Method and apparatus for searching application function information

Abstract

The application discloses a retrieval method and device of application function information. It is related to the field of artificial intelligence, and the method comprises the following steps: receiving a retrieval request initiated by a requester; in response to the retrieval request, generating a reference specification set of a target application by using a first generation model, wherein the first generation model is obtained by training a first initial generation model by using an application attribute library and a technical specification library which are marked with application specification set samples of each application in a plurality of applications; generating a target specification set according to the reference specification set and a target technical document of the target application by using a second generation model, wherein the second generation model is obtained by training a second initial generation model by using application specification set samples which are marked with function specification set samples of each function in each application and function technical documents; and sending the target specification set to the requester. Through the application, the problem of low retrieval efficiency of application function-related technical specifications in the related art is solved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and more specifically, to a method and apparatus for retrieving application function information. Background Technology

[0002] Whether in financial institutions or other companies, enterprises typically establish various enterprise-level technical specifications, stipulating the guidelines that applications must follow during development. These specifications are usually tailored to different domains and technology platforms. Therefore, when designing an application architecture, architects need to determine all the technical specifications that must be considered. However, these specifications are scattered across various enterprise-level technical specification documents, making it difficult for architects to quickly find the technical specifications that the application's functionalities need to focus on.

[0003] There is currently no effective solution to the problem of low retrieval efficiency of technical specifications related to application functions in related technologies. Summary of the Invention

[0004] The main objective of this application is to provide a method and apparatus for retrieving application function information, so as to solve the problem of low retrieval efficiency of application function-related technical specifications in related technologies.

[0005] To achieve the above objectives, according to one aspect of this application, a method for retrieving application function information is provided. This method is applied to an application information retrieval device. The method includes: receiving a retrieval request initiated by a requester, wherein the retrieval request requests the retrieval of application function information required to implement a target function in a target application, and the application function information indicates the technical specifications required to implement the target function in the target application; responding to the retrieval request by generating a reference specification set for the target application through a first generation model, wherein the first generation model is trained using an application attribute library and a technical specification library that annotates application specification set samples of various applications across multiple applications, the application attribute library stores application attributes of various applications, the technical specification library stores the technical specifications required to be followed when implementing each application attribute, and the application specification set samples include the technical specifications required to be followed when implementing the corresponding application; generating a target specification set based on the reference specification set and the target technical document of the target application through a second generation model, wherein the second generation model is trained using an application specification set sample and a functional technical document that annotates functional specification set samples of various functions in various applications, the functional technical document records the implementation method of the corresponding function, the target technical document records the implementation method of the target function in the target application, and the functional specification set sample includes the technical specifications required to be followed when implementing the corresponding function in the corresponding application; and sending the target specification set to the requester.

[0006] Optionally, generating a reference specification set for the target application through a first generative model includes: obtaining a target application identifier for the target application; adding the target application identifier to the target position of a first preset prompt word to obtain a first target prompt word, wherein the first preset prompt word is used to describe the requirement for generating technical specifications corresponding to the content added at the target position using natural language; and executing the first target prompt word using an application attribute library and a technical specification library through the first generative model to obtain a reference specification set, wherein the first generative model is obtained by training the first model function of the first initial generative model using application prompt words labeled with application specification set samples of various applications, and the first model function includes executing the received prompt words that conform to the semantic expression of the application prompt words using the application attribute library and the technical specification library to obtain an application specification set, wherein the application prompt words are used to describe the requirement for generating technical specifications corresponding to the corresponding application using natural language.

[0007] Optionally, the first target prompt word is executed using the first generation model with an application attribute library and a technical specification library to obtain a reference specification set, including: responding to the first target prompt word with the first generation model and extracting the target application identifier from the target position in the first target prompt word; using the application attribute library as a knowledge base with the first generation model and extracting the target application attribute corresponding to the target application identifier from the application attribute library; using the technical specification library as a knowledge base with the first generation model and extracting the technical specification that conforms to the target application attribute from the technical specification library to obtain a reference specification set.

[0008] Optionally, the technical specification library is used as a knowledge base through the first generative model. Technical specifications that conform to the target application attributes are extracted from the technical specification library to obtain a reference specification set, including: using the technical specification library as a knowledge base through the first generative model, extracting technical specifications that conform to the target application attributes from the technical specification library to obtain each reference specification item; and converting each reference specification item into a corresponding specification vector through the first generative model to obtain a reference specification set including multiple reference specification vectors.

[0009] Optionally, a target specification set is generated by a second generation model based on a reference specification set and a target technical document of the target application. This includes: obtaining the target technical document; adding the target technical document to a reference position of a second preset prompt word to obtain a second target prompt word, wherein the second preset prompt word is used to describe in natural language the generation of technical specifications corresponding to the content added at the reference position; and executing the second target prompt word using the reference specification set through the second generation model to obtain the target specification set. The second generation model is trained using functional prompt words labeled with functional specification set samples for each function to train the second model function of the second initial generation model. The second model function includes executing the received prompt words that conform to the semantic expression of the functional prompt words using application specification set samples to obtain the functional specification set. The functional prompt words are used to describe in natural language the requirements for generating technical specifications corresponding to the corresponding functions.

[0010] Optionally, the second target prompt word is executed using the reference specification set through the second generation model to obtain the target specification set, including: responding to the second target prompt word through the second generation model and extracting the target technical document from the reference position in the second target prompt word; using the reference specification set as a knowledge base through the second generation model and extracting the technical specifications that conform to the target technical document from the reference specification set to obtain the target specification set.

[0011] Optionally, before generating the reference specification set of the target application through the first generative model, the method further includes: obtaining application attribute update information and technical specification update information of each application from the application database through the target interface, wherein the application database is used to store all application running information of multiple applications, the application attribute update information is the information on changes in the current application attributes of each application relative to the application attributes stored in the application attribute library, and the technical specification update information is the information on changes in the current technical specifications that each application attribute needs to follow when implementing the application relative to the technical specifications stored in the technical specification library; converting the application attribute update information into application attribute vectors and converting the technical specification update information into technical specification vectors; updating the current application attribute library using the application attribute vectors and updating the current technical specification library using the technical specification vectors.

[0012] To achieve the above objectives, according to another aspect of this application, an application function information retrieval device is provided. This device is applied to an application information retrieval equipment and includes: a receiving module for receiving a retrieval request initiated by a requester, wherein the retrieval request is for requesting the retrieval of application function information required to implement a target function in a target application, and the application function information is for indicating the technical specifications required to implement the target function in the target application; and a first generation module for responding to the retrieval request and generating a reference specification set for the target application through a first generation model, wherein the first generation model is obtained by training a first initial generation model using an application attribute library and a technical specification library that annotates application specification set samples of various applications in multiple applications, the application attribute library stores the application attributes of each application, and the technical specification library stores the technical specifications required for the implementation of each application attribute. The application specification set includes technical specifications that the corresponding application must follow when implementing it. The second generation module generates a target specification set based on the reference specification set and the target application's target technical documents using a second generation model. The second generation model is trained using application specification set samples labeled with functional specification sets for each function in each application and functional technical documents. The functional technical documents record the implementation methods of the corresponding functions, and the target technical documents record the implementation methods of the target functions in the target application. The functional specification set includes technical specifications that must be followed when implementing the corresponding functions in the corresponding application. The sending module sends the target specification set to the requester.

