes, the high-efficiency construction of the multi-scene application platform is provided, and the platform functions and services are developed or optimized accurately according to the requirements and the application feedback.
Example two
Based on the same inventive concept as the knowledge service platform zero code construction method based on scene driving in the foregoing embodiment, as shown in fig. 5, the present application provides a knowledge service platform zero code construction system based on scene driving, the system includes:
the system comprises a list creation module 1, wherein the list creation module 1 is used for acquiring scene demand information of a first user and creating a scene application template list according to the scene demand information of the first user;
the first feature disassembly module 2 is used for obtaining a key feature disassembly result by carrying out key requirement feature disassembly on the scene requirement information of the first user;
the first screening module 3 is used for screening the scene application template list according to the key feature dismantling result to obtain an associated application template list;
the feature generation module 4 is configured to send the associated application template list to the first user, and generate a component feedback screening feature according to component feedback information of the first user;
the second screening module 5 is used for carrying out secondary screening on the application template list based on the component feedback screening characteristics and the key characteristic dismantling results, and outputting a first scene application template;
the loading module 6 is used for carrying out visual display loading on the first scene application template through accessing the external interface into the thematic field knowledge base, and generating a target knowledge service platform.
Further, the system further comprises:
the second feature disassembly module is used for inputting the scene demand information of the first user into a feature disassembly model, wherein the feature disassembly model comprises a first disassembly branch and a second disassembly branch, the first disassembly branch performs feature disassembly according to scene relevance, and the second disassembly branch performs feature disassembly according to user demand custom degree;
the first analysis module is used for analyzing the scene demand information of the first user according to the first disassembly branch and outputting a first disassembly feature set comprising scene relevance grade identification;
the third analysis module is used for analyzing the scene demand information of the first user according to the second disassembly branch and outputting a second disassembly feature set comprising a user-defined demand level identifier;
the result generation module is used for generating key feature disassembly results based on the first disassembly feature set and the second disassembly feature set.
Further, the system further comprises:
the first matrix module is used for generating a first feature identification matrix according to the first disassembled feature set with the scene relevance grade identification;
the second matrix module is used for generating a second characteristic identification matrix according to the second disassembled characteristic set with the user-defined requirement level identification;
the third matrix module is used for performing matrix calculation according to the first characteristic identification matrix and the second characteristic identification matrix, multiplying the overlapped vectors in the first characteristic identification matrix and the second characteristic identification matrix, and outputting a third characteristic identification matrix;
and the third screening module is used for screening the scene application template list according to the third characteristic identification matrix to obtain an associated application template list.
Further, the system further comprises:
a first vector module for initializing a key feature combination vector, wherein the key feature combination vector comprises at least two key feature vectors;
the second vector module is used for generating a search space according to the vectors in the third feature identification matrix;
and the combination changing module is used for carrying out template searching and predicting by taking the initialized key feature combination vector as an initial variable to obtain a searching and predicting result, and carrying out combination changing on the key feature combination vector based on the searching space when the searching and predicting result return value is empty until the searching and predicting result return value is not empty.
Further, the system further comprises:
the first random selection module is used for randomly selecting N key features based on the first feature identification matrix, wherein the scene relevance levels of the N key features are all larger than the preset scene relevance, and N is a positive integer larger than or equal to 0;
the second random selection module is used for randomly selecting M key features based on the second feature identification matrix, wherein the custom demand level of the M key features is larger than the preset demand level, and M is a positive integer larger than or equal to 0;
and the third vector module is used for obtaining an initialized key feature combination vector according to the M key features and the N key features.
Further, the system further comprises:
the fourth analysis module is used for storing the component feedback information of the first user to the feedback storage module, and calling all the component feedback information in the feedback storage module to analyze, so as to obtain negative feedback quantity distribution density, negative feedback frequency distribution density and feedback proportion of positive feedback and negative feedback;
the set generation module is used for generating an evaluation index set of each component object according to the negative feedback quantity distribution density, the negative feedback frequency distribution density and the feedback proportion of positive feedback and negative feedback;
the application field module is used for acquiring scene application fields of components corresponding to each evaluation index in the evaluation index set;
the regularization processing module is used for regularizing the evaluation index set based on the scene application field and judging whether to activate correction conditions according to regularized processing results;
and the activation module is used for generating component feedback screening characteristics according to the regularization processing result if the correction condition is activated.
Further, the system further comprises:
the model building module is used for building a template recommendation identification model, wherein the template recommendation identification model comprises an assembly recommendation index, an application rate recommendation index and an update recommendation index;
the output module is used for connecting the template recommendation recognition model with the associated application template list and outputting recommendation index lists respectively corresponding to the associated application template list according to the template recommendation recognition model;
and the recommending module is used for recommending the associated application template list according to the recommendation index list.
