View warehouse system and device based on multi-mode digital object conversion

By designing a view warehouse system for multi-modal digital object conversion in digital networking, and using the conversion framework to automatically process multi-modal digital objects, the problem of inefficient access to multi-modal digital objects is solved, and more efficient data processing is achieved. Improved user experience.

CN118939739APending Publication Date: 2024-11-12PEKING UNIV
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411122131.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

In digital networking, the process of users accessing multimodal digital objects is cumbersome and inefficient, especially when processing large batches of digital objects, which leads to an increase in time costs and seriously affects the user experience.

Method used

Design a view warehouse system based on multi-modal digital object conversion, including storage units, conversion frameworks, view warehouses and interface layers. By storing conversion rules and message templates in view digital objects, users can be automatically processed to reduce the different Direct access requirements for digital object warehouses, using the conversion framework to call conversion functions for data processing.

Benefits of technology

It simplifies user operations, saves time and costs, improves the efficiency of users to access multi-modal digital objects, meets diverse access needs, and eliminates the steps of users to manually process original data objects, improving user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118939739A_ABST
    Figure CN118939739A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of digital networking, and discloses a view warehouse system and device based on multi-mode digital object conversion. The system is deployed in a digital network, and comprises a storage unit which comprises at least one key value database and is used for storing an original digital object; the data type of the original digital object comprises at least one of character strings, numerical values, audios or images; the conversion framework is used for loading / unloading the conversion function and executing the conversion function; the conversion function is used for processing the original digital object according to a conversion rule to obtain a result digital object; the view warehouse is used for storing view digital objects; a conversion rule and a message template are stored in the view digital object; the view digital object is used for acquiring an original digital object from a key value database according to an access request sent by a user side and the message template; based on the conversion rule, calling a conversion function through the conversion framework to process the original digital object; the interface layer is used for providing an HTTP interface and a DOIP interface for accessing the view warehouse for the display layer; and the display layer is deployed at a user side and is used for providing a browser user interface. By adopting the system, the data can be automatically acquired from the plurality of digital object warehouses and processed according to the access request of the user side, so that the efficiency of accessing the multi-modal digital object by the user is improved, and the user experience is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of digital networking technology, and in particular to a view warehouse system and device based on multimodal digital object conversion. Background Art

[0002] The digital object warehouse is the core system of the digital network. The digital objects in the digital network are stored in different digital object warehouses, including different types of multimodal digital objects, such as numerical digital objects, audio digital objects, video digital objects, etc. When a user accesses a digital object in the digital network, an access request based on DOIP (Digital Object Interface Protocol, DOIP) needs to be sent to the corresponding digital object warehouse.

[0003] In the application scenario of data analysis, users need to access a large number of multimodal digital objects in the digital object warehouse for further analysis, which requires users to manually send a large number of access requests to different digital object warehouses. After obtaining the multimodal digital objects in different digital object warehouses, users also need to process each digital object, perform operations such as data encryption and decryption, data type conversion, data format conversion, data calculation, etc., so as to obtain the final required data. The current process of accessing multimodal digital objects is very cumbersome and inefficient, especially when processing large quantities of digital objects. The time cost consumed is greatly increased, which seriously affects the user experience. In view of this, how to improve the efficiency of user access to multimodal digital objects and enhance the user experience is a problem that needs to be solved. Summary of the invention

[0004] In view of this, the present application aims to propose a view warehouse system and device based on multimodal digital object conversion to improve the efficiency of user access to multimodal digital objects and enhance user experience.

[0005] To achieve the above purpose, the technical solution of this application is as follows:

[0006] In a first aspect, an embodiment of the present application provides a view warehouse system based on multimodal digital object conversion, wherein the system is deployed in a digital network and includes:

[0007] A storage unit, comprising at least one key-value database, for storing original digital objects; the data type of the original digital objects comprises at least one of the following: a string, a value, an audio, or an image;

[0008] A conversion framework, used for loading / unloading conversion functions and executing conversion functions; the conversion functions are used for processing the original digital object according to the conversion rules to obtain a result digital object;

[0009] A view warehouse is used to store view digital objects; the view digital objects store conversion rules and message templates; the view digital objects are used to obtain original digital objects from a key-value database according to an access request sent by a user terminal and the message template; based on the conversion rules, the conversion function is called through the conversion framework to process the original digital objects;

[0010] The interface layer is used to provide the presentation layer with an HTTP interface and a DOIP interface for accessing the view repository;

[0011] The presentation layer is deployed on the user side and is used to provide a browser user interface.

[0012] According to a second aspect of an embodiment of the present application, a multimodal digital object conversion device is provided, which is deployed in the system provided in the first aspect of the embodiment of the present application, and includes:

[0013] A user interface module is configured to receive an access request sent by a user terminal, and send the access request to a view digital object in the system provided by the first aspect of the embodiment of the present application;

[0014] The message generation module is configured to obtain the request parameters in the access request through the view digital object; generate at least one sub-request according to the request parameters and the message template; send the at least one sub-request to a corresponding digital object repository, wherein the digital object repository is a key-value database in the system provided by the first aspect of the embodiment of the present application; and receive the original digital objects returned by each digital object repository;

[0015] A conversion module is configured to call a conversion framework to execute a corresponding conversion function to process the original digital object according to the conversion rule to obtain a result digital object;

[0016] The user interface module is also configured to return the result digital object to the user end.

[0017] Optionally, the message generating module is configured to generate at least one sub-request according to the request parameters and the message template, specifically including:

[0018] Replace the reserved parameter field in the message template with the request parameter;

[0019] Based on the number of the identifiers of the digital object repositories in the message template, a corresponding number of sub-requests are generated; wherein each sub-request is used to obtain the original digital object from a different digital object repository.