[0013] Optionally, the first generation module includes: a first acquisition unit, used to acquire a target application identifier of the target application; a first addition unit, used to add the target application identifier to the target position of a first preset prompt word to obtain a first target prompt word, wherein the first preset prompt word is used to describe the requirement of generating technical specifications corresponding to the content added at the target position using natural language; and a first execution unit, used to execute the first target prompt word through an application attribute library and a technical specification library using a first generation model to obtain a reference specification set, wherein the first generation model is obtained by training the first model function of the first initial generation model using application prompt words labeled with application specification set samples of various applications, and the first model function includes executing the received prompt words that conform to the semantic expression of the application prompt words to obtain an application specification set using the application attribute library and the technical specification library, wherein the application prompt words are used to describe the requirement of generating technical specifications corresponding to the corresponding application using natural language.

[0014] Optionally, the first execution unit is further configured to: respond to the first target prompt word through the first generation model, and extract the target application identifier from the target position in the first target prompt word; use the application attribute library as a knowledge base through the first generation model, and extract the target application attribute corresponding to the target application identifier from the application attribute library; use the technical specification library as a knowledge base through the first generation model, and extract the technical specifications that conform to the target application attributes from the technical specification library, thereby obtaining a reference specification set.

[0015] Optionally, the first execution unit is further configured to: use the technical specification library as a knowledge base through the first generation model, extract technical specifications that conform to the target application attributes from the technical specification library to obtain various reference specification entries; and convert each reference specification entry into a corresponding specification vector through the first generation model to obtain a reference specification set including multiple reference specification vectors.

[0016] Optionally, the second generation module includes: a second acquisition unit for acquiring a target technical document; a second adding unit for adding the target technical document to a reference position of a second preset prompt word to obtain a second target prompt word, wherein the second preset prompt word is used to describe the generation of technical specifications corresponding to the content added at the reference position using natural language; and a second execution unit for executing the second target prompt word using a reference specification set through a second generation model to obtain a target specification set, wherein the second generation model is trained using functional prompt words labeled with functional specification set samples of each function to train the second model function of the second initial generation model, and the second model function includes executing the received prompt words that conform to the semantic expression of the functional prompt words using the application specification set samples to obtain a functional specification set, wherein the functional prompt words are used to describe the requirements for generating technical specifications corresponding to the corresponding functions using natural language.

[0017] Optionally, the second execution unit is further configured to: respond to the second target prompt word through the second generation model, extract the target technical document from the reference position in the second target prompt word; use the reference specification set as a knowledge base through the second generation model, extract the technical specifications that conform to the target technical document from the reference specification set, and obtain the target specification set.

[0018] Optionally, the application function information retrieval device further includes: an acquisition module, used to acquire application attribute update information and technical specification update information of each application from the application database through the target interface before generating the reference specification set of the target application through the first generation model, wherein the application database is used to store all application operation information of multiple applications, the application attribute update information is the information on the changes in the current application attributes of each application relative to the application attributes stored in the application attribute library, and the technical specification update information is the information on the changes in the current technical specifications that each application attribute needs to follow when implementing the application relative to the technical specifications stored in the technical specification library; a conversion module, used to convert the application attribute update information into application attribute vectors and the technical specification update information into technical specification vectors; and an update module, used to update the current application attribute library using the application attribute vectors and update the current technical specification library using the technical specification vectors.

[0019] In this embodiment, upon receiving a retrieval request from a requester requesting information on application functions required to implement a target function in a target application, the system first responds to the retrieval request and generates a reference specification set for the target application using a first generation model. This first generation model is trained on an application attribute library and a technical specification library that annotates application specification set samples from multiple applications. Then, a target specification set is generated using a second generation model based on the reference specification set and the target technical document of the target application. This second generation model is trained on a second initial generation model using application specification set samples annotated with function specification set samples from various applications and the functional technical document. Finally, the target specification set is sent to the requester. This achieves the goal of quickly finding the technical specifications required to implement the target function of the target application recorded in the target technical document using the first and second generation models, the application attribute library, and the technical specification library. This improves the retrieval efficiency of application function-related technical specifications and solves the problem of low retrieval efficiency caused by the need to manually review a large number of enterprise-level technical specification documents. Attached Figure Description

[0020] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0021] Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for retrieving application function information is shown.

[0022] Figure 2This is a flowchart of the application function information retrieval method according to Embodiment 1 of this application;

[0023] Figure 3 This is a schematic diagram of an application function information retrieval system according to an embodiment of this application;

[0024] Figure 4 This is a schematic diagram of an application function information retrieval device provided according to an embodiment of this application;

[0025] Figure 5 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] It should be noted that the application function information retrieval method and apparatus in this application can be used in the field of financial technology when designing applications for financial institutions, or in any field other than financial technology when designing applications. This application does not limit the application field of the application function information retrieval method and apparatus.

[0029] It should be noted that all relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, and displayed data) involved in this application are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, processing, transmission, provision, disclosure, use, and handling of such data comply with the laws, regulations, and standards of the relevant regions, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding access points are provided for users to choose to authorize or refuse access. For example, this system has interfaces with relevant users or organizations. Before obtaining relevant information, a request to obtain the information needs to be sent to the aforementioned user or organization through the interface, and the relevant information is obtained only after receiving consent from the aforementioned user or organization.

[0030] The information collection (e.g., user voice, video, and text collection) and analysis operations involved in this application have provided users with corresponding operation entry points during execution, allowing users to choose to agree to or reject the automated decision results; if the user chooses to reject, the process will proceed to the expert decision-making process.

[0031] The present application will now be described in detail with reference to various embodiments.

[0032] Example 1

[0033] According to an embodiment of this application, an embodiment of a method for retrieving application function information is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0034] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal (or mobile device) for implementing a method for retrieving application function information is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0035] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0036] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method for retrieving application function information in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned method for retrieving application function information. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0037] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0038] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0039] Under the aforementioned operating environment, this application provides the following: Figure 2The method for retrieving application function information is shown. Figure 2 This is a flowchart of the application function information retrieval method according to Embodiment 1 of this application. For example... Figure 2 As shown, the method for retrieving application function information may include, but is not limited to, the following steps:

[0040] Step S202: Receive a search request initiated by the requester, wherein the search request is used to request the search for application function information required to implement the target function in the target application, and the application function information is used to indicate the technical specifications required to implement the target function in the target application.