Through the foregoing detailed description of a knowledge service platform zero code construction method based on scene driving, those skilled in the art can clearly know that a knowledge service platform zero code construction system based on scene driving in this embodiment, and for the device disclosed in the embodiment, since the device corresponds to the method disclosed in the embodiment, the description is simpler, and relevant places refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (7)
1. A knowledge service platform zero code construction method based on scene driving is characterized by comprising the following steps:
acquiring scene demand information of a first user, and creating a scene application template list according to the scene demand information of the first user;
the key demand feature disassembly is carried out on the scene demand information of the first user, so that a key feature disassembly result is obtained;
screening the scene application template list according to the key feature disassembly result to obtain an associated application template list;
the associated application template list is sent to the first user, and component feedback screening characteristics are generated according to the component feedback information of the first user;
performing secondary screening on the application template list based on the component feedback screening characteristics and the key characteristic dismantling results, and outputting a first scene application template;
accessing an external interface into a thematic field knowledge base, and carrying out visual display loading on the first scene application template to generate a target knowledge service platform;
the method comprises the following steps of:
inputting the scene demand information of the first user into a feature disassembly model, wherein the feature disassembly model comprises a first disassembly branch and a second disassembly branch, the first disassembly branch performs feature disassembly according to scene relevance, and the second disassembly branch performs feature disassembly according to user demand custom degree;
analyzing the scene demand information of the first user according to the first disassembly branch, and outputting a first disassembly feature set comprising scene relevance grade identifiers;
analyzing scene demand information of the first user according to the second dismantling branch, and outputting a second dismantling feature set comprising a user-defined demand level identifier;
and generating a key feature disassembly result based on the first disassembly feature set and the second disassembly feature set.
2. The method of claim 1, wherein the list of scene application templates is filtered based on the key feature decomposition result, the method further comprising:
generating a first feature identification matrix according to the first disassembled feature set with the scene relevance grade identification;
generating a second feature identification matrix according to the second disassembled feature set with the user-defined requirement level identification;
performing matrix calculation according to the first characteristic identification matrix and the second characteristic identification matrix, multiplying the overlapping vectors in the first characteristic identification matrix and the second characteristic identification matrix, and outputting a third characteristic identification matrix;
and screening the scene application template list according to the third characteristic identification matrix to obtain an associated application template list.
3. The method of claim 2, wherein the scene application template list is filtered according to the third feature identification matrix, the method further comprising:
initializing key feature combination vectors, wherein the key feature combination vectors at least comprise two key feature vectors;
generating a search space according to the vector in the third feature identification matrix;
and carrying out template searching and predicting by taking the initialized key feature combination vector as an initial variable, obtaining a searching and predicting result, and carrying out combination change on the key feature combination vector based on the searching space when the searching and predicting result return value is empty until the searching and predicting result return value is not empty.
4. The method of claim 3, wherein initializing a key feature combination vector comprises:
n key features randomly selected based on the first feature identification matrix, wherein the scene relevance levels of the N key features are all larger than the preset scene relevance, and N is a positive integer larger than or equal to 0;
m key features randomly selected based on the second feature identification matrix, wherein the custom demand level of the M key features is larger than the preset demand level, and M is a positive integer larger than or equal to 0;
and obtaining an initialization key feature combination vector by using the M key features and the N key features.
5. The method of claim 1, wherein generating component feedback screening features based on the component feedback information of the first user, the method further comprising:
storing the component feedback information of the first user to a feedback storage module, and calling all the component feedback information in the feedback storage module to analyze to obtain negative feedback quantity distribution density, negative feedback frequency distribution density and feedback proportion of positive feedback and negative feedback;
generating an evaluation index set of each component object according to the negative feedback quantity distribution density, the negative feedback frequency distribution density and the feedback proportion of positive feedback and negative feedback;
acquiring scene application fields of components corresponding to all evaluation indexes in the evaluation index set;
regularizing the evaluation index set based on the scene application field, and judging whether to activate correction conditions according to regularized results;
and if the correction condition is activated, generating a component feedback screening feature according to the regularization processing result.
6. The method of claim 2, wherein the method further comprises:
building a template recommendation recognition model, wherein the template recommendation recognition model comprises a component recommendation index, an application rate recommendation index and an update recommendation index;
connecting the template recommendation recognition model with the associated application template list, and outputting recommendation index lists respectively corresponding to the associated application template list according to the template recommendation recognition model;
and recommending the associated application template list according to the recommendation index list.