[0020] Optionally, when the data type of the original digital object is a string, the conversion module is configured to process the original digital object, specifically comprising:

[0021] Convert the original digital object into keywords or abstracts using a TextRank algorithm;

[0022] Or, de-identify and encrypt the keywords in the original digital object;

[0023] or, performing conformal encryption on the original digital object;

[0024] Or, performing random replacement encryption on the original digital object;

[0025] Or, the designated position of the original digital object is coded and encrypted.

[0026] Optionally, when the data type of the original digital object is a numerical value, the conversion module is configured to process the original digital object, specifically comprising:

[0027] Performing a rounding operation on the original digital object;

[0028] Or, perform a differential privacy operation on the original digital object.

[0029] Optionally, when the data type of the original digital object is audio, the conversion module is configured to process the original digital object, specifically comprising:

[0030] Calling the Vosk API to convert the original digital object into a text string;

[0031] Or, the audio segment at a specific position of the original digital object is set to blank.

[0032] Optionally, when the data type of the original digital object is an image, the conversion module is configured to process the original digital object, specifically comprising:

[0033] Convert the original digital object into a label string;

[0034] Or, scaling the original digital object;

[0035] Or, blurring a designated pixel area in the original digital object.

[0036] Optionally, the view digital object further stores an address of a custom program package; the custom program package has a plurality of custom conversion functions;

[0037] The conversion module is configured to process the original digital object, and further includes:

[0038] Calling the conversion framework to load the corresponding custom program package according to the address of the custom program package;

[0039] After the custom program package is loaded, the conversion framework is called to execute the corresponding custom conversion function according to the conversion rule.

[0040] According to a third aspect of an embodiment of the present application, a multimodal digital object conversion method is provided, which is applied to the device provided in the second aspect of the embodiment of the present application, including:

[0041] receiving an access request sent by a user terminal, and sending the access request to a view digital object; the view digital object stores a conversion rule and a message template;

[0042] Obtaining request parameters in the access request through the view digital object; generating at least one sub-request according to the request parameters and the message template; sending the at least one sub-request to a corresponding digital object repository, and receiving the original digital object returned by each digital object repository;

[0043] According to the conversion rule, calling the conversion framework to execute the corresponding conversion function to process the original digital object to obtain a result digital object;

[0044] The result digital object is returned to the user end.

[0045] Optionally, the multimodal digital object conversion method further includes:

[0046] Predefine conversion functions used in different application scenarios, including the following types: string type conversion function, numeric type conversion function, image type conversion function, and audio type conversion function; each type of conversion function can be used to perform common conversion or privacy conversion on digital objects;

[0047] Corresponding view digital objects are created in advance according to different application scenarios, and corresponding conversion rules and message templates are created and stored in each view digital object; the conversion rules are used to call at least one conversion function to process the acquired original digital object.

[0048] This system provides a view warehouse and conversion framework. The view digital object in the view warehouse stores the corresponding conversion rules and message templates. The view digital object receives the access request from the user end and automatically requests at least one corresponding key-value database to obtain the original digital object according to the message template. The user only needs to initiate a DOIP access request to the view warehouse once, and no longer needs to manually request data from different key-value databases, which greatly simplifies user operations and saves time costs.

[0049] After obtaining the original digital object, according to the conversion rules, the conversion framework calls the conversion function to process the original digital object, and obtains the result digital object required by the user, eliminating the step of manually processing the original data object, improving the efficiency of user access to the result digital object, and enhancing the user experience. In addition, the conversion framework can support multi-modal digital object processing, including the conversion of two structured digital objects of string type and numeric type, as well as the conversion of two unstructured digital objects of pictures and audio, which can meet the diverse access needs of multi-modal digital objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0051] Figure 1 It is a schematic diagram of the architecture of a view warehouse system based on multimodal digital object conversion proposed in an embodiment of the present application;

[0052] Figure 2 is a schematic diagram of a process for accessing a multimodal digital object in an embodiment of the present application;

[0053] Figure 3 is a schematic diagram of a multimodal digital object conversion device proposed in an embodiment of the present application;

[0054] Figure 4 It is a flowchart of a multimodal digital object conversion method proposed in one embodiment of the present application. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0056] It should be understood that the references to "one embodiment" or "an embodiment" throughout the specification mean that the specific features, structures, or characteristics associated with the embodiment are included in at least one embodiment of the present application. Therefore, the references to "in one embodiment" or "in an embodiment" appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0057] In the various embodiments of the present application, it should be understood that the size of the serial numbers of the following processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0058] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects as detailed in the present application.

[0059] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application may be combined with each other.

[0060] Multimodal digital objects are stored in the digital object warehouse, and data is requested from the digital object warehouse through DOIP messages. There are three scenarios when users access the digital object warehouse: first, the digital objects of the type required by the user are stored in other modal forms in the digital object warehouse, such as images stored in numerical coding form; second, the digital objects of the type required by the user need to be converted from multiple digital objects located in different digital object warehouses, such as the final result needs to be calculated from multiple digital objects; third, the digital objects required by the user need to be digital objects that have been processed for privacy information to ensure that sensitive information is not leaked or abused.

[0061] The JSON formatted representation of the DOIP message is divided into two parts: Header and Body. The Header mainly consists of Id and Attribute. Id and Body are in string format, Attribute is in JSON format, and digital objects are stored in the Body in the form of strings. For example, the user's data is abstracted into 4 DOs (DigitalObjects) DO1, DO2, DO3, and DO4, which are located in different digital object warehouses. When they circulate through the Body of the DOIP message, the Body of DO1 is in JSON format, with two fields: student ID and math score. The Body of DO2 is in JSON format, with two fields: student ID and Chinese score. The Body of DO3 is a Base64 string, which represents a slogan of the user. The Body of DO4 is a Base64 string, which represents the user's avatar. The user wants to convert the original digital object: sum the scores of DO1 and DO2 and round them down; encode the student ID part of DO1 to protect personal privacy; convert the audio of DO3 into a text string; convert the image of DO4 into a label, and then form a new digital object to return, completing the digital object access operation. In the existing solution, in order to complete the conversion of the above original digital objects, the user needs to write four digital object conversion rules to describe the required digital objects, which is cumbersome and has low access efficiency.