[0041] Step S204: In response to the retrieval request, a reference specification set for the target application is generated through the first generation model. The first generation model is trained using an application attribute library and a technical specification library that annotates application specification set samples of each application in multiple applications. The application attribute library stores the application attributes of each application, and the technical specification library stores the technical specifications that each application attribute must follow when implementing the application. The application specification set samples include the technical specifications that the corresponding application must follow when implementing it.

[0042] Step S206: Generate a target specification set using the second generation model based on the reference specification set and the target technical document of the target application. The second generation model is obtained by training the second initial generation model with application specification set samples and functional technical documents that are labeled with functional specification set samples of each function in each application. The functional technical documents record the implementation method of the corresponding function, the target technical documents record the implementation method of the target function in the target application, and the functional specification set samples include the technical specifications that need to be followed when implementing the corresponding function in the corresponding application.

[0043] Step S208: Send the target specification set to the requester.

[0044] Through the above steps, upon receiving a retrieval request from the requester requesting information on application functions required to implement the target function in the target application, the system first responds to the retrieval request. A reference specification set for the target application is generated using a first generation model, which is trained on an application attribute library and a technical specification library that annotates application specification set samples from multiple applications. Then, a target specification set is generated using a second generation model based on the reference specification set and the target technical document of the target application. This second generation model is trained on a second initial generation model using application specification set samples annotated with function specification set samples from various applications and the functional technical document. Finally, the target specification set is sent to the requester. This achieves the goal of quickly finding the technical specifications required to implement the target function of the target application recorded in the target technical document using the first and second generation models, the application attribute library, and the technical specification library. This improves the retrieval efficiency of application function-related technical specifications and solves the problem of low retrieval efficiency caused by manually reviewing a large number of enterprise-level technical specification documents.

[0045] Optionally, in this embodiment, the aforementioned method for retrieving application function information may be applied to, but is not limited to, an application information retrieval device. Here, the application information retrieval device may refer to, but is not limited to, hardware and / or software systems specifically designed to process and execute application function information retrieval tasks. The application information retrieval device may exist as a standalone device, or it may be integrated into a system that requires application function information retrieval, such as, but not limited to, an application design system. This application does not specifically limit the form in which the application information retrieval device is designed.

[0046] Optionally, in the embodiment provided in step S202, the requesting party may refer to, but is not limited to, the device or system that initiates the search request. The requesting party may be, but is not limited to, a display system, which may receive the search task triggered by the user on the display screen and convert the search task into a search request and send it to the aforementioned application information search device.

[0047] Optionally, in this embodiment, the retrieval request may, but is not limited to, directly carry the target application identifier of the target application. The target application identifier may, but is not limited to, be the target application name. By sending the target application name to the application information retrieval device, the application information retrieval device may, but is not limited to, trigger the application information retrieval device to take the latest received technical document among one or more technical documents corresponding to the target application name as the target technical document. The application information retrieval device then retrieves the application function information required to implement the target function in the target application based on the target application name and the target technical document.

[0048] Optionally, in this embodiment, the search request may also include, but is not limited to, directly carrying the target application name and target technical document of the target application. The application information retrieval device may, but is not limited to, directly extract the target application name and target technical document from the search request after receiving it.

[0049] Optionally, in this embodiment, the search request may also include, but is not limited to, directly carrying the target technical document. The application information retrieval device may, but is not limited to, extract the target technical document from the search request after receiving the search request, and then extract the target application name from the target technical document.

[0050] Optionally, in this embodiment, the application function information may, but is not limited to, indicate the technical specifications required to implement the target function in the target application in various ways. For example, when the target function includes multiple functions, the application function information may, but is not limited to, indicate the technical specifications required to implement each function in the target application in the form of a list, which facilitates developers to check and implement each function one by one. Alternatively, the application function information may also be returned in the form of structured data, such as JSON or XML, in which the relevant technical specification entries for each function are clearly marked and organized, which facilitates automated processing and analysis by the program.

[0051] Optionally, in the embodiment provided in step S204, a reference specification set for the target application may be generated using the first generative model, but is not limited to. When generating the reference specification set for the target application, the first generative model may, but is not limited to, retrieve relevant information about the target application from the latest application attribute library and technical specification library outside the first generative model, and then use this retrieved relevant information to guide the generation of text, thereby improving the quality and accuracy of the prediction.

[0052] Optionally, in this embodiment, the application attribute library may store, but is not limited to, various application attributes of each application, such as the application's disaster recovery level, whether the application is on the cloud, the programming language used by the application, etc.

[0053] Optionally, in this embodiment, the technical specification library may, but is not limited to, storing the technical specifications that need to be followed when implementing applications for various application attributes. For example, what specifications need to be followed for applications with a disaster recovery level of 3, what specifications need to be followed for applications with a disaster recovery level of 4, what specifications need to be followed for cloud-based applications, and so on.

[0054] It should be noted that the aforementioned technical specifications corresponding to different disaster recovery levels, different cloud migration requirements, and different development programming languages ​​may be stored in different enterprise specification documents, but are not limited to. For example, technical specifications corresponding to different disaster recovery levels may be stored in the "High Availability Design Specification," and technical specifications corresponding to different cloud migration requirements may be stored in the "Application Design Specification." Different enterprise specification documents may be created and written by different people, and therefore their writing and expression habits may also differ. Manually searching for the required technical specification items would be very time-consuming and labor-intensive for the above reasons. However, the method provided in this application can fully utilize the model's language understanding and data processing capabilities to quickly and efficiently find the technical specification items required for application design.

[0055] Optionally, in this embodiment, a knowledge base for application attributes and technical specifications may be constructed first, but is not limited to this. This includes converting application attribute information and related technical specification documents into vector representations. Vector retrieval techniques such as BERT-as-Service and FAISS may be used to convert text into dense vectors and store them in a vector database. A large pre-trained language model (such as BERT or T5) may be selected as the first initial generation model, and this first initial generation model may or may not have strong language understanding and generation capabilities. A series of application specification set samples may be prepared, each sample containing an application identifier (target application identifier) ​​and a set of related technical specifications. These samples are used to guide the model to learn how to retrieve appropriate technical specifications based on application attributes. A first preset prompt word may be redesigned, the structure of which includes a position for inserting the target application identifier so that the model can understand the retrieval requirements. For example, the prompt word could be designed as: "[Application Identifier] What technical specifications need to be considered when implementing the application?" Retrieval training using the knowledge base enables the model to retrieve relevant application attributes and corresponding specifications from the application attribute and technical specification bases when it receives a prompt word with an application identifier. Based on the retrieved specifications, the model is trained using samples from the application specification set to generate a reference specification set that matches the attributes of the target application and covers all necessary technical specifications. After training, the model's retrieval and generation capabilities are evaluated, particularly the accuracy in generating the reference specification set. Fine-tuning is then performed based on the evaluation results until the model achieves satisfactory performance, resulting in the first generative model.

[0056] Optionally, in the embodiment provided in step S206, the second generative model may be used to generate a target specification set for the target technical document. When generating the target specification set for the target technical document, the second generative model may, but is not limited to, retrieve relevant information of the target technical document from the latest reference specification set outside the second generative model, and then use this retrieved relevant information to guide the generation of text, thereby improving the quality and accuracy of the prediction.

[0057] Optionally, in this embodiment, the target technical document may include, but is not limited to, various design documents related to application development, such as system overall plan, preliminary design document, detailed design document, white paper, etc. The target technical document may be, but is not limited to, a design document for a specific system version, and may record, but is not limited to, new features that the target application needs to update during a particular update iteration. For example, the target application may have included cloud computing functionality and Level 4 disaster recovery functionality in its initial design, but the cloud computing functionality was not implemented in the initial version. However, it needs to be implemented in a later version update. In this case, the target technical document records the implementation method of the cloud computing functionality in the target application, and the retrieval method provided in this application can be used to retrieve the technical specifications that need to be followed when implementing the cloud computing functionality in the target application.

[0058] Optionally, in this embodiment, similar to the construction of a technical specification library, a vectorized knowledge base of application specification sets is created to ensure that the second generative model can access the technical specifications related to the implementation of all relevant functions of the application. A large pre-trained language model (such as BERT, T5, etc.) can be selected as the second initial generative model, and this second initial generative model can, but is not limited to, possess strong language understanding and generation capabilities. Functional specification set samples can, but are not limited to, be prepared, with each sample containing a technical document for a function and a set of technical specifications that must be followed when implementing that function. A prompt word can, but is not limited to, be designed to guide the model to generate technical specifications based on the target technical document, such as: "Please provide the technical specifications that need to be considered in the [functional description]". Retrieval training is performed using the functional technical document and the reference specification set, enabling the model to retrieve the technical specifications associated with the target technical document from the reference specification set. Generative training can, but is not limited to, use functional specification set samples to teach the model how to generate a set of technical specifications for a specific function implementation, which should be refined and specific to every detail of the target technical document, resulting in the second generative model.

[0059] Optionally, in the embodiment provided in step S208, when the target specification set is generated, it can be packaged and returned to the requester via the same API or communication protocol. The requester (such as the architect's workstation) receives this specification set to guide subsequent design and development work and ensure that all necessary technical specifications are followed when the application functions are implemented.

[0060] Optionally, in the application function information retrieval method provided in this application embodiment, generating a reference specification set for the target application through a first generation model may include, but is not limited to: obtaining the target application identifier of the target application; adding the target application identifier to the target position of a first preset prompt word to obtain a first target prompt word, wherein the first preset prompt word is used to describe the requirement for generating technical specifications corresponding to the content added at the target position using natural language; and executing the first target prompt word through the first generation model using an application attribute library and a technical specification library to obtain a reference specification set, wherein the first generation model is obtained by training the first model function of the first initial generation model using application prompt words labeled with application specification set samples of various applications. The first model function includes executing the received prompt words that conform to the semantic expression of the application prompt words using the application attribute library and the technical specification library to obtain an application specification set, wherein the application prompt words are used to describe the requirement for generating technical specifications corresponding to the corresponding application using natural language.

[0061] Optionally, in this embodiment, the target application identifier may be, but is not limited to, the application name or application number of the target application, etc.

[0062] Optionally, in this embodiment, the first preset prompt word may be, but is not limited to, the aforementioned "[Application Identifier] What technical specifications need to be considered when implementing the application?" or any other prompt word that uses natural language to express the technical specifications required to generate the content added at the target location.

[0063] Based on the above, by combining customized first preset prompts with target application identifiers, the first generation model can efficiently retrieve technical specifications that match the application attributes of the target application, improving the targeting and accuracy of specification retrieval, and ensuring that the generated set of reference specifications can accurately reflect the specific needs of the application, thereby optimizing design and development decisions.

[0064] Optionally, in the application function information retrieval method provided in this application embodiment, the first target prompt word is executed through the first generation model using the application attribute library and technical specification library to obtain a reference specification set, which may include, but is not limited to: responding to the first target prompt word through the first generation model and extracting the target application identifier from the target position in the first target prompt word; using the application attribute library as a knowledge base through the first generation model and extracting the target application attribute corresponding to the target application identifier from the application attribute library; using the technical specification library as a knowledge base through the first generation model and extracting the technical specification that conforms to the target application attribute from the technical specification library to obtain a reference specification set.

[0065] Optionally, in this embodiment, the application attribute library can be used as an external knowledge base for the first generation model, allowing the first generation model to use information from the external knowledge base to generate target application attributes; similarly, the technical specification library can be used as an external knowledge base for the first generation model, allowing the first generation model to use information from the external knowledge base to generate a set of reference specifications.

[0066] Optionally, in this embodiment, compared to directly retraining a traditional large model using the application attribute library and technical specification library, this application uses the application attribute library and technical specification library as the knowledge base of the first generative model, which can realize the rapid generation of the required information using the model without incurring significant costs.

[0067] Based on the above, when performing a retrieval, the first generative model can accurately extract and match application attributes and technical specifications by using an external application attribute library and technical specification library. This ensures that the generated reference specification set not only covers all application attributes but also focuses on specific needs, improving the intelligence level of the specification retrieval process, reducing interference from useless information, and enhancing the practical value of specification recommendations.

[0068] Optionally, in the application function information retrieval method provided in this application embodiment, the technical specification library is used as a knowledge base through the first generation model, and technical specifications that conform to the target application attributes are extracted from the technical specification library to obtain a reference specification set. This may include, but is not limited to, the following: using the technical specification library as a knowledge base through the first generation model, extracting technical specifications that conform to the target application attributes from the technical specification library to obtain each reference specification item; and converting each reference specification item into a corresponding specification vector through the first generation model to obtain a reference specification set including multiple reference specification vectors.

[0069] Through the above, technical specification items that conform to the target application attributes are converted into vector form, constructing a structured knowledge representation of the reference specification set. This greatly improves the speed and accuracy of the second generative model in retrieving and matching the target specification set, promotes the immediacy of specification generation, and enables specification recommendations to quickly respond to the specific needs of applications and functions.

[0070] Optionally, in the application function information retrieval method provided in this application embodiment, generating a target specification set based on a reference specification set and a target technical document of the target application through a second generation model may include, but is not limited to: obtaining the target technical document; adding the target technical document to a reference position of a second preset prompt word to obtain a second target prompt word, wherein the second preset prompt word is used to describe in natural language the generation of technical specifications corresponding to the content added at the reference position; executing the second target prompt word using the reference specification set through the second generation model to obtain the target specification set, wherein the second generation model is obtained by training the second model function of the second initial generation model using functional prompt words labeled with functional specification set samples of each function, and the second model function includes executing the received prompt words that conform to the semantic expression of the functional prompt words using application specification set samples to obtain the functional specification set, wherein the functional prompt words are used to describe in natural language the requirement to generate technical specifications corresponding to the corresponding function.

[0071] Optionally, in this embodiment, the second preset prompt may be, but is not limited to, the aforementioned "Please provide the technical specifications that need to be considered in the [Function Description]" or any other prompt that uses natural language to express the need to generate technical specifications corresponding to the content added at the reference position.

[0072] Based on the above, the second generative model can deeply understand the content of the target technical document, accurately identify the functional attributes mentioned in the document, and retrieve the most suitable specification entries from a large set of reference specifications, avoiding interference from irrelevant specifications. The model's execution process is highly automated, generating a set of technical specifications for specific functional implementations without human intervention, significantly reducing the burden on architects and developers in compliance checks. Through the model's intelligent retrieval and vector matching technologies, it ensures that the generated specification set is closely related to the functional implementation of the target technical document, improving the accuracy of compliance with technical specifications and reducing the risks caused by misinterpretation or neglect of specifications.

[0073] Optionally, in the application function information retrieval method provided in the embodiments of this application, the second generation model uses the reference specification set to execute the second target prompt word to obtain the target specification set, which may include, but is not limited to: responding to the second target prompt word through the second generation model and extracting the target technical document from the reference position in the second target prompt word; using the reference specification set as a knowledge base through the second generation model and extracting the technical specifications that conform to the target technical document from the reference specification set to obtain the target specification set.

[0074] Optionally, in this embodiment, the reference specification set may be used as an external knowledge base for the second generative model, allowing the second generative model to use the information in the external knowledge base to generate technical specifications that conform to the target technical document.

[0075] Optionally, in this embodiment, compared to directly retraining a traditional large model using the reference specification set, this application uses the reference specification set as the knowledge base of the second generative model, which can realize the rapid generation of the required information using the model without incurring significant costs.

[0076] Based on the above, the second generative model can accurately retrieve and extract relevant technical specifications from the reference specification set according to the specific needs of the target technical document. The generated functional specification set is more specific and relevant, and can directly guide the implementation of functions, greatly improving the efficiency of developers and reducing the risk of incorrect implementation due to misunderstanding of technical specifications.

[0077] Optionally, in the application function information retrieval method provided in this application embodiment, before generating the reference specification set of the target application through the first generation model, the application attribute library and technical specification library may be updated in the following ways, but not limited to: obtaining application attribute update information and technical specification update information of each application from the application database through the target interface, wherein the application database is used to store all application running information of multiple applications, the application attribute update information is the information on the changes that have occurred in the current application attributes of each application relative to the application attributes stored in the application attribute library, and the technical specification update information is the information on the changes that have occurred in the current technical specifications that each application attribute needs to follow when implementing the application relative to the technical specifications stored in the technical specification library; converting the application attribute update information into application attribute vectors and converting the technical specification update information into technical specification vectors; updating the current application attribute library using the application attribute vectors and updating the current technical specification library using the technical specification vectors.

[0078] Optionally, in this embodiment, the application database may, but is not limited to, store all application runtime information for multiple applications, such as application information ledgers and enterprise-level technical specification document libraries. The application database may, but is not limited to, information maintained internally by the enterprise and can be used directly as basic data. The application database may, but is not limited to, provide external calling interfaces (i.e., target interfaces) through APIs (Application Programming Interfaces), enabling external devices / systems to access the data stored in the application database via APIs.

[0079] Optionally, in this embodiment, since the enterprise's technical specifications and application information are constantly changing, it is necessary to update the application attribute database and technical specification database in a timely manner to ensure the timeliness and accuracy of technical specification retrieval.

[0080] As an alternative implementation, this application also provides a system for retrieving application function information. Figure 3 This is a schematic diagram of an application function information retrieval system according to an embodiment of this application. For example... Figure 3 As shown, a retrieval system may include, but is not limited to, the following components:

[0081] (1) Enterprise assets: This includes, but is not limited to, information maintained internally by the enterprise that can be used directly as basic data. Typically, this data is stored in a system that provides an API (Application Programming Interface) for external systems to access.

[0082] (2) Basic Knowledge Base: Essentially a vector database, it stores the enterprise's application information and technical specification documents as vectors, serving as an external knowledge base for the model. The basic knowledge base may include, but is not limited to, an application information knowledge base (i.e., an application attribute base) and a specification knowledge base (i.e., a technical specification base).

[0083] (3) Synchronizer: Since a company's technical specifications and application information are constantly changing, and these changes only manifest in the company's assets and do not affect the content of the basic knowledge base, a synchronization mechanism needs to be established between the basic knowledge base and the company's assets. This mechanism ensures that when company assets change, the changed content can be synchronized to the basic knowledge base in real time. The synchronizer may include, but is not limited to, the following three components:

[0084] ① Scheduler: Enables scheduled tasks (e.g., every day at 4 AM) to trigger synchronous tasks;

[0085] ② Preprocessing module: Cleans and segments the text in the application information ledger and enterprise-level technical document library of enterprise assets to prepare for vectorization;

[0086] ③ Vectorization engine: Converts the cleaned text into high-dimensional vectors and writes them into a vector database (i.e., a basic knowledge base).

[0087] (4) Application-level specification model (i.e., the first generative model): This model may consist of, but is not limited to, a pre-trained and usable large language model, a set of preset prompt words (i.e., the first preset prompt words), and an external knowledge base (i.e., the basic knowledge base). The input of this model is the application name. After receiving the input, the model first obtains the basic information of the application (such as whether it is on the cloud, disaster recovery level, etc.) through the external knowledge base (application information knowledge base, specification knowledge base), then performs vector retrieval from the specification knowledge base, outputs all specification entries involved in the application, and then stores these entries as a temporary knowledge base in the application-level specification knowledge base (i.e., the reference specification set).

[0088] (5) Application-level Specification Knowledge Base: This is a temporary knowledge base, serving as the core model's knowledge base, storing the specification content related to the application input by the user. When the user inputs an application name, the application-level specification model outputs all specification entries related to that application. These entries, after being vectorized, form the application-level specification knowledge base. The reason for establishing such a temporary knowledge base is that the data in this knowledge base is based on the application name and enterprise assets input by the user. However, enterprise assets and user input are dynamically changing, and the data cannot be prepared in advance. Therefore, the content can only be determined each time user input is received.

[0089] (6) Core Model: This model consists of a pre-trained, usable multimodal large language model, a set of pre-set prompt words (i.e., the second pre-set prompt words), and an external knowledge base. The input of this model is a technical document. By analyzing the content of the technical document and combining it with the application-level specification knowledge base, the model outputs all the technical specification items that the technical document needs to consider.

[0090] Optionally, in this embodiment, the main process of the retrieval system may include, but is not limited to, two aspects. The first aspect is the system background operation process, which primarily aims to implement and maintain the knowledge base. This process may include, but is not limited to, the following steps:

[0091] S1. After the timer triggers the scheduled task, it calls the API of the enterprise assets to obtain the application information ledger and the content of the enterprise-level technical specification document.

[0092] S2. The timer scheduler transmits the acquired content to the preprocessing module;

[0093] S3, the preprocessing module cleans and segments the content of the application information ledger and enterprise-level technical specification documents, and then submits them to the vectorization engine;

[0094] S4. The vectorization engine vectorizes the preprocessed content and then writes the vectorized content into the vector database to keep the knowledge base fresh.

[0095] On the other hand, there is the system front-end operation flow. Through the system front-end operation flow, users can retrieve the technical specifications involved in the technical documents. This flow may include, but is not limited to, the following steps:

[0096] S① The user inputs a technical document (i.e., the target technical document) + application name (i.e., the target application identifier). The core module first calls the application-level specification model (i.e., the first generation model) through the application name.

[0097] S② The application-level specification model retrieves application information and full technical specification content by searching the vector database (i.e., the application attribute database and the technical specification database) by application name;

[0098] S③ The application-level specification model is based on application information. It selects the specification items related to the application from the full set of technical specifications and outputs them to the application-level specification knowledge base (i.e., the reference specification set).

[0099] S④ The core module inputs the content of the technical document into the core model (i.e., the second generation model) and designates the application-level specification knowledge base as the external knowledge base of the core model;

[0100] S⑤ The core model, based on the content of the technical document, retrieves all the specification entries involved in the technical document from the application-level specification knowledge base (to obtain the target specification set) and outputs it to the user.

[0101] The system described above primarily addresses the need for rapid retrieval of enterprise-level technical specification documents during the construction of subsystems (applications) within enterprise-level IT systems. This allows for the prior acquisition of all relevant specification items before application system design. By applying the methods and system described in this application, the need for lengthy post-application specification checks and rectifications can be avoided, as well as preventing the application system from being at risk for extended periods (due to design non-compliance resulting in system risks). Users (typically application architects) only need to provide technical documents to obtain a list of relevant specification items. By using an external enterprise knowledge base as a plug-in knowledge base for the model, retraining existing large language models is avoided, saving significant costs and enabling rapid deployment. The configuration information required for the methods and system described in this application (such as technical specifications and application information) is all internal to the enterprise. The entire system is built with the enterprise-level system as its implementation goal. System deployment and maintenance are handled by the enterprise itself, without the need for individual applications to concern themselves with the process. System usage is application-centric; application architects only need to focus on using the applications they are responsible for. The system's ecosystem is enterprise-centric, allowing applications to exchange usage experiences, opinions, and suggestions within the enterprise through forums and other means, free from external information interference.

[0102] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0103] Example 2

[0104] This application also provides an application function information retrieval device. It should be noted that the application function information retrieval device of this application can be used to execute the application function information retrieval method provided in this application. The application function information retrieval device provided in this application is described below.

[0105] According to embodiments of this application, an apparatus for implementing the above-described method for retrieving application function information is also provided, such as... Figure 4 As shown, the device can be applied to, but is not limited to, application information retrieval devices, and the device includes:

[0106] The receiving module 402 is used to receive a retrieval request initiated by the requester, wherein the retrieval request is used to request the retrieval of application function information required to implement the target function in the target application, and the application function information is used to indicate the technical specifications required to implement the target function in the target application.

[0107] The first generation module 404 is used to respond to the retrieval request and generate a reference specification set for the target application through the first generation model. The first generation model is obtained by training the first initial generation model using an application attribute library and a technical specification library that annotates application specification set samples of each application in multiple applications. The application attribute library stores the application attributes of each application, and the technical specification library stores the technical specifications that each application attribute needs to follow when implementing the application. The application specification set samples include the technical specifications that the corresponding application needs to follow when implementing it.

[0108] The second generation module 406 is used to generate a target specification set based on the reference specification set and the target technical document of the target application through the second generation model. The second generation model is obtained by training the second initial generation model with application specification set samples and functional technical documents that are labeled with functional specification set samples of each function in each application. The functional technical documents record the implementation method of the corresponding function, the target technical documents record the implementation method of the target function in the target application, and the functional specification set samples include the technical specifications that need to be followed when implementing the corresponding function in the corresponding application.

[0109] The sending module 408 is used to send the target specification set to the requester.

[0110] The application function information retrieval device provided in this application embodiment, upon receiving a retrieval request from a requester for application function information required to implement a target function in a target application, first responds to the retrieval request and generates a reference specification set for the target application through a first generation model. The first generation model is trained using an application attribute library and a technical specification library that annotates application specification set samples from multiple applications. Then, a target specification set is generated using a second generation model based on the reference specification set and the target technical document of the target application. The second generation model is trained using an application specification set sample annotated with function specification set samples from various applications and the functional technical document. Finally, the target specification set is sent to the requester. This achieves the goal of quickly finding the technical specifications required to implement the target function of the target application recorded in the target technical document through the first and second generation models, the application attribute library, and the technical specification library. This improves the retrieval efficiency of application function-related technical specifications and solves the technical problem of low retrieval efficiency caused by the need to manually review a large number of enterprise-level technical specification documents.

[0111] Optionally, in the application function information retrieval device provided in this application embodiment, the first generation module includes: a first acquisition unit, used to acquire the target application identifier of the target application; a first addition unit, used to add the target application identifier to the target position of the first preset prompt word to obtain the first target prompt word, wherein the first preset prompt word is used to describe the requirements for generating technical specifications corresponding to the content added at the target position using natural language; and a first execution unit, used to execute the first target prompt word through the first generation model using the application attribute library and technical specification library to obtain a reference specification set, wherein the first generation model is obtained by training the first model function of the first initial generation model using application prompt words labeled with application specification set samples of various applications, and the first model function includes using the application attribute library and technical specification library to execute the received prompt words that conform to the semantic expression of the application prompt words to obtain the application specification set, wherein the application prompt words are used to describe the requirements for generating technical specifications corresponding to the corresponding application using natural language.

[0112] Optionally, in the application function information retrieval device provided in the embodiments of this application, the first execution unit is further configured to: respond to the first target prompt word through the first generation model, and extract the target application identifier from the target position in the first target prompt word; use the application attribute library as a knowledge base through the first generation model, and extract the target application attribute corresponding to the target application identifier from the application attribute library; use the technical specification library as a knowledge base through the first generation model, and extract the technical specifications that conform to the target application attributes from the technical specification library to obtain a reference specification set.

[0113] Optionally, in the application function information retrieval device provided in the embodiments of this application, the first execution unit is further configured to: use the technical specification library as a knowledge base through the first generation model, extract technical specifications that conform to the target application attributes from the technical specification library to obtain various reference specification entries; and convert each reference specification entry into a corresponding specification vector through the first generation model to obtain a reference specification set including multiple reference specification vectors.

