Code-free application generation method and system based on artificial intelligence

Abstract

The invention relates to the technical field of data processing, in particular to a code-free application generation method and system based on artificial intelligence, and the method comprises the steps: constructing a front-end module library, obtaining the data information of a module established by each user in code-free application development and the occurrence frequency of the data information, counting the number of established modules, complex coefficients and development efficiency of each user during code-free front-end application development, calculating and normalizing influence coefficients of each user, then classifying to form clusters, acquiring update coefficients of various data information of each cluster, sorting the update coefficients, and storing the sorted update coefficients in a database; and carrying out module library building on the data information corresponding to the sorted preset number of updating coefficients to complete updating of the module library. Through the influence coefficient of the user, the user group which contributes to the product greatly is identified, and the development and optimization of the module are guided, so that the actual demand of the user is met, and the development quality is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology. More specifically, the present invention relates to a method and system for generating code-free applications based on artificial intelligence. Background Art

[0002] Code-free application generation refers to the process of directly generating software applications based on user requirements using artificial intelligence technology in a non-programming way. This technology lowers the threshold for software development, allowing users with non-technical backgrounds to quickly build and deploy applications.

[0003] The existing Chinese patent application document with publication number CN117331559A discloses a method for front-end codeless generation of application interface, which belongs to the field of application interface generation. A method for front-end codeless generation of application interface, comprising the following steps: S1: receiving front-end application interface generation request information and analyzing it, S2: sending the drawn initial front-end application interface to the user end for confirmation, S3: making corresponding adjustments based on the initial front-end application interface, S4: generating an updated front-end application interface for the initial front-end application interface, S5: testing and optimizing the functions of the final front-end application interface, S6: final processing of the display background of the front-end application interface. The present invention solves the problem that the existing patent generates application interfaces through code, which is costly and greatly increases the construction time and difficulty.

[0004] The application document receives and analyzes the front-end application interface generation request information, sends the initial interface to the user for confirmation, and then makes adjustments based on the feedback to generate an updated interface, test and optimize the functions, thereby simplifying the traditional coding process, saving costs, and greatly improving the efficiency of application development. The final front-end application interface functions are more complete, ensuring the feasibility of the front-end application interface. Currently, there are tools that implement front-end application development by dragging and dropping front-end modules. However, when generating codeless applications based on front-end modules, since the front-end models are often provided by the company that develops the tools, when conducting codeless development, not only professional developers but also new users need to be faced. In addition, when the company creates front-end modules, it is often subject to greater thinking limitations, which makes it difficult for the front-end modules to meet the differentiated needs of users and the user experience is poor. When tool development companies develop new front-end modules, it is difficult to obtain effective front-end module construction guidance, which reduces the user experience. Summary of the invention

[0005] In order to solve the problem that the front-end module creation is often subject to large thinking limitations, and at the same time, it is difficult to obtain effective front-end module construction guidance when developing a new front-end module, which reduces the user experience, the present invention provides solutions in the following aspects.

[0006] In the first aspect, a method for generating a codeless application based on artificial intelligence includes: building a front-end module library, obtaining data information of modules established by each user in the development of codeless applications and the number of occurrences of the data information, wherein the data information includes: attribute sequence, function sequence, input data and output data pair sequence; counting the number of modules established by each user in the development of codeless front-end applications, complexity coefficient and development efficiency, and calculating the influence coefficient of each user; normalizing the influence coefficient of each user and classifying them to obtain multiple clusters, obtaining the update coefficient of each data information of each cluster, and sorting the update coefficients in order from large to small, and building a module library for the data information corresponding to the preset number of update coefficients after sorting, and completing the update of the module library; wherein the update coefficient of the data information satisfies the following relationship: Where P n，w，x Represents the update coefficient of the xth data of the Wth data information in the nth cluster, R′ m，n Represents the frequency value of the xth data item in the Wth data item in the nth cluster, R′ m Represents the total frequency value of the Wth item of data information in the nth cluster, F n represents the mean of the influence coefficients of all users in the nth cluster, Q n Represents the weight of the nth cluster.

[0007] The effect is: by analyzing user activities in codeless application development, the system can identify the module attributes, functions and data interfaces that users need most, thereby realizing personalized module development. According to changes in user needs, the module library can be dynamically updated to ensure that the product always meets the latest needs of the market and users.

[0008] Preferably, the functional sequence is a text description of the specific functions that the module needs to implement, and the functional sequence includes the behavior of the user interface, data processing logic, and interaction with other modules;

[0009] Among them, it is necessary to define the input data and output data pairs of the module, and one of the input interfaces / output interfaces of the module is interconnected with one of the input and output interfaces / input interfaces of other modules to form a complete input interface / output interface.

[0010] Preferably, obtaining the number of occurrences of the data information includes:

[0011] Count the number of occurrences of each attribute, perform semantic extraction based on the function using a pre-trained semantic network, calculate the cosine similarity of the semantic vectors between the text descriptions of any two functions, and if the cosine similarity is less than a preset similarity threshold, it means that the two functions belong to the same function, otherwise, they belong to different functions, and count the number of occurrences;

[0012] Based on the input data, in response to the data of the input data and output data pair being opposite, they are not considered as the same input data and output data pair sequence, wherein the input data / output data pair sequence refers to the type of the input data and output data of the module, and the types include: character, integer and Boolean types.

[0013] The effect is: by counting the number of occurrences of attributes and functions, the most commonly used features and functional requirements of users in codeless application development can be accurately identified. Using pre-trained semantic networks to perform semantic extraction and comparison of functional descriptions, similar functions can be classified to avoid duplicate development and optimize the structure of the module library. Identifying and developing the most commonly used and most needed functions can improve development efficiency because development resources are concentrated on the features that are most valuable to users. By identifying and merging similar functions, redundancy in the module library can be reduced, making the library neater and easier to manage.

[0014] Preferably, the influence coefficient includes:

[0015] Taking any user as the target user, calculate the product of the number of modules constructed by the target user in each codeless front-end application development and the complexity coefficient between the constructed modules to obtain the development complexity. The product of the largest development complexity among all target users and the average efficiency of the target user in all codeless front-end application development is taken as the influence coefficient of the target user.

[0016] The effect is: by calculating the product of the number of modules built by the user in each development and the complexity coefficient, the user's contribution to the front-end module library can be quantified. By comparing the user's maximum development complexity with the average development efficiency, the key users who have a significant impact on the construction of the module library can be identified. Based on the key users, it can be ensured that the update and expansion of the module library can cover more complex application scenarios, thereby improving the development quality.

[0017] Preferably, the influence coefficient also satisfies the following relationship:

[0018]

[0019] In the formula, F i represents the user influence coefficient of the i-th user, J represents the total number of times the user has developed front-end applications, and E i，j G represents the number of modules built by the i-th user in the j-th codeless front-end application development.i，j represents the complexity coefficient between modules constructed by the i-th user during the j-th codeless front-end application development, H i Represents the mean efficiency of the i-th user in front-end application development.

[0020] Preferably, the complexity coefficient includes:

[0021] Taking any user as the target user, the total number of links between construction modules when the target user develops the j-th codeless front-end application is used as the complexity coefficient when the target user develops the j-th codeless front-end application.

[0022] Preferably, the efficiency mean includes:

[0023] Take any user as the target user, obtain the number of building modules when the target user develops the j-th codeless front-end application, and the effective development time when the target user develops the j-th codeless front-end application, use the exponential decay function to calculate the product of the effective development time and the hyperparameter, multiply the number of building modules by the result of the exponential decay function as the efficiency of the codeless front-end application development, calculate the average efficiency of all codeless front-end application development, and obtain the efficiency mean.

[0024] The effect is that by considering the number of building blocks and the effective development time, the user's efficiency in codeless front-end application development can be more accurately evaluated. By improving development efficiency, users can complete their application development faster, thereby improving user satisfaction.

[0025] In the second aspect, a codeless application generation system based on artificial intelligence includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned codeless application generation method based on artificial intelligence is implemented.

[0026] The present invention has the following effects:

[0027] 1. The present invention calculates the user influence coefficient and considers the development complexity coefficient and efficiency mean, thereby identifying the user groups that contribute more to the product, and guides the development and optimization of modules based on the data of these user groups, which not only improves the development efficiency, but also ensures that the newly developed modules can better meet the actual needs of users and improves the development quality.

[0028] 2. The present invention calculates the update coefficient of each data information of each cluster, thereby obtaining the priority of the update coefficient, concentrating resources on developing high-priority modules, more effectively utilizing limited development resources, reducing investment in low-priority functions or functions that are less needed by users, dynamically adjusting and updating the front-end module library, and more accurately meeting the needs of different user groups, thereby improving user experience. At the same time, through the cosine similarity of the semantic vectors between the text descriptions of the function sequence, similar functions can be merged to avoid redundancy. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0030] Figure 1 It is a method flow chart of steps S1 to S3 in the method for generating code-free applications based on artificial intelligence in an embodiment of the present invention.

[0031] Figure 2 It is a structural block diagram of an artificial intelligence-based codeless application generation system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0033] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0034] Description of specific implementation scenarios: In the codeless application development interface, users select components and drag different modules to realize the combination and link of different front-end modules. However, due to the different needs of different users, users require different modules when developing and generating codeless applications. Although most module functions can often meet user needs, the characteristic of codeless application generation is the personalized differences of generated applications, which requires continuous expansion of the front-end module library to meet the personalized needs of users.

[0035] Reference Figure 1 The method for generating codeless applications based on artificial intelligence includes steps S1 to S3, which are as follows:

[0036] S1: Build a front-end module library to obtain the data information and the number of occurrences of the data information when each user creates a module during codeless application development, where the data information includes but is not limited to: attribute sequence, function sequence, input data and output data pair sequence.

[0037] It should be noted that when the front-end module library is expanded, it is necessary to record the module requirements of users during the front-end codeless application development. Therefore, this solution selects users to establish modules when developing front-end codeless applications to record the development requirements of users, wherein the attributes / functions / input data and output data of the front-end modules are marked by users, thereby obtaining the attribute sequence and function sequence and input data and output data pair sequence corresponding to the t-th module of the i-th user in the j-th codeless front-end application development.

[0038] The function sequence is a textual description of the specific functions that the module needs to implement. The function sequence includes the behavior of the user interface, data processing logic, and interaction with other modules;

[0039] Among them, it is necessary to define the input data and output data pairs of the module, and one of the module input interface / output interface is connected to one of the input and output interfaces / input interfaces of other modules to form a complete input interface / output interface.

[0040] To explain further, to establish the data information of a module, you first need to determine what data the module receives as input data, that is, these input data can be user input, output of other modules or system data; other modules include but are not limited to: visual components, layout components, functional components, interactive components, data components, blank modules and expandable modules; secondly, determine what data will be generated as output data after the module is processed. Similarly, these output data can be used as input by other modules or used to update the user interface.

[0041] Define input and output interfaces for modules and link the modules. Through the links between modules, a complete input and output interface chain is formed. In this way, data can flow from one module to another, realizing the logical flow of the application.

[0042] Get the number of occurrences of data information, including steps:

[0043] Count the number of occurrences of each attribute, perform semantic extraction based on the function using a pre-trained semantic network, calculate the cosine similarity of the semantic vectors between the text descriptions of any two functions, and if the cosine similarity is less than a preset similarity threshold, it means that the two functions belong to the same function, otherwise, they belong to different functions, and count the number of occurrences;

[0044] Based on the input data, in response to the data of the input data and output data pair being opposite, they are not considered the same input data and output data pair sequence, wherein the input data / output data pair sequence refers to the type of the input data and output data of the module, and the types include but are not limited to: character, integer and Boolean types.

[0045] It should be noted that the preset similarity threshold is 0.2, which can be adjusted by the implementer according to the specific implementation scenario. By using the pre-trained semantic network to perform semantic extraction and comparison of function descriptions, similar functions can be classified to avoid duplicate development. While meeting the actual needs of users, it also improves development efficiency and user experience, optimizes the module library, and enhances the personalization and efficiency of codeless application generation.

[0046] Further analysis shows that if a new front-end module is built for each blank module of each user, it will lead to inefficiency, which may lead to uneven resource allocation and fail to meet the needs of all users, thus affecting user satisfaction. Therefore, we choose to classify users in order to more effectively meet the needs of different user groups. Through classification, we can identify the user groups that are most valuable to the company, that is, those users with larger module library requirements and high development efficiency. The specific steps are as follows.

[0047] S2: Count the number of modules, complexity coefficient, and development efficiency of each user when developing codeless front-end applications, and calculate the impact coefficient of each user.

[0048] Impact factors, including:

[0049] Taking any user as the target user, calculate the product of the number of modules constructed by the target user in each codeless front-end application development and the complexity coefficient between the constructed modules to obtain the development complexity. The product of the largest development complexity among all target users and the average efficiency of the target user in all codeless front-end application development is taken as the influence coefficient of the target user.

[0050] Specifically, the influence coefficient satisfies the following relationship:

[0051]

[0052] In the formula, F i represents the influence coefficient of the i-th user, E i，j G represents the number of modules built by the i-th user in the j-th codeless front-end application development. i，j represents the complexity coefficient between modules constructed by the i-th user during the j-th codeless front-end application development, H i represents the mean efficiency of the i-th user in front-end application development, max represents the maximum function, and J represents the total number of times the user develops front-end applications.

[0053] That is to say, the larger the number of modules built during codeless front-end application development, the more likely the user is to have a larger module library requirement and may be a potential high-quality user. If the development efficiency is high at the same time and the use of modules during codeless front-end application development is more complex, the user is a high-quality potential customer.

[0054] By classifying users and user impact coefficients, resource allocation is optimized and construction efficiency is improved, which helps reduce user churn, improve user satisfaction, and ensure the effective use of company resources.

[0055] In addition, another embodiment further includes:

[0056] The influence coefficient also satisfies the following relationship:

[0057]

[0058] In the formula, F i represents the user influence coefficient of the i-th user, J represents the total number of times the user has developed front-end applications, and E i，j G represents the number of modules built by the i-th user in the j-th codeless front-end application development. i，j represents the complexity coefficient between modules constructed by the i-th user during the j-th codeless front-end application development, H i Represents the mean efficiency of the i-th user in front-end application development.

[0059] That is, by using the geometric mean, the impact of extreme values ​​is reduced, ensuring that the contribution of each development activity is fairly considered. This is suitable for processing data sets containing extreme values, and implementers can choose the calculation method of the impact coefficient based on the characteristics of the data information obtained.

[0060] The complexity coefficient is:

[0061] Taking any user as the target user, the total number of links between construction modules when the target user develops the j-th codeless front-end application is used as the complexity coefficient when the target user develops the j-th codeless front-end application.

[0062] In addition, the number of times the module is reused can be used as the complexity coefficient. A high degree of reuse means that the module is well designed, but it may also increase the complexity of the system. The number of input and output interfaces of the module can also be counted. In other words, the more interfaces, the more complex the interaction between modules.

[0063] Efficiency mean, including steps:

[0064] Take any user as the target user, obtain the number of building modules when the target user develops the j-th codeless front-end application, and the effective development time when the target user develops the j-th codeless front-end application, use the exponential decay function to calculate the product of the effective development time and the hyperparameter, multiply the number of building modules by the result of the exponential decay function as the efficiency of the codeless front-end application development, calculate the average efficiency of all codeless front-end application development, and obtain the efficiency mean.

[0065] Specifically, the efficiency satisfies the following relationship:

[0066] H i，j =E i，j ×exp(-K i，j ×L);

[0067] In the formula, H i，j represents the efficiency of the i-th user in the j-th codeless front-end application development, E i，j K represents the number of modules built by the i-th user in the j-th codeless front-end application development. i，j represents the effective development time of the i-th user in the j-th codeless front-end application development, L represents a hyperparameter, and exp() represents an exponential function with a natural number e as the base.

[0068] That is to say, the empirical value of L is 0.1, which can be adjusted by the implementer according to the specific implementation scenario. Since the user development efficiency is more related to the user, the mean efficiency value should be taken.

[0069] S3: Normalize the influence coefficient of each user and classify them to obtain multiple clusters, obtain the update coefficient of each data information of each cluster, and sort the update coefficients in order from large to small, build a module library for the data information corresponding to the preset number of update coefficients after sorting, and complete the update of the module library.

[0070] In this embodiment, the mean shift algorithm is used to classify the user influence coefficients of all users, wherein the clustering radius is 0.05.

[0071] Specifically, the update coefficient of the data information satisfies the following relationship:

[0072]

[0073] Where P n，w，x Represents the update coefficient of the xth data of the Wth data information in the nth cluster, R′ m，n Represents the frequency value of the xth data item in the Wth data item in the nth cluster, R′ m Represents the total frequency value of the Wth item of data information in the nth cluster, F nrepresents the mean of the influence coefficients of all users in the nth cluster, Q n Represents the weight of the nth cluster.

[0074] That is to say, The larger the value, the higher the module development demand of users in this category. In this case, you can consider building this module for codeless application development first. n is a hyperparameter sequence, that is, a weight is assigned to each category of users. The weight is determined by the company according to the specific implementation scenario, and then the cumulative value of the Q value corresponding to all categories is required to be 1. N represents the total number of clusters, which can be adjusted by implementers according to specific circumstances.