[0062] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0063] Figure 1 This is a schematic diagram of the architecture of a view warehouse system based on multimodal digital object conversion proposed in an embodiment of the present application. The system is deployed in a digital network and includes:

[0064] A storage unit, comprising at least one key-value database, for storing original digital objects; the data type of the original digital objects comprises at least one of the following: a string, a value, an audio, or an image;

[0065] A conversion framework, used for loading / unloading conversion functions and executing conversion functions; the conversion functions are used for processing the original digital object according to the conversion rules to obtain a result digital object;

[0066] A view warehouse is used to store view digital objects; the view digital objects store conversion rules and message templates; the view digital objects are used to obtain original digital objects from a key-value database according to an access request sent by a user terminal and the message template; based on the conversion rules, the conversion function is called through the conversion framework to process the original digital objects;

[0067] The interface layer is used to provide the presentation layer with an HTTP interface and a DOIP interface for accessing the view repository;

[0068] The presentation layer is deployed on the user side and is used to provide a browser user interface.

[0069] In this embodiment, a conversion framework capable of processing multi-modal digital objects is proposed. In order to realize the processing of digital objects of string type, numerical type, image type and audio type, the conversion framework needs to be able to support the definition of digital object conversion functions, add / unload conversion functions and execute conversion functions according to conversion rules. Its functional implementation is as follows:

[0070] (1) Define built-in conversion functions: Conversion functions are Java-based functions, which are divided into four categories: string type conversion functions, numeric type conversion functions, image type conversion functions, and audio type conversion functions. Each type of conversion function is further divided into two types: common conversion and privacy conversion. Common conversion is to process the original digital object in an unencrypted manner, and privacy conversion is to process the original digital object in an encrypted manner.

[0071] In this embodiment, an abstract class and an abstract method of the conversion function are constructed, and the conversion framework implements the above conversion function based on the above method, and is built into the conversion framework for calling. Specifically, a NashornScriptEngineUtil tool class is written, and a Nashorn JavaScript engine instance is created through the tool class. The implementation class where the conversion function is located is instantiated and stored in the Nashorn JavaScript engine instance, and the above conversion function is subsequently used in the engine through the class instance. It is worth noting that the defined conversion function needs to be loaded into the conversion framework before it can be used.

[0072] (2) Execute the conversion function: Pre-store the instance of the implementation class of the above built-in conversion function into the Nashorn JavaScript engine instance, so that the conversion function can be used in the Nashorn JavaScript engine instance in the form of "instance name.function name". Write a ForkVisitor class to store the original digital object in the form of a JSON array, and implement the execution of the conversion function through the ForkVisitor class.

[0073] (3) Add / unload conversion functions: Conversion functions are divided into built-in conversion functions and custom conversion functions. Built-in conversion functions are pre-defined and instantiated and stored in the Nashorn JavaScript engine instance, while custom conversion functions can be programs provided by users. Custom conversion functions can be loaded into the conversion framework in the form of Jar packages (custom program packages) to facilitate subsequent use according to conversion rules.

[0074] In this embodiment, a conversion framework is constructed based on the Nashorn JavaScript engine. Various Java conversion functions are built into the framework and used through the Nashorn JavaScript engine. The framework supports the conversion of structured digital objects of string type and numerical type, as well as the conversion of unstructured digital objects such as images and audio. Specifically, the processing methods supported by the conversion framework include: text string conversion to keywords, text string conversion to summary, text string de-identification, string type-preserving encryption, string coding, the first N bits of the string remain unchanged, numerical data rounding up / down, numerical data differential privacy, image conversion to label, image scaling, image global blurring and specific position blurring, audio conversion to text, and audio specific segment blanking. NashornScriptEngineUtil, JsonVisitor, and ForkVisitor are the three core classes of the conversion framework, among which ForkVisitor inherits from JsonVisitor. The input of ForkVisitor is the digital object and conversion rule to be converted. It uses NashornScriptEngineUtil to call the conversion function to complete the execution of the conversion rule, and then returns the execution result of the conversion rule, that is, the result digital object finally obtained by the user.

[0075] In this embodiment, a view warehouse is constructed based on a conversion framework.

[0076] A view repository (ViewRepo) is constructed based on the conversion framework to process the DOIP access request sent by the user end. Multiple view digital objects (ViewDo) are created in the view repository in advance for different access scenarios, and conversion rules and message templates are stored therein. In this embodiment, a key-value database is used as the digital object party library. The message template of the view digital object is used to generate a sub-request for accessing the digital object warehouse to obtain the original digital object. The conversion rule is used to describe what type of processing and conversion needs to be done on the obtained original digital object, that is, what type of result digital object the user ultimately needs. In addition, when creating a view digital object, the conversion function required by the conversion rule must also be defined. The view warehouse supports the creation, deletion, update, and query functions of view digital objects, where the fields of the view digital object are shown in Table 1 below.

[0077] Table 1

[0078] Field Name Field Type illustrate id string View Digital Object Identifier enableIndex boolean Whether to add index basicInfo jsonObject Public parameters mappingRule jsonObject Matching rules searchMeta jsonObject Search Metadata authMeta jsonObject Permission metadata schemaMeta jsonArray Field metadata

[0079] Among them, the core fields stored in the matching rule (mappingRule) include conversion rules and message templates. The public parameters store a Type field, which has two values, namely, the value corresponding to the field match and the value corresponding to the Jar package match. Among them, field matching means calling the built-in digital object conversion function to process and convert the initial digital object, and Jar package matching means that the custom program package needs to be loaded into the Nashorn JavaScript engine. The message template is used to generate a sub-request to request the original digital object from at least one digital object repository in the current access scenario, and the conversion rule is used to determine the conversion function to be used. The view warehouse interacts directly with the user end, receives the access request sent by the user, and returns the resulting digital object to the user.

[0080] In this embodiment, by constructing a view warehouse, it is possible to automatically request at least one corresponding key-value database to obtain the original digital object according to an access request sent by the user. The user only needs to initiate a DOIP access request to the view warehouse, and no longer needs to manually request data from different key-value databases, which greatly simplifies user operations and saves time costs. By constructing a conversion framework for processing multi-modal digital objects, flexible processing of different types of data is achieved to meet the diverse access needs of users, and the user's manual processing of the original data object is eliminated, thereby improving the efficiency of user access to the resulting digital object and enhancing the user experience.

[0081] Based on the same inventive concept, an embodiment of the present application provides a multi-modal digital object conversion device. Figure 3 , Figure 3 FIG. 1 is a schematic diagram of a multimodal digital object conversion device 100 proposed in an embodiment of the present application. Figure 3 As shown, the device comprises:

[0082] The user interface module 101 is configured to receive an access request sent by a user terminal, and send the access request to a view digital object in the system provided in the above embodiment;

[0083] The message generation module 102 is configured to obtain the request parameters in the access request through the view digital object; generate at least one sub-request according to the request parameters and the message template; and send the at least one sub-request to a corresponding digital object repository, wherein the digital object repository is a key-value database in the system provided in the above embodiment;

[0084] The conversion module 103 is configured to receive the original digital objects returned by each digital object repository; according to the conversion rule, call the conversion framework to execute the corresponding conversion function to process the original digital objects and obtain the result digital objects;

[0085] The user interface module 101 is further configured to return the result digital object to the user end.

[0086] In this embodiment, the view warehouse stores view digital objects for different access scenarios, and each view digital object stores corresponding message templates and conversion rules. When the user terminal initiates an access request, the user interface module receives the access request sent by the user terminal and forwards it to the corresponding view digital object in the view warehouse. The message generation module obtains the request parameters in the access request through the view digital object, and generates multiple sub-requests to request the original digital object from different digital object warehouses according to the stored message template and request parameters. For example, for the total score access scenario, it is necessary to obtain the user's Chinese score and math score at the same time, and then perform a sum operation to obtain the result digital object. The digital object of the Chinese score and the digital object of the math score are stored in two digital object warehouses respectively, so two sub-requests need to be generated in the message template, which are used to obtain the Chinese score and the math score from the two digital object warehouses respectively. In response to the sub-request, the digital object warehouse returns the corresponding original digital object to the view digital object. The conversion module calls the conversion framework to execute the corresponding conversion function (for example, the conversion function of the summation operation) according to the conversion rules stored in the view digital object, processes the acquired original digital object, and thus obtains the result digital object required by the user, and the user interface module returns the result digital object to the user end.

[0087] As an implementation mode of the present application, the view digital object further stores the address of a custom program package; the custom program package has a plurality of custom conversion functions;

[0088] The conversion module is configured to process the original digital object, and further includes:

[0089] Calling the conversion framework to load the corresponding custom program package according to the address of the custom program package;

[0090] After the custom program package is loaded, the conversion framework is called to execute the corresponding custom conversion function according to the conversion rule.

[0091] In one embodiment, the conversion framework also supports the functionality of extending custom conversion functions, and custom program packages are added / unloaded through the conversion framework, so that the original data can be flexibly processed according to needs, thereby improving the flexibility and extensibility of the view repository and the conversion framework.

[0092] Before loading a custom conversion function, you need to package the custom conversion function into a custom program package (Jar package). When a user creates a view digital object, if you need to use a custom conversion function, store the Jar package address where the custom conversion function is located and the corresponding class name in the matching rule field of the view digital object, and set the value of the Type field in the public parameters of the view digital object to the corresponding value of the Jar package match. When loading the Jar package, transfer the Jar package address and the corresponding class name to the NashornScriptEngineUtil tool class, and then use the class loader in the NashornScriptEngineUtil tool class to load the Jar package, and use reflection technology to instantiate the class in the Jar package and put it into the Nashorn JavaScript engine instance, so that the conversion function in the Jar package can be used in the Nashorn JavaScript engine through the class instance.

[0093] When deleting a view digital object, you need to uninstall the corresponding custom conversion function, and simultaneously delete the Jar package address and the corresponding class name stored in the NashornScriptEngineUtil tool class to complete the uninstallation of the custom conversion function.

[0094] Figure 2 1 is a flow chart of accessing a multimodal digital object in an embodiment of the present application. Figure 2 As shown, the client (user end) sends an access request to the view repository (ViewRepo) in accordance with the DOIP protocol. The view digital object (ViewDo) in the view repository that processes the request generates multiple DOIP sub-requests based on the request and obtains the original digital object from different digital object repositories (Repo). The conversion framework runs the Nashorn Javascript engine through the Java virtual machine, calls the built-in conversion function to process the original digital object, generates the result digital object and returns it to the user client. Alternatively, a custom program package (Jar package) is pre-loaded, and the original digital object is processed by calling the custom conversion function in the Jar package, and the result digital object is generated and returned to the user client.

[0095] As an implementation manner of the present application, the message generation module is configured to generate at least one sub-request according to the request parameters and the message template, specifically including:

[0096] Replace the reserved parameter field in the message template with the request parameter;

[0097] Based on the number of the identifiers of the digital object repositories in the message template, a corresponding number of sub-requests are generated; wherein each sub-request is used to obtain the original digital object from a different digital object repository.

[0098] In one embodiment, the user sends an access request to ViewDo, and ViewDo generates multiple sub-requests according to the DOIP request parameters and the message template. Specifically, the request parameters are obtained from the access request and used to replace the reserved parameter fields in the message template, and then the corresponding number of sub-requests are generated according to the number of identifiers of the digital object repositories in the message template for obtaining data from the corresponding digital object repositories.

[0099] For example, Zhang San needs to obtain the total score composed of Chinese score and math score. The DOIP request sent by the user contains the request parameter: Zhang San's student number "12345678". The message template of the view digital object stores the program templates for applying for Chinese score and math score from two digital object repositories (yuwen and shuxue) respectively. The reserved parameter field "${input.attributes.student number}" for replacement is reserved in the program template. An example of the core fields of the message template is as follows:

[0100] / / The first template is used to obtain the original digital object DO1 (i.e., a certain row of data in the SQL-type digital object storing math score), and the second template is used to obtain the original digital object DO2 (i.e., a certain row of data in the SQL-type digital object storing Chinese score)

[0102] {

[0103] "id":"score / math",

[0104] "op":"Retrieve",

[0105] "attributes":{"student number":"${input.attributes.student number}"}

[0106] },

[0107] {

[0108] "id":"score / chinese",

[0109] "op":"Retrieve",

[0110] "attributes":{"student number":"${input.attributes.student number}"}

[0111] }

[0112] ​Use the request parameters to replace the reserved parameter fields in the program template, and generate a corresponding number of sub-requests according to the digital object warehouse identifier (yuwen, shuxue) in the message template. In this case, two sub-requests are generated, as follows:

[0113] / / Get Zhang San's math score DO1

[0114] {

[0115] "id":"view / viewdo1",

[0116] "op":"Retrieve",

[0117] "attributes":{"student number":"12345678"}

[0118] }

[0119] / / Get Zhang San's Chinese score DO2

[0120] {

[0121] "id":"view / viewdo2",

[0122] "op":"Retrieve",

[0123] "attributes":{"student number":"12345678"}

[0124] }

[0125] The generated sub-request is sent to the corresponding digital object warehouse to obtain the original digital object. After obtaining the original digital object (i.e., math score DO1, Chinese score DO2), the conversion framework calls the corresponding conversion function according to the conversion rule to process the original digital object. In this case, the math score and Chinese score need to be summed to obtain Zhang San's total score (i.e., the result digital object).

[0126] In this case, the conversion rule uses the object of the FloorConverter implementation class in the conversion framework and the "int convert(int,int)" conversion function in the instance. Specifically, the core fields of the conversion rule are as follows:

[0127] / / Get the math score of digital object do1 and the Chinese score of digital object do2, sum them up and adjust them to a number close to the sum result and divisible by 10, and return the resulting digital object to the client in the form of {"total score": conversion result}.

[0128] {

[0129] "Total Score": "${floorConverter.convert(do1.mathScore + do2.chineseScore, 10)}"

[0130] }

[0131] When executing the conversion function, the conversion framework calls the NashornScriptEngineUtil tool class through the ForkVisito class to create a JavaScript function in the Nashorn JavaScript engine instance. This JavaScript function calls the instance floorConverter of the FloorConverter class and uses the convert function of this object to replace ${floorConverter.convert(do1.mathScore + do2.chineseScore, 10)} in the conversion rule with the execution result of the JavaScript function, and executes to obtain the result digital object required by the user.

[0132] As an implementation manner of this application, when the data type of the original digital object is a string, the conversion module is configured to process the original digital object, specifically including:

[0133] Using the TextRank algorithm to convert the original digital object into keywords or abstracts;

[0134] Or, de-identifying and encrypting the keywords in the original digital object;

[0135] Or, performing conformal encryption on the original digital object;

[0136] Or, performing random replacement encryption on the original digital object;

[0137] Or, performing censor encryption on the specified position of the original digital object.

[0138] In one embodiment, the conversion framework calls the conversion function to process the string original digital object, specifically including the following:

[0139] (1) Using the TextRank algorithm to convert the text string into keywords. The specific steps executed by the conversion function are as follows:

[0140] 1) Segment the text string according to complete sentences;

[0141] 2) Perform word segmentation, part-of-speech tagging, and stop word filtering on each sentence in turn. Common words, punctuation marks, and "words other than verbs, adverbs, adjectives, and nouns" will be filtered out to obtain candidate keywords;

[0142] 3) Create two windows of size 5, and each candidate keyword votes for the candidate keywords with a distance of 5 before and after it;

[0143] 4) According to the sorted voting results, the three most important words are obtained as the TOP-3 keywords.

[0144] (2) Use the TextRank algorithm to convert the text string into a summary. The specific steps performed by the conversion function are as follows:

[0145] 1) Split the given text string into complete sentences, and perform word segmentation and stop word filtering on each sentence.

[0146] 2) Each sentence is used as a node, and the BM25 algorithm is used to calculate the correlation between every two sentences;

[0147] 3) Perform iterative calculations on each sentence until convergence, obtain the three most important sentences in order of relevance, and return them as summaries.

[0148] (3) For the three types of sensitive information in the text string, namely name, place name, and organizational structure name, de-identification is performed through Chinese word segmentation, part-of-speech tagging, and named entity recognition functions based on conditional random fields (CRF). The specific steps are as follows:

[0149] 1) Public corpus is used for Chinese word segmentation, part-of-speech tagging and named entity recognition. The public corpus requires that words and parts of speech be separated by " / " and any word must have a part of speech. Words are separated by spaces and support the use of "[]" to merge multiple words into a compound word.

[0150] 2) Use CRF++ to train the Chinese word segmentation model, part-of-speech tagging model, and named entity recognition model. The training process of the above models is similar. Take the Chinese word segmentation model training as an example. First, convert the public corpus into CRF++ format, then prepare a feature template, use CRF++'s crf_learn for training, and obtain the text format model "cws.txt". Create the corresponding word segmenter through this model.

[0151] In one embodiment, a ".txt.bin" binary cache may also be created and used to reduce the duration of the next load, further improving the execution efficiency of the conversion function and reducing the delay in the user obtaining the resulting digital object.

[0152] 3) Load the trained model and construct a CRF lexical analyzer. Through the three steps of Chinese word segmentation, part-of-speech tagging, and named entity recognition, identify the three types of sensitive identifying information in the text string: name, place name, and organizational structure name. Replace them with "[Name]", "[Place]", and "[Organization]" in the text string respectively.

[0153] (4) Implement corresponding conformal encryption for numeric strings, English letter strings, and mixed numeric and English letter strings. The above three encryption principles are the same, the difference lies in the different Alphabet dictionary tables and Tweak values ​​used. The specific steps are as follows:

[0154] 1) Design an Alphabet dictionary table. Numeric strings include bank card numbers, mobile phone numbers, ID card numbers, etc. For the conformal encryption of numeric strings, a numeric alpha dictionary table (0-9) can be used. English alphabet strings include English names, etc. For the conformal encryption of English alphabet strings, an English alphabet dictionary table (lowercase letters az, uppercase letters AZ) can be used. Mixed numeric and English alphabet strings are a mixture of the above two, and a dictionary table (0-9, lowercase letters az, uppercase letters AZ) is used.

[0155] 2) Design Tweak values ​​and use them in conjunction with encryption algorithm keys. Tweak is used to resolve conflicting results caused by local encryption. Usually, the immutable part of the data is used as Tweak. For conformal encryption of numeric strings such as mobile phone numbers, ID numbers, and bank card numbers, you can take the first M bits and the last N bits together as Tweak, and encrypt the middle part conformally. For conformal encryption of alphabetic strings such as English names, take the first and last letters of the word as Tweak, and encrypt the remaining English letters conformally. For conformal encryption of mixed strings of numbers and English letters, you can randomly take N bits as Tweak, and then encrypt the entire thing conformally.

[0156] 3) Design the Key value and use the FF1 algorithm to encrypt the data. The Key length is one of 16 bits, 24 bits, and 32 bits. If decryption is not required, the key can be randomly selected to obtain higher security. For example, a secure random number generator can be used to obtain a 16-bit Key.

[0157] (5) Keep the first few characters of the string unchanged and replace the remaining characters (numbers and letters) randomly. The specific steps of the conversion function are as follows:

[0158] If it is a number, it is randomly replaced with another number between 0 and 9. If it is a letter (uppercase or lowercase), it is randomly offset by a certain position (within the current uppercase and lowercase letter range) and returns the new character.

[0159] (6) Encode and encrypt the specified position of the digital object. The specific steps of the conversion function are as follows:

[0160] Replace the characters in the specified range of the string with the "*" symbol. The user needs to enter the original string, the start position and the end position of the coding.

[0161] As an implementation manner of the present application, when the data type of the original digital object is a numerical value, the conversion module is configured to process the original digital object, specifically including:

[0162] Performing a rounding operation on the original digital object;

[0163] Or, perform a differential privacy operation on the original digital object.

[0164] In one embodiment, the conversion framework calls the conversion function to process the numerical digital object, which specifically includes the following:

[0165] (1) Rounding up / down of Int / Double numeric data:

[0166] The user enters numeric data of type Int or Double, and a "count" value, and the conversion function adjusts the value to a number that is close to the original value and can be divided by "count".

[0167] (2) Differential privacy of Int / Double numeric data:

[0168] The maximum difference in statistical query results caused by changing a single record in the data set is called sensitivity Δf:

[0169] Δf=max|f(M)-f(M')|.

[0170] Among them, M and M' are data sets that are the same except for one data item, and f is the query function. When setting the sensitivity, it is necessary to determine it according to the characteristics of the query function: if f is a summation function, then the sensitivity is the possible maximum value of a single data item; if f is a function that calculates an average, then the sensitivity is the possible maximum change of a single data item divided by the total amount of data. The sensitivity is set in this way to quantify the maximum impact that a change in a single data item may have on the query results. This quantitative value (sensitivity) further determines the amount of noise that needs to be added to ensure that the predetermined level of privacy protection is achieved. The higher the sensitivity, the more noise needs to be added to maintain the privacy of the data. Conversely, less noise can be added to ensure data availability while maintaining data privacy. When adding noise locally to a single numerical data item, the sensitivity can be set to 1, and then the amount of noise to be added can be controlled by other parameters.

[0171] The process of adding Laplace noise to single numerical data is as follows:

[0172] Determine the privacy budget ε. The smaller the ε value, the more noise is added, the higher the privacy protection is, but the practicality of the data will be reduced. According to the sensitivity and privacy budget, use Laplace distribution to generate noise. The generation formula is usually Laplace(Δf / ε). Add the generated Laplace noise to the original numerical data.

[0173] The process of adding Gaussian noise to single numerical data is as follows:

[0174] Determine the privacy budget ε and δ. The parameter δ represents the upper limit of the probability of violating the privacy protection of ε. According to the sensitivity, privacy budget and δ value, the Gaussian distribution formula Gaussian(σ 2 ) generates noise and adds it to the original numerical data, where σ is the standard deviation calculated based on Δf, ε and δ.

[0175] As an implementation manner of the present application, when the data type of the original digital object is audio, the conversion module is configured to process the original digital object, specifically including:

[0176] Calling the Vosk API to convert the original digital object into a text string;

[0177] Or, the audio segment at a specific position of the original digital object is set to blank.

[0178] In one embodiment, the audio digital object is processed by calling the conversion function through the conversion framework, which specifically includes the following:

[0179] (1) Use the Vosk API to convert the Base64-encoded audio string into a text string. The specific steps are as follows:

[0180] 1) Preprocess the audio data. Set the encoding format, number of channels, sampling rate, and output format, and convert the audio according to the requirements of the Vosk API.