[0114] Optionally, in the application function information retrieval device provided in this application embodiment, the second generation module includes: a second acquisition unit, used to acquire a target technical document; a second addition unit, used to add the target technical document to the reference position of a second preset prompt word to obtain a second target prompt word, wherein the second preset prompt word is used to describe the generation of technical specifications corresponding to the content added at the reference position using natural language; and a second execution unit, used to execute the second target prompt word using a reference specification set through a second generation model to obtain a target specification set, wherein the second generation model is obtained by training the second model function of the second initial generation model using functional prompt words labeled with functional specification set samples of each function, and the second model function includes executing the received prompt words that conform to the semantic expression of the functional prompt words using application specification set samples to obtain a functional specification set, wherein the functional prompt words are used to describe the requirements for generating technical specifications corresponding to the corresponding functions using natural language.

[0115] Optionally, in the application function information retrieval device provided in the embodiments of this application, the second execution unit is further configured to: respond to the second target prompt word through the second generation model, extract the target technical document from the reference position in the second target prompt word; use the reference specification set as a knowledge base through the second generation model, extract the technical specifications that conform to the target technical document from the reference specification set, and obtain the target specification set.

[0116] Optionally, in the application function information retrieval device provided in this application embodiment, the application function information retrieval device further includes: an acquisition module, used to acquire application attribute update information and technical specification update information of each application from the application database through the target interface before generating the reference specification set of the target application through the first generation model, wherein the application database is used to store all application running information of multiple applications, the application attribute update information is the information on the changes in the current application attributes of each application relative to the application attributes stored in the application attribute library, and the technical specification update information is the information on the changes in the current technical specifications that each application attribute needs to follow when implementing the application relative to the technical specifications stored in the technical specification library; a conversion module, used to convert the application attribute update information into application attribute vectors and convert the technical specification update information into technical specification vectors; and an update module, used to update the current application attribute library using the application attribute vectors and update the current technical specification library using the technical specification vectors.

[0117] It should be noted that the receiving module 402 to the sending module 408 mentioned above correspond to steps S202 to S208 in Embodiment 1. Each module and the corresponding step implement the same instance and application scenario, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware components or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above modules can also be part of the device and run in the computer terminal 10 provided in Embodiment 1.

[0118] Example 3

[0119] Embodiments of this application may provide an electronic device. Figure 5 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 5 As shown, the electronic device may include: one or more ( Figure 5 (Only one is shown) processor 502, memory 504, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0120] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the methods and apparatus in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the above-described methods. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0121] The processor can access information and application programs stored in memory via a transmission device to perform the following steps: receiving a retrieval request initiated by a requester, wherein the retrieval request is used to request the retrieval of application function information required to implement the target function in the target application, and the application function information is used to indicate the technical specifications required to implement the target function in the target application; responding to the retrieval request, generating a reference specification set for the target application through a first generative model, wherein the first generative model is trained using an application attribute library and a technical specification library that annotates application specification set samples of various applications from multiple applications, the application attribute library storing application attributes of each application, and the technical specification library storing application attributes of each application. The application implementation must adhere to specific technical specifications. The application specification set sample includes the technical specifications that the corresponding application must follow during implementation. A target specification set is generated using a second generation model based on the reference specification set and the target application's target technical documents. This second generation model is trained using an application specification set sample annotated with functional specification sets for each function in each application, along with functional technical documents. The functional technical documents record the implementation methods of the corresponding functions, and the target technical documents record the implementation methods of the target functions in the target application. The functional specification set sample includes the technical specifications that must be followed when implementing the corresponding functions in the corresponding application. The target specification set is then sent to the requesting party.

[0122] The processor can also invoke information and applications stored in the memory via a transmission device to perform the following steps: obtaining the target application identifier of the target application; adding the target application identifier to the target position of the first preset prompt word to obtain the first target prompt word, wherein the first preset prompt word is used to describe the requirement for generating technical specifications corresponding to the content added at the target position using natural language; executing the first target prompt word through the first generation model using the application attribute library and technical specification library to obtain a reference specification set, wherein the first generation model is obtained by training the first model function of the first initial generation model using application prompt words labeled with application specification set samples of various applications, and the first model function includes executing the received prompt words that conform to the semantic expression of the application prompt words to obtain the application specification set, wherein the application prompt words are used to describe the requirement for generating technical specifications corresponding to the corresponding application using natural language.

[0123] The processor can also invoke information and applications stored in the memory via the transmission device to perform the following steps: responding to the first target prompt word through the first generation model, extracting the target application identifier from the target location in the first target prompt word; using the application attribute library as a knowledge base through the first generation model, extracting the target application attribute corresponding to the target application identifier from the application attribute library; using the technical specification library as a knowledge base through the first generation model, extracting the technical specifications that conform to the target application attributes from the technical specification library, and obtaining a reference specification set.

[0124] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: using the technical specification library as a knowledge base through the first generation model, extracting technical specifications that conform to the target application attributes from the technical specification library to obtain various reference specification entries; converting each reference specification entry into a corresponding specification vector through the first generation model to obtain a set of reference specifications including multiple reference specification vectors.

[0125] The processor can also invoke information and applications stored in the memory via a transmission device to perform the following steps: obtaining a target technical document; adding the target technical document to a reference position of a second preset prompt word to obtain a second target prompt word, wherein the second preset prompt word is used to describe the generation of technical specifications corresponding to the content added at the reference position using natural language; executing the second target prompt word using a reference specification set through a second generation model to obtain a target specification set, wherein the second generation model is obtained by training the second model function of the second initial generation model using functional prompt words labeled with functional specification set samples of each function, and the second model function includes executing the received prompt words that conform to the semantic expression of the functional prompt words using application specification set samples to obtain a functional specification set, wherein the functional prompt words are used to describe the requirements for generating technical specifications corresponding to the corresponding functions using natural language.

[0126] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: responding to the second target prompt word through the second generation model, extracting the target technical document from the reference position in the second target prompt word; using the reference specification set as a knowledge base through the second generation model, extracting the technical specifications that conform to the target technical document from the reference specification set, and obtaining the target specification set.

[0127] The processor can also invoke information and applications stored in memory via a transmission device to perform the following steps: Retrieve application attribute update information and technical specification update information for each application from the application database through a target interface. The application database stores all application runtime information for multiple applications. The application attribute update information represents changes to the current application attributes of each application relative to the application attributes stored in the application attribute library. The technical specification update information represents changes to the current technical specifications that each application attribute must follow during implementation relative to the technical specifications stored in the technical specification library. Convert the application attribute update information into application attribute vectors and the technical specification update information into technical specification vectors. Update the current application attribute library using the application attribute vectors and update the current technical specification library using the technical specification vectors.

[0128] In this embodiment, upon receiving a retrieval request from a requester for application function information required to implement a target function in a target application, the system first responds to the retrieval request and generates a reference specification set for the target application using a first generation model. This first generation model is trained on an application attribute library and a technical specification library that annotates application specification set samples from multiple applications. Then, a target specification set is generated using a second generation model based on the reference specification set and the target technical document of the target application. This second generation model is trained on a second initial generation model using application specification set samples annotated with function specification set samples from various applications and the functional technical document. Finally, the target specification set is sent to the requester. This achieves the goal of quickly finding the technical specifications required to implement the target function of the target application recorded in the target technical document using the first and second generation models, the application attribute library, and the technical specification library. This improves the retrieval efficiency of application function-related technical specifications and solves the problem of low retrieval efficiency caused by the need to manually review a large number of enterprise-level technical specification documents.

[0129] Those skilled in the art will understand that Figure 5 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones (such as Android phones, iOS phones, etc.), tablets, PDAs, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 5 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 5 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 5 The different configurations shown.

[0130] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0131] Example 4

[0132] Embodiments of this application also provide a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the application function information retrieval method provided in Embodiment 1.

[0133] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0134] This application also provides a computer program product, which, when executed on a data processing device, is suitable for performing the steps of a method for retrieving application function information.

[0135] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0136] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0137] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0138] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0139] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0140] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0141] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for retrieving application function information, characterized in that, The method, applied to an application information retrieval device, includes: Receive a search request initiated by a requester, wherein the search request is used to request the retrieval of application function information required to implement the target function in the target application, and the application function information is used to indicate the technical specifications required to implement the target function in the target application; In response to the retrieval request, a reference specification set for the target application is generated through a first generation model. The first generation model is trained using an application attribute library and a technical specification library that annotates application specification set samples of each application in multiple applications. The application attribute library stores the application attributes of each application, and the technical specification library stores the technical specifications that each application attribute must follow when implementing the application. The application specification set samples include the technical specifications that the corresponding application must follow when implementing it. A target specification set is generated by a second generative model based on the reference specification set and the target technical document of the target application. The second generative model is obtained by training a second initial generative model using the application specification set sample, which is labeled with the functional specification set sample of each function in each application, and the functional technical document. The functional technical document records the implementation method of the corresponding function, and the target technical document records the implementation method of the target function in the target application. The functional specification set sample includes the technical specifications that need to be followed when implementing the corresponding function in the corresponding application. The target specification set is sent to the requester.

2. The method according to claim 1, characterized in that, The step of generating the reference specification set for the target application through the first generative model includes: Obtain the target application identifier of the target application; The target application identifier is added to the target position of the first preset prompt word to obtain the first target prompt word, wherein the first preset prompt word is used to describe the technical specification requirements corresponding to the content added at the target position using natural language; The reference specification set is obtained by executing the first target prompt word using the application attribute library and the technical specification library through the first generation model. The first generation model is trained by using application prompt words labeled with application specification set samples of various applications to train the first model function of the first initial generation model. The first model function includes using the application attribute library and the technical specification library to execute the received prompt words that conform to the semantic expression of the application prompt words to obtain the application specification set. The application prompt words are used to describe the requirements for generating technical specifications corresponding to the corresponding application using natural language.

3. The method according to claim 2, characterized in that, The step of using the application attribute library and the technical specification library through the first generation model to execute the first target prompt word and obtain the reference specification set includes: The target application identifier is extracted from the target location in the first target prompt word by responding to the first target prompt word through the first generation model; The first generation model uses the application attribute library as a knowledge base to extract the target application attributes corresponding to the target application identifier from the application attribute library. The first generation model uses the technical specification library as a knowledge base, and extracts technical specifications that conform to the target application attributes from the technical specification library to obtain the reference specification set.

4. The method according to claim 3, characterized in that, The first generation model is used to treat the technical specification library as a knowledge base, and technical specifications that conform to the target application attributes are extracted from the technical specification library to obtain the reference specification set, including: The first generation model uses the technical specification library as a knowledge base, and extracts technical specifications that conform to the target application attributes from the technical specification library to obtain various reference specification entries. The first generation model converts each reference specification entry into a corresponding specification vector, resulting in a reference specification set that includes multiple reference specification vectors.

5. The method according to claim 1, characterized in that, The step of generating a target specification set based on the reference specification set and the target technical document of the target application using a second generation model includes: Obtain the target technical document; The target technical document is added to the reference position of the second preset prompt word to obtain the second target prompt word, wherein the second preset prompt word is used to use natural language to describe and generate technical specifications corresponding to the content added at the reference position; The second generation model executes the second target prompt word using the reference specification set to obtain the target specification set. The second generation model is trained on the second model function of the second initial generation model using function prompt words labeled with function specification set samples of each function. The second model function includes executing the received prompt words that conform to the semantic expression of the function prompt words using the application specification set samples to obtain the function specification set. The function prompt words are used to describe the requirements for generating technical specifications corresponding to the corresponding functions using natural language.

6. The method according to claim 5, characterized in that, The step of using the reference specification set through the second generation model to execute the second target prompt word, and obtaining the target specification set, includes: The target technical document is extracted from the reference position in the second target prompt word by responding to the second target prompt word through the second generation model; The second generation model uses the reference specification set as a knowledge base to extract technical specifications that conform to the target technical document from the reference specification set, thereby obtaining the target specification set.

7. The method according to claim 1, characterized in that, Before generating the reference specification set for the target application through the first generative model, the method further includes: The application attribute update information and technical specification update information of each application are obtained from the application database through the target interface. The application database is used to store all application running information of the multiple applications. The application attribute update information is the information on the changes that have occurred in the current application attributes of each application relative to the application attributes stored in the application attribute library. The technical specification update information is the information on the changes that have occurred in the current technical specifications that each application attribute needs to follow when implementing the application relative to the technical specifications stored in the technical specification library. The application attribute update information is converted into an application attribute vector, and the technical specification update information is converted into a technical specification vector; Update the current application attribute library using the application attribute vector, and update the current technical specification library using the technical specification vector.

8. A device for retrieving application function information, characterized in that, The device is used in application information retrieval equipment and includes: A receiving module is used to receive a retrieval request initiated by a requester, wherein the retrieval request is used to request the retrieval of application function information required to implement the target function in the target application, and the application function information is used to indicate the technical specifications required to implement the target function in the target application; The first generation module is used to respond to the retrieval request and generate a reference specification set for the target application through a first generation model. The first generation model is obtained by training a first initial generation model using an application attribute library and a technical specification library that annotates application specification set samples of each application in multiple applications. The application attribute library stores the application attributes of each application, and the technical specification library stores the technical specifications that each application attribute needs to follow when implementing the application. The application specification set samples include the technical specifications that the corresponding application needs to follow when implementing it. The second generation module is used to generate a target specification set based on the reference specification set and the target technical document of the target application using a second generation model. The second generation model is obtained by training the second initial generation model with the application specification set sample, which is labeled with the functional specification set sample of each function in each application, and the functional technical document. The functional technical document records the implementation method of the corresponding function, and the target technical document records the implementation method of the target function in the target application. The functional specification set sample includes the technical specifications that need to be followed when implementing the corresponding function in the corresponding application. The sending module is used to send the target specification set to the requester.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the computer-readable storage medium is located to perform the application function information retrieval method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program executes the application function information retrieval method according to any one of claims 1 to 7 when it runs.

11. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the application function information retrieval method according to any one of claims 1 to 7.

Images