[0075] Further explanation: the preset number is to sort the update coefficients in descending order, obtain the data information corresponding to the first 20% of the update coefficients to build the module library, and complete the update of the module library.

[0076] The present invention also provides a code-free application generation system based on artificial intelligence. Figure 2 As shown, the system includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method for generating a code-free application based on artificial intelligence according to the first aspect of the present invention is implemented.

[0077] The system also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0078] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that may be used to store the required information and may be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0079] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0080] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A code-free application generation method based on artificial intelligence, characterized in that: include: Construct a front-end module library, obtain data information and the number of occurrences of the data information of the modules established by each user in the codeless application development, wherein the data information includes: attribute sequence, function sequence, input data and output data pair sequence; Statistics on the number of modules, complexity coefficient and development efficiency of each user when developing codeless front-end applications, and calculate the impact coefficient of each user; Normalize the influence coefficient of each user and classify them to obtain multiple clusters, obtain the update coefficient of each data information of each cluster, and sort the update coefficients in descending order, build a module library for the data information corresponding to the preset number of update coefficients after sorting, and complete the update of the module library; The update coefficient of the data information satisfies the following relationship: Where P n，w，x Represents the update coefficient of the xth data of the Wth data information in the nth cluster, R′ n，w，x Represents the frequency value of the xth data item in the Wth data item in the nth cluster, R′ n，w Represents the total frequency value of the Wth item of data information in the nth cluster, F n represents the mean of the influence coefficients of all users in the nth cluster, Q n Represents the weight of the nth cluster.

2. The method for generating code-free applications based on artificial intelligence according to claim 1, characterized in that: The functional sequence is a textual description of the specific functions that the module needs to implement, including the behavior of the user interface, data processing logic, and interaction with other modules; Among them, it is necessary to define the input data and output data pairs of the module, and one of the input interfaces / output interfaces of the module is interconnected with one of the input and output interfaces / input interfaces of other modules to form a complete input interface / output interface.

3. The method for generating code-free applications based on artificial intelligence according to claim 1, characterized in that: Obtain the number of occurrences of the data information, including: Count the number of occurrences of each attribute, perform semantic extraction based on the function using a pre-trained semantic network, calculate the cosine similarity of the semantic vectors between the text descriptions of any two functions, and if the cosine similarity is less than a preset similarity threshold, it means that the two functions belong to the same function, otherwise, they belong to different functions, and count the number of occurrences; Based on the input data, in response to the data of the input data and output data pair being opposite, they are not considered as the same input data and output data pair sequence, wherein the input data / output data pair sequence refers to the type of the input data and output data of the module, and the types include: character, integer and Boolean types.

4. The method for generating code-free applications based on artificial intelligence according to claim 1, characterized in that: The influence coefficient includes: Taking any user as the target user, calculate the product of the number of modules constructed by the target user in each codeless front-end application development and the complexity coefficient between the constructed modules to obtain the development complexity. The product of the largest development complexity among all target users and the average efficiency of the target user in all codeless front-end application development is taken as the influence coefficient of the target user.

5. The method for generating code-free applications based on artificial intelligence according to claim 1, characterized in that: The influence coefficient also satisfies the following relationship: In the formula, F i represents the user influence coefficient of the i-th user, J represents the total number of times the user has developed front-end applications, and E i，j G represents the number of modules built by the i-th user in the j-th codeless front-end application development. i，j represents the complexity coefficient between modules constructed by the i-th user during the j-th codeless front-end application development, H i Represents the mean efficiency of the i-th user in front-end application development.

6. The method for generating codeless applications based on artificial intelligence according to claim 4, characterized in that: The complexity coefficients include: Taking any user as the target user, the total number of links between construction modules when the target user develops the j-th codeless front-end application is used as the complexity coefficient when the target user develops the j-th codeless front-end application.

7. The method for generating code-free applications based on artificial intelligence according to claim 4, characterized in that: The efficiency mean includes: Take any user as the target user, obtain the number of building modules when the target user develops the j-th codeless front-end application, and the effective development time when the target user develops the j-th codeless front-end application, use the exponential decay function to calculate the product of the effective development time and the hyperparameter, multiply the number of building modules by the result of the exponential decay function as the efficiency of the codeless front-end application development, calculate the average efficiency of all codeless front-end application development, and obtain the efficiency mean.

8. The code-free application generation system based on artificial intelligence is characterized by: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method for generating a code-free application based on artificial intelligence according to any one of claims 1 to 7 is implemented.

Images