[0181] 2) Prepare a Kaldi model that can be used in Vosk, which is compiled from three data sources: dictionary, acoustic model, and language model. The acoustic model is trained with a large amount of speech data and is used to identify phonemes or sound features in speech signals. The language model is used to understand how words are combined into sentences and predict the possibility of word sequences based on statistical probability. Language models of different sizes and complexities can be used in Vosk to balance recognition accuracy and performance. The acoustic model usually outputs a series of identifiers, each of which represents an acoustic unit. The mapping relationship in the dictionary is used to convert these identifiers into corresponding words.

[0182] 3) Extract the Mel-frequency cepstral coefficients (MFCCs) and convert the audio signal into a series of feature vectors that can represent the signal. Then use the acoustic model to evaluate the matching degree between different sound fragments and various phonemes, and generate a series of probability scores to indicate the possibility of a certain phoneme appearing. Finally, the most likely word sequence is determined through a search algorithm combined with a dictionary, acoustic score, and language model to obtain the final text result.

[0183] (2) Setting a specific position of the audio digital object to empty. The specific steps are as follows:

[0184] The conversion function replaces the Base64-encoded audio string with a blank audio segment at the specified start and end positions (defined by seconds), and then returns the converted audio in Base64-encoded form.

[0185] As an implementation manner of the present application, when the data type of the original digital object is an image, the conversion module is configured to process the original digital object, specifically including:

[0186] Convert the original digital object into a label string;

[0187] Or, scaling the original digital object;

[0188] Or, blurring a designated pixel area in the original digital object.

[0189] In one embodiment, the conversion function image digital object is processed by calling the conversion framework, specifically including the following:

[0190] (1) Use DJL to call the PyTorch model and convert the Base64-encoded image string into a label string. The specific conversion process is as follows:

[0191] 1) Use Tracing and Scripting to convert the PyTorch model to the TorchScript format supported by DJL;

[0192] 2) Create a pipeline (i.e., the preprocessing that needs to be done on each image) and use the pipeline to create a converter;

[0193] 3) Use DJL’s core class object Criteria to load the PyTorch model in TorchScript format;

[0194] 4) Load the image, then use the model and transformer to create predictor variables and call the predict method on the image.

[0195] (2) Scaling the Base64-encoded image string. The specific steps are as follows:

[0196] The image is converted to the specified size according to the final required image size (defined by the width and height of pixels) and returned in Base64 encoded form.

[0197] (3) Obfuscate the Base64-encoded image string. The specific steps are as follows:

[0198] Performs Gaussian blur on the specified area of ​​the image according to the specified coordinate parameters and blur area size parameters. Alternatively, if no coordinate parameters and blur area size parameters are specified, performs Gaussian blur on the entire image. After blurring, the image is returned in Base64 encoded form.

[0199] Based on the same inventive concept, an embodiment of the present application provides a multimodal digital object conversion method. Figure 4 , Figure 4 FIG. 1 is a flow chart of a multimodal digital object conversion method proposed in an embodiment of the present application. Figure 4 As shown, the method includes:

[0200] S1: receiving an access request sent by a user terminal, and sending the access request to a view digital object; the view digital object stores a conversion rule and a message template;

[0201] S2: obtaining request parameters in the access request through the view digital object; generating at least one sub-request according to the request parameters and the message template; sending the at least one sub-request to a corresponding digital object repository, and receiving the original digital object returned by each digital object repository;

[0202] S3: According to the conversion rule, calling the conversion framework to execute the corresponding conversion function to process the original digital object to obtain a result digital object;

[0203] S4: Return the result digital object to the user end.

[0204] As an implementation manner of the present application, the multimodal digital object conversion method further includes:

[0205] Predefine conversion functions used in different application scenarios, including the following types: string type conversion function, numeric type conversion function, image type conversion function, and audio type conversion function; each type of conversion function can be used to perform common conversion or privacy conversion on digital objects;

[0206] Corresponding view digital objects are created in advance according to different application scenarios, and corresponding conversion rules and message templates are created and stored in each view digital object; the conversion rules are used to call at least one conversion function to process the acquired original digital object.

[0207] As an implementation manner of the present application, at least one sub-request is generated according to the request parameter and the message template, including:

[0208] Replace the reserved parameter field in the message template with the request parameter;

[0209] Based on the number of the identifiers of the digital object repositories in the message template, a corresponding number of sub-requests are generated; wherein each sub-request is used to obtain the original digital object from a different digital object repository.

[0210] As an implementation manner of the present application, when the data type of the original digital object is a string, processing the original digital object includes:

[0211] Convert the original digital object into keywords or abstracts using a TextRank algorithm;

[0212] Or, de-identify and encrypt the keywords in the original digital object;

[0213] or, performing conformal encryption on the original digital object;

[0214] Or, performing random replacement encryption on the original digital object;

[0215] Or, the designated position of the original digital object is coded and encrypted.

[0216] As an implementation manner of the present application, when the data type of the original digital object is a numerical value, processing the original digital object includes:

[0217] Performing a rounding operation on the original digital object;

[0218] Or, perform a differential privacy operation on the original digital object.

[0219] As an implementation manner of the present application, when the data type of the original digital object is audio, processing the original digital object includes:

[0220] Calling the Vosk API to convert the original digital object into a text string;

[0221] Or, the audio segment at a specific position of the original digital object is set to blank.

[0222] As an implementation manner of the present application, when the data type of the original digital object is an image, processing the original digital object includes:

[0223] Convert the original digital object into a label string;

[0224] Or, scaling the original digital object;

[0225] Or, blurring a designated pixel area in the original digital object.

[0226] As an implementation mode of the present application, the view digital object further stores the address of a custom program package; the custom program package has a plurality of custom conversion functions;

[0227] Processing the original digital object further includes:

[0228] Calling the conversion framework to load the corresponding custom program package according to the address of the custom program package;

[0229] After the custom program package is loaded, the conversion framework is called to execute the corresponding custom conversion function according to the conversion rule.