7. A knowledge service platform zero code construction system based on scene driving, the system comprising:
the system comprises a list creation module, a scene application template module and a scene application template module, wherein the list creation module is used for acquiring scene demand information of a first user and creating a scene application template list according to the scene demand information of the first user;
the first feature disassembly module is used for obtaining a key feature disassembly result by carrying out key requirement feature disassembly on the scene requirement information of the first user;
the first screening module is used for screening the scene application template list according to the key feature dismantling result to obtain an associated application template list;
the feature generation module is used for sending the associated application template list to the first user and generating component feedback screening features according to the component feedback information of the first user;
the second screening module is used for carrying out secondary screening on the application template list based on the component feedback screening characteristics and the key characteristic dismantling results, and outputting a first scene application template;
the loading module is used for carrying out visual display loading on the first scene application template through accessing an external interface into a thematic field knowledge base to generate a target knowledge service platform;
the second feature disassembly module is used for inputting the scene demand information of the first user into a feature disassembly model, wherein the feature disassembly model comprises a first disassembly branch and a second disassembly branch, the first disassembly branch performs feature disassembly according to scene relevance, and the second disassembly branch performs feature disassembly according to user demand custom degree;
the first analysis module is used for analyzing the scene demand information of the first user according to the first disassembly branch and outputting a first disassembly feature set comprising scene relevance grade identification;
the third analysis module is used for analyzing the scene demand information of the first user according to the second disassembly branch and outputting a second disassembly feature set comprising a user-defined demand level identifier;
the result generation module is used for generating key feature disassembly results based on the first disassembly feature set and the second disassembly feature set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310813982.5A CN116521159B (en) | 2023-07-05 | 2023-07-05 | Knowledge service platform zero code construction method and system based on scene driving |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310813982.5A CN116521159B (en) | 2023-07-05 | 2023-07-05 | Knowledge service platform zero code construction method and system based on scene driving |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116521159A CN116521159A (en) | 2023-08-01 |
| CN116521159B true CN116521159B (en) | 2023-09-01 |
Family
ID=87408633
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310813982.5A Active CN116521159B (en) | 2023-07-05 | 2023-07-05 | Knowledge service platform zero code construction method and system based on scene driving |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116521159B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113434844A (en) * | 2021-06-23 | 2021-09-24 | 青岛海尔科技有限公司 | Intelligent scene building method and device, storage medium and electronic equipment |
| CN115202868A (en) * | 2022-06-09 | 2022-10-18 | 中国电子科技集团公司第十五研究所 | Autonomous controllable heterogeneous intelligent computing service platform and intelligent scene matching method |
| WO2023060578A1 (en) * | 2021-10-15 | 2023-04-20 | Baidu.Com Times Technology (Beijing) Co., Ltd. | Systems and methods for multi-task and multi-scene unified ranking |
| CN116204721A (en) * | 2023-03-10 | 2023-06-02 | 南京邮电大学 | Concept lattice recommendation method and device based on user record feedback and search content |
-
2023
- 2023-07-05 CN CN202310813982.5A patent/CN116521159B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113434844A (en) * | 2021-06-23 | 2021-09-24 | 青岛海尔科技有限公司 | Intelligent scene building method and device, storage medium and electronic equipment |
| WO2023060578A1 (en) * | 2021-10-15 | 2023-04-20 | Baidu.Com Times Technology (Beijing) Co., Ltd. | Systems and methods for multi-task and multi-scene unified ranking |
| CN115202868A (en) * | 2022-06-09 | 2022-10-18 | 中国电子科技集团公司第十五研究所 | Autonomous controllable heterogeneous intelligent computing service platform and intelligent scene matching method |
| CN116204721A (en) * | 2023-03-10 | 2023-06-02 | 南京邮电大学 | Concept lattice recommendation method and device based on user record feedback and search content |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116521159A (en) | 2023-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110968767B (en) | Ranking engine training method and device, and business card ranking method and device | |
| CN110738577A (en) | Community discovery method, device, computer equipment and storage medium | |
| CN107402745A (en) | The mapping method and device of DFD | |
| CN108664651A (en) | A kind of pattern recommends method, apparatus and storage medium | |
| KR101411321B1 (en) | Method and apparatus for managing neighbor node having similar characteristic with active node and computer readable medium thereof | |
| CN111459820B (en) | Model application method and device and data analysis processing system | |
| CN113206830B (en) | Data processing method and device and electronic equipment | |
| CN110929806A (en) | Picture processing method and device based on artificial intelligence and electronic equipment | |
| CN112487018B (en) | Method and device for generating list, electronic equipment and computer readable storage medium | |
| CN116521159B (en) | Knowledge service platform zero code construction method and system based on scene driving | |
| CN115422169A (en) | Data warehouse construction method and device based on commercial scene | |
| CN105608118B (en) | Result method for pushing based on customer interaction information | |
| CN110309948A (en) | Complete vehicle logistics order forecast method and device, logistics system and computer-readable medium | |
| CN111539517A (en) | Graph convolution neural network generation method based on graph structure matrix characteristic vector | |
| CN112541556A (en) | Model construction optimization method, device, medium, and computer program product | |
| KR20120079630A (en) | Method and system for indexing and searching in multi-modality data | |
| CN114691630B (en) | Smart supply chain big data sharing method and system | |
| CN111259480A (en) | Engineering cost consultation management method, system and computer storage medium | |
| CN109558420A (en) | Data query method, apparatus, electronic equipment and storage medium | |
| CN116361546A (en) | Method and device for processing search request, electronic equipment and storage medium | |
| CN115983377A (en) | Automatic learning method, device, computing equipment and medium based on graph neural network | |
| CN112101419A (en) | Efficient classification system based on crowdsourcing and creating and using methods thereof | |
| CN116561735B (en) | Mutual trust authentication method and system based on multiple authentication sources and electronic equipment | |
| CN112817635B (en) | Model processing method and data processing system | |
| CN113779263B (en) | Decision engine detection method, device and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |