Low-code development intelligent auxiliary system and method based on machine learning

Abstract

The invention relates to the technical field of low-code development, and discloses a low-code development intelligent auxiliary system and method based on machine learning, and the method comprises the steps: collecting a target database, target department data and habit data, analyzing the use habit of a target user for a preliminary integration interface, forming an analysis result, forming the preliminary integration interface, and storing the preliminary integration interface; according to the low-code development intelligent auxiliary system and method based on machine learning, a low-code database is formed according to use functions of all departments, corresponding low codes are automatically extracted and the initial use interface is formed according to different departments, and the low-code development intelligent auxiliary system and method based on machine learning can be used for realizing low-code development according to use habit differences of users. According to the method and the device, the use interface is automatically adjusted, so that the user can quickly and effectively find the function key required by the user, the frequency of mistakenly clicking other function keys is reduced, the use habit of the user is better met, and the working efficiency of the user is improved.

Description

Technical Field

[0001] The present invention relates to the field of low-code development technology, and specifically to a low-code development intelligent assistance system and method based on machine learning. Background Art

[0002] The low-code development intelligent assistance system based on machine learning is a system that combines machine learning technology and low-code development platform to improve development efficiency and reduce development difficulty. It aims to use machine learning technology to intelligentize the low-code development process, allowing developers to create applications more quickly and accurately while reducing coding complexity. Low-code platforms usually allow users to build applications by dragging and dropping components and configuring parameters, while machine learning provides intelligent assistance for this process, which can accelerate the development and deployment of internal enterprise applications, reduce dependence on professional developers, help enterprises quickly create and optimize business process management systems, and provide customized application development assistance for different industries. By reducing development difficulty and improving efficiency, it provides developers with more powerful tools and allows non-technical personnel to participate in application development;

[0003] The system can automatically generate corresponding codes according to user needs and configurations, reducing manual coding work. It can provide intelligent recommendations for components, modules, or configurations based on user historical operations and needs, detect errors in the development process in real time, and provide repair suggestions or automatic repair functions. It can analyze user usage habits and project data, provide performance optimization and design improvement suggestions, and allow users to describe requirements in natural language. The system can automatically convert them into code or configuration. It uses historical project data and user behavior data to train machine learning models to improve the accuracy of recommended and automatically generated code, identify common development patterns and best practices, and use this information to optimize code generation and configuration suggestions. It continuously adjusts models and recommendation strategies based on user feedback and operating habits to improve the intelligence level of the system, integrates machine learning modules into the development environment, and provides real-time intelligent assistance functions.

[0004] The existing low-code development intelligent assistance systems and methods based on machine learning cannot form a low-code database based on the usage functions of all departments, cannot automatically extract the corresponding low-code and form the initial usage interface according to the different departments, and cannot automatically adjust the usage interface according to the differences in users' usage habits. As a result, users cannot find the function buttons they need to use more promptly and are prone to accidentally click on other function keys. This does not conform to the users' usage habits, reduces the users' work efficiency, and has certain limitations in its practicality. Summary of the invention

[0005] The present invention provides a low-code development intelligent auxiliary system and method based on machine learning, which has the advantages of forming a low-code database according to the usage functions of all departments, automatically extracting the corresponding low-code and forming an initial usage interface according to different departments, and automatically adjusting the usage interface according to the differences in the usage habits of users, so that users can find the function keys they need to use more quickly and effectively, reduce the number of times other function keys are accidentally clicked, and are more in line with the usage habits of users, improve the user's work efficiency, prevent people from different departments from entering the system of the department, and improve the confidentiality of department data. It solves the problems mentioned in the above background technology that it is impossible to form a low-code database according to the usage functions of all departments, it is impossible to automatically extract the corresponding low-code and form an initial usage interface according to different departments, and it is impossible to automatically adjust the usage interface according to the differences in the usage habits of users, so that users cannot find the function keys they need to use more promptly, are prone to accidentally click other function keys, are not in line with the user's usage habits, reduce the user's work efficiency, and its practicality has certain limitations.