[0230] Based on the same inventive concept, an embodiment of the present application provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps in the method provided in the above embodiment of the present application are implemented.

[0231] Based on the same inventive concept, an embodiment of the present application provides a readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the method provided in the above embodiment of the present application are implemented.

[0232] Based on the same inventive concept, an embodiment of the present application provides an electronic device. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the computer program is executed by the processor, the steps in the method provided in the above embodiment are implemented.

[0233] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0234] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0235] For the method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and components involved are not necessarily required by the present application.

[0236] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, devices, or computer program products. Therefore, the embodiments of the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the embodiments of the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0237] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0238] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0239] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0240] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they are aware of the basic creative concepts. Therefore, the present application is interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0241] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0242] The view warehouse system and device based on multimodal digital object conversion provided by the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A view warehouse system based on multimodal digital object conversion, characterized in that: Deployed in the Internet of Things, including: A storage unit, comprising at least one key-value database, for storing original digital objects; the data type of the original digital objects comprises at least one of the following: a string, a value, an audio, or an image; A conversion framework, used for loading / unloading conversion functions and executing conversion functions; the conversion functions are used for processing the original digital object according to the conversion rules to obtain a result digital object; A view warehouse is used to store view digital objects; the view digital objects store conversion rules and message templates; the view digital objects are used to obtain original digital objects from a key-value database according to an access request sent by a user terminal and the message template; based on the conversion rules, the conversion function is called through the conversion framework to process the original digital objects; The interface layer is used to provide the presentation layer with an HTTP interface and a DOIP interface for accessing the view repository; The presentation layer is deployed on the user side and is used to provide a browser user interface.

2. A multimodal digital object conversion device, characterized in that: Deployed in the system as claimed in claim 1, comprising: A user interface module, configured to receive an access request sent by a user terminal, and send the access request to a view digital object in the system according to claim 1; a message generation module configured to obtain request parameters in the access request through the view digital object; generate at least one sub-request according to the request parameters and the message template; and send the at least one sub-request to a corresponding digital object repository, wherein the digital object repository is a key-value database in the system according to claim 1; The conversion module is configured to receive the original digital objects returned by each digital object repository; according to the conversion rule, call the conversion framework to execute the corresponding conversion function to process the original digital objects to obtain the result digital objects; The user interface module is also configured to return the result digital object to the user end.

3. The multimodal digital object conversion device according to claim 2, characterized in that: The message generation module is configured to generate at least one sub-request according to the request parameters and the message template, specifically including: Replace the reserved parameter field in the message template with the request parameter; Based on the number of the identifiers of the digital object repositories in the message template, a corresponding number of sub-requests are generated; wherein each sub-request is used to obtain the original digital object from a different digital object repository.

4. The multimodal digital object conversion device according to claim 2, characterized in that: In the case where the data type of the original digital object is a string, the conversion module is configured to process the original digital object, specifically including: Convert the original digital object into keywords or abstracts using a TextRank algorithm; Or, de-identify and encrypt the keywords in the original digital object; or, performing conformal encryption on the original digital object; Or, performing random replacement encryption on the original digital object; Or, the designated position of the original digital object is coded and encrypted.

5. The multimodal digital object conversion device according to claim 2, characterized in that: In the case where the data type of the original digital object is a numerical value, the conversion module is configured to process the original digital object, specifically comprising: Performing a rounding operation on the original digital object; Or, perform a differential privacy operation on the original digital object.

6. The multimodal digital object conversion device according to claim 2, characterized in that: In the case where the data type of the original digital object is audio, the conversion module is configured to process the original digital object, specifically comprising: Calling the Vosk API to convert the original digital object into a text string; Or, the audio segment at a specific position of the original digital object is set to blank.

7. The multimodal digital object conversion device according to claim 2, characterized in that: In the case where the data type of the original digital object is an image, the conversion module is configured to process the original digital object, specifically comprising: Convert the original digital object into a label string; Or, scaling the original digital object; Or, blurring a designated pixel area in the original digital object.

8. The multimodal digital object conversion device according to claim 2, characterized in that: The view digital object also stores the address of a custom program package; the custom program package has a plurality of custom conversion functions; The conversion module is configured to process the original digital object, and further includes: Calling the conversion framework to load the corresponding custom program package according to the address of the custom program package; After the custom program package is loaded, the conversion framework is called to execute the corresponding custom conversion function according to the conversion rule.

9. A multimodal digital object conversion method, characterized in that: The device as claimed in any one of claims 2 to 8 comprises: receiving an access request sent by a user terminal, and sending the access request to a view digital object; the view digital object stores a conversion rule and a message template; Obtaining request parameters in the access request through the view digital object; generating at least one sub-request according to the request parameters and the message template; sending the at least one sub-request to a corresponding digital object repository, and receiving the original digital object returned by each digital object repository; According to the conversion rule, calling the conversion framework to execute the corresponding conversion function to process the original digital object to obtain a result digital object; The result digital object is returned to the user end.

10. The multimodal digital object conversion method according to claim 9, characterized in that: Also includes: Predefine conversion functions used in different application scenarios, including the following types: string type conversion function, numeric type conversion function, image type conversion function, and audio type conversion function; each type of conversion function can be used to perform common conversion or privacy conversion on digital objects; Corresponding view digital objects are created in advance according to different application scenarios, and corresponding conversion rules and message templates are created and stored in each view digital object; the conversion rules are used to call at least one conversion function to process the acquired original digital object.

Citation Information

Patent Citations

  • Heterogeneous data integration platform

    CN107895046A

  • Big data multi-center heterogeneous dynamic data conversion method

    CN110442629A

  • Public legal service data processing method and device, electronic equipment and storage medium

    CN115827753A

  • Scientific research number networking service method and system based on scientific research number object pragmatic relationship

    CN117271460A