[0006] The present invention provides the following technical solution: a low-code development intelligent auxiliary method based on machine learning, comprising:

[0007] Get the target database;

[0008] Get target department data;

[0009] Based on the target department data, a preliminary integration interface is formed through preliminary integration strategies;

[0010] Obtain target users’ habit data on the preliminary integration interface;

[0011] The habit data includes first habit data and second habit data;

[0012] If the target user's usage habits for the preliminary integration interface are the first habit data, no adjustment will be made;

[0013] If the target user's usage habit for the preliminary integration interface is the second habit data, the data adjustment strategy is executed to form adjustment data;

[0014] An adjustment usage interface is formed based on the adjustment data and the preliminary integration interface.

[0015] As a low-code development intelligent auxiliary method based on machine learning described in the present invention, wherein: the acquisition of the target database is specifically:

[0016] Obtain all departments of the target enterprise to form a department set;

[0017] Get all the functions used by each department in the department set, and form a function set for each department;

[0018] According to the department set and the function set corresponding to each department, integrate to form a department function set;

[0019] Obtain the low code corresponding to each element in the department function set respectively to form a target database.

[0020] As a low-code development intelligent auxiliary method based on machine learning described in the present invention, wherein: the acquisition of target department data is specifically:

[0021] Get the target department;

[0022] Extract the function set corresponding to the target department in the target database and define it as the target function set;

[0023] Extract the low code corresponding to the target function set and target department and set it as the target low code;

[0024] Get all target low codes to form a target low code collection.

[0025] As a low-code development intelligent auxiliary method based on machine learning described in the present invention, wherein: the preliminary integration strategy is specifically:

[0026] Get the target low-code collection;

[0027] Set the default interface layout;

[0028] Based on the default interface layout and the target low-code set, a new default interface layout is integrated and defined as the preliminary integration interface.

[0029] As a low-code development intelligent auxiliary method based on machine learning described in the present invention, wherein: the acquisition of the target user's habit data on the preliminary integration interface is specifically:

[0030] Obtain the target user's assigned business and set it as the target business;

[0031] Extract all elements corresponding to the target business in the target function set to form a business function set;

[0032] Implement habit analysis strategies;

[0033] The element corresponding to the first duration in the business function set is defined as the first business function;

[0034] The element corresponding to the query duration of the second duration in the business function set is defined as the second business function;

[0035] Get the total number of elements in the business function set, which is defined as the total number of functions;

[0036] Obtain the quantity of the first business function, which is defined as the first function quantity;

[0037] Obtain the number of the second business function and set it as the second function number;

[0038] If the first function number ÷ the total number of functions ≥ (the second function number ÷ the total number of functions) × (1 + 50%), it is determined to be the first habit data;

[0039] If the first function number ÷ the total number of functions < (the second function number ÷ the total number of functions) × (1 + 50%), it is determined to be the second habit data.

[0040] As a low-code development intelligent auxiliary method based on machine learning described in the present invention, wherein: the habit analysis strategy is specifically:

[0041] Extract the key position of each element in the business function set on the preliminary integration interface and set it as the target position;

[0042] Obtain all sub-businesses under the target business and define them as target sub-businesses;

[0043] Obtain all target positions corresponding to each target sub-business in the business function set and define them as target sub-positions;

[0044] Obtaining the order of each target sub-location corresponding to each target sub-business to form a sub-location order set;

[0045] Extract the first element and the last element in the sub-position sequence set, and define them as the sub-first element and the sub-last element respectively;

[0046] Execute the duration analysis strategy for each target sub-business separately.

[0047] As a low-code development intelligent auxiliary method based on machine learning described in the present invention, wherein: the duration analysis strategy is specifically:

[0048] Get the time point when the target user selects the sub-first element, and set it as the sub-first time point;

[0049] Get the time point when the target user selects the sub-end element, and set it as the sub-end time point;

[0050] Calculate the query duration, query duration = sub-end time point - sub-start time point;

[0051] Obtain all personnel corresponding to the target business in the target department and set them as reference personnel;

[0052] Obtain the query duration of each reference person and set it as the reference duration;

[0053] Calculate the judgment time, judgment time = sum of all reference times ÷ number of reference personnel;

[0054] If the query duration ≤ determination duration × (1 + 3%), the query duration is determined to be the first duration;

[0055] If the query duration>determination duration×(1+3%), the query duration is determined to be the second duration.

[0056] As a low-code development intelligent auxiliary method based on machine learning described in the present invention, wherein: the data adjustment strategy is specifically:

[0057] Get the initial integration interface;

[0058] Get all target sub-businesses;

[0059] Extracting a sub-location sequence set corresponding to each target sub-business;

[0060] Execute steps S1 to S13 for each sub-position sequence set respectively;

[0061] S1. Identify the first sub-element in the sub-position sequence set as the previous sub-element;

[0062] S2, define the next element in the sub-position sequence set that is adjacent to the previous sub-element as the next sub-element;

[0063] S3, obtaining the time point at which the target user selects the previous sub-element, and setting it as the previous time point;

[0064] S4, obtaining the cursor coordinates of the target user and setting them as the target coordinates;

[0065] S5. Set unit duration;

[0066] S6. Collect the target coordinates once every unit time until the target user selects the rear sub-element;

[0067] S7, obtaining the time point at which the target user selects the latter sub-element, and setting it as the latter time point;

[0068] S8, the previous time point is used as the starting time point, and the later time point is used as the ending time point to form an analysis period;

[0069] S9, obtaining the number of each target coordinate in the analysis period, and setting it as the analysis number;

[0070] S10, extracting target coordinates with an analysis quantity greater than 1 and defining them as determination coordinates;

[0071] S11, obtaining the coordinates corresponding to the rear sub-element in the preliminary integration interface, and setting them as reference coordinates;

[0072] S12, if the number of determined coordinates = 1, and the determined coordinates = reference coordinates, the latter sub-element is identified as the former sub-element, and steps S2-S13 are repeated until all elements in the sub-position sequence set have their corresponding coordinates determined;

[0073] S13, if the number of determination coordinates is greater than 1, extract all determination coordinates that are different from the reference coordinates and define them as analysis coordinates;

[0074] Obtain the time point corresponding to each analysis coordinate and set it as the analysis time point;

[0075] Calculate the analysis time corresponding to each analysis coordinate, analysis time = analysis time point - previous time point;

[0076] The analysis coordinate corresponding to the analysis duration with the smallest value is extracted and identified as the adjustment coordinate;

[0077] Change the reference coordinates corresponding to the rear sub-element to the adjusted coordinates;

[0078] The latter sub-element is identified as the former sub-element, and steps S2-S13 are repeatedly executed until the corresponding coordinates of all elements in the sub-position sequence set are determined.

[0079] The present invention also discloses a system for executing a low-code development intelligent auxiliary method based on machine learning, which includes:

[0080] Collection module: used to collect target database, target department data and habit data;

[0081] Analysis module: used to analyze the target users' usage habits of the preliminary integration interface and form analysis results;

[0082] Integration module: used to form a preliminary integration interface and adjust the user interface;

[0083] Adjustment module: used to judge and adjust the preliminary integration interface according to the analysis results.

[0084] The present invention has the following beneficial effects:

[0085] 1. The low-code development intelligent auxiliary system and method based on machine learning obtains the functions needed by all departments, integrates them into a low-code database according to different functions, obtains the department where the user is located, automatically extracts the functions corresponding to the department in the low-code database, and automatically extracts the low-code corresponding to each function, and automatically generates the initial usage interface corresponding to the department, thereby improving development efficiency and reducing development time. It can automatically generate usage interfaces according to different departments, prevent people from different departments from entering the department's system, and improve the confidentiality of department data.

[0086] 2. The low-code development intelligent assistance system and method based on machine learning determines whether the target user can find the button corresponding to the required function as soon as possible by obtaining the target user's usage habits of the initial usage interface. If the target user can find the button corresponding to the required function in time, it means that the position of the button corresponding to the function is more in line with the target user's usage habits, and there is no need to adjust the position of the button corresponding to the function. If the target user cannot find the button corresponding to the required function in time, it means that the position of the button corresponding to the function does not conform to the target user's usage habits, and it is necessary to adjust the position of the button corresponding to the function. The usage interface is automatically adjusted according to the differences in the user's usage habits, so that the user can find the function button they need to use more quickly and effectively, reduce the number of accidental clicks on other function keys, be more in line with the user's usage habits, and improve the user's work efficiency.

[0087] 3. This low-code development intelligent assistance system and method based on machine learning obtains the target user's clicking speed for the key flow corresponding to each function, determines the position where the target user subconsciously moves the cursor after clicking a key, and thus determines the adjustment position of the next key. According to the differences in user usage habits, the user interface is automatically adjusted, so that users can find the function keys they need to use more quickly and effectively, reduce the number of accidental clicks on other function keys, be more in line with the user's usage habits, and improve the user's work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 A flow chart of the low-code development intelligent auxiliary method based on machine learning of the present invention;

[0089] Figure 2 A system block diagram for executing the machine learning-based low-code development intelligent assistance method for the present invention. DETAILED DESCRIPTION

[0090] 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 only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0091] Embodiment 1: A low-code development intelligent assistance method based on machine learning, see Figure 1 ,include:

[0092] Get the target database;

[0093] Get target department data;

[0094] Based on the target department data, a preliminary integration interface is formed through preliminary integration strategies;

[0095] Obtain target users’ habit data on the preliminary integration interface;

[0096] The habit data includes first habit data and second habit data;

[0097] If the target user's usage habits for the preliminary integration interface are the first habit data, no adjustment will be made;

[0098] If the target user's usage habit for the preliminary integration interface is the second habit data, the data adjustment strategy is executed to form adjustment data;

[0099] An adjustment usage interface is formed based on the adjustment data and the preliminary integration interface.

[0100] Through the above method, a low-code database is formed according to the usage functions of all departments. According to different departments, the corresponding low-code is automatically extracted and the initial usage interface is formed. According to the differences in user usage habits, the usage interface is automatically adjusted, so that users can find the function buttons they need to use more quickly and effectively, reduce the number of accidental clicks on other function keys, be more in line with the user's usage habits, improve the user's work efficiency, prevent people from different departments from entering the department's system, and improve the confidentiality of department data.

[0101] Embodiment 2: This embodiment is an improvement made on the basis of embodiment 1. In this low-code development intelligent auxiliary method based on machine learning, the acquisition of the target database is specifically:

[0102] Obtain all departments of the target enterprise to form a department set;

[0103] Get all the functions used by each department in the department set, and form a function set for each department;

[0104] According to the department set and the function set corresponding to each department, integrate to form a department function set;

[0105] Obtain the low code corresponding to each element in the department function set respectively to form a target database.

[0106] This embodiment also provides that the acquisition of target department data is specifically:

[0107] Get the target department;

[0108] Extract the function set corresponding to the target department in the target database and define it as the target function set;

[0109] Extract the low code corresponding to the target function set and target department and set it as the target low code;

[0110] Get all target low codes to form a target low code collection.

[0111] This embodiment also provides that the preliminary integration strategy is specifically:

[0112] Get the target low-code collection;

[0113] Setting a default interface layout, where the default interface layout is the interface layout set by the developer;

[0114] Based on the default interface layout and the target low-code set, a new default interface layout is integrated and defined as the preliminary integration interface.

[0115] Embodiment 3: This embodiment is an improvement made on the basis of Embodiment 2. In this embodiment, the acquisition of the habit data of the target user for the preliminary integration interface is specifically as follows:

[0116] Obtain the target user's assigned business and set it as the target business;

[0117] Extract all elements corresponding to the target business in the target function set to form a business function set;

[0118] Implement habit analysis strategies;

[0119] The element corresponding to the first query duration in the business function set is defined as the first business function, that is, the target user has a short search time for the key that needs to be clicked for the business, that is, the target user can quickly find the key corresponding to the function, and there is no need to adjust the position of the key corresponding to the function;

[0120] The element corresponding to the second query duration in the business function set is defined as the second business function, that is, the target user searches for the key that needs to be clicked for the business. That is, the target user cannot quickly find the key corresponding to the function, and the key position corresponding to the function needs to be adjusted at this time;

[0121] Get the total number of elements in the business function set, which is defined as the total number of functions;

[0122] Obtain the quantity of the first business function, which is defined as the first function quantity;

[0123] Obtain the number of the second business function and set it as the second function number;

[0124] If the first function number ÷ the total number of functions ≥ (the second function number ÷ the total number of functions) × (1 + 50%), it is determined to be the first habit data;

[0125] If the first function number ÷ the total number of functions < (the second function number ÷ the total number of functions) × (1 + 50%), it is determined to be the second habit data.

[0126] The habit analysis strategy is specifically as follows:

[0127] Extract the key position of each element in the business function set on the preliminary integration interface and set it as the target position;

[0128] Obtain all sub-businesses under the target business and define them as target sub-businesses;

[0129] Obtain all target positions corresponding to each target sub-business in the business function set and define them as target sub-positions;

[0130] Obtaining the order of each target sub-location corresponding to each target sub-business to form a sub-location order set;

[0131] Extract the first element and the last element in the sub-position sequence set, and define them as the sub-first element and the sub-last element respectively;

[0132] Execute the duration analysis strategy for each target sub-business separately.

[0133] The duration analysis strategy is specifically as follows:

[0134] Get the time point when the target user selects the sub-first element, and set it as the sub-first time point;

[0135] Get the time point when the target user selects the sub-end element, and set it as the sub-end time point;

[0136] Calculate the query duration, query duration = sub-end time point - sub-start time point;

[0137] Obtain all personnel corresponding to the target business in the target department and set them as reference personnel;

[0138] Obtain the query duration of each reference person and set it as the reference duration;

[0139] Calculate the judgment time, judgment time = sum of all reference times ÷ number of reference personnel;

[0140] If the query duration ≤ determination duration × (1 + 3%), the query duration is determined to be the first duration;

[0141] If the query duration>determination duration×(1+3%), the query duration is determined to be the second duration.

[0142] Embodiment 4: This embodiment is an improvement made on the basis of Embodiment 3. In this embodiment, the data adjustment strategy is specifically as follows:

[0143] Get the initial integration interface;

[0144] Get all target sub-businesses;

[0145] Extracting a sub-location sequence set corresponding to each target sub-business;

[0146] Execute steps S1 to S13 for each sub-position sequence set respectively;

[0147] S1. Identify the first sub-element in the sub-position sequence set as the previous sub-element;

[0148] S2, define the next element in the sub-position sequence set that is adjacent to the previous sub-element as the next sub-element;

[0149] S3, obtaining the time point at which the target user selects the previous sub-element, and setting it as the previous time point;

[0150] S4, obtaining the cursor coordinates of the target user and setting them as the target coordinates;

[0151] S5. Set the unit time length, which is 0.1S;

[0152] S6. Collect the target coordinates once every unit time until the target user selects the rear sub-element;

[0153] S7, obtaining the time point at which the target user selects the latter sub-element, and setting it as the latter time point;

[0154] S8, the previous time point is used as the starting time point, and the later time point is used as the ending time point to form an analysis period;

[0155] S9, obtaining the number of each target coordinate in the analysis period, and setting it as the analysis number;

[0156] S10, extracting target coordinates with an analysis quantity greater than 1 and defining them as determination coordinates;

[0157] S11, obtaining the coordinates corresponding to the rear sub-element in the preliminary integration interface, and setting them as reference coordinates;

[0158] S12, if the number of determined coordinates = 1, and the determined coordinates = reference coordinates, the latter sub-element is identified as the former sub-element, and steps S2-S13 are repeated until all elements in the sub-position sequence set have their corresponding coordinates determined;

[0159] S13, if the number of determination coordinates is greater than 1, extract all determination coordinates that are different from the reference coordinates and define them as analysis coordinates;

[0160] Obtain the time point corresponding to each analysis coordinate and set it as the analysis time point;

[0161] Calculate the analysis time corresponding to each analysis coordinate, analysis time = analysis time point - previous time point;

[0162] The analysis coordinate corresponding to the analysis duration with the smallest value is extracted and identified as the adjustment coordinate;

[0163] Change the reference coordinates corresponding to the rear sub-element to the adjusted coordinates;

[0164] The latter sub-element is identified as the former sub-element, and steps S2-S13 are repeatedly executed until the corresponding coordinates of all elements in the sub-position sequence set are determined.

[0165] Embodiment 5: This embodiment also discloses a system for executing a low-code development intelligent assistance method based on machine learning, see Figure 2 ,include:

[0166] Collection module: used to collect target database, target department data and habit data;

[0167] Analysis module: used to analyze the target users' usage habits of the preliminary integration interface and form analysis results;

[0168] Integration module: used to form a preliminary integration interface and adjust the user interface;

[0169] Adjustment module: used to judge and adjust the preliminary integration interface according to the analysis results.

[0170] In this embodiment, a low-code database is formed according to the usage functions of all departments. According to different departments, the corresponding low-code is automatically extracted and the initial usage interface is formed. According to the differences in user usage habits, the usage interface is automatically adjusted, so that users can find the function buttons they need to use more quickly and effectively, reduce the number of accidental clicks on other function keys, be more in line with the user's usage habits, improve the user's work efficiency, prevent people from different departments from entering the department's system, and improve the confidentiality of department data.

[0171] The embodiments of the present disclosure also provide a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the low-code development intelligent assistance system and method based on machine learning in the aforementioned method embodiments.

[0172] It should be noted that more specific examples of computer-readable storage media may include portable computer disks, hard disks, erasable programmable read-only memories (EPROMs or flash memories), optical storage devices, magnetic storage devices, or any suitable combination thereof. In the present disclosure, computer-readable storage media may be any tangible medium that contains or stores a program, which may be used by or in conjunction with an instruction execution system, apparatus, or device, and the computer-readable medium may be contained in the electronic device; or it may exist independently without being assembled into the electronic device.

[0173] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device can implement the solution provided by the method embodiment.

[0174] Computer program code for performing operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, and may be executed entirely on a user's computer, partially on a user's computer, as a stand-alone software package, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0175] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0176] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A low-code development intelligent assistance method based on machine learning, characterized by: include: Get the target database; Get target department data; Based on the target department data, a preliminary integration interface is formed through preliminary integration strategies; Obtain target users’ habit data on the preliminary integration interface; The habit data includes first habit data and second habit data; If the target user's usage habits for the preliminary integration interface are the first habit data, no adjustment will be made; If the target user's usage habit for the preliminary integration interface is the second habit data, the data adjustment strategy is executed to form adjustment data; Based on the adjustment data and the preliminary integration interface, an adjustment user interface is formed; The obtaining of the target database is specifically as follows: Obtain all departments of the target enterprise to form a department set; Get all the functions used by each department in the department set, and form a function set for each department; According to the department set and the function set corresponding to each department, integrate to form a department function set; Obtain the low code corresponding to each element in the department function set respectively to form a target database.

2. According to claim 1, a low-code development intelligent auxiliary method based on machine learning is characterized in that: The acquisition of target department data is specifically as follows: Get the target department; Extract the function set corresponding to the target department in the target database and define it as the target function set; Extract the low code corresponding to the target function set and target department and set it as the target low code; Get all target low codes to form a target low code collection.

3. According to the low-code development intelligent auxiliary method based on machine learning according to claim 1, it is characterized by: The preliminary integration strategy is specifically: Get the target low-code collection; Set the default interface layout; Based on the default interface layout and the target low-code set, a new default interface layout is integrated and defined as the preliminary integration interface.

4. According to the low-code development intelligent auxiliary method based on machine learning according to claim 1, it is characterized by: The step of obtaining the target user's habit data on the preliminary integration interface is as follows: Obtain the target user's assigned business and set it as the target business; Extract all elements corresponding to the target business in the target function set to form a business function set; Implement habit analysis strategies; The element corresponding to the first duration in the business function set is defined as the first business function; The element corresponding to the query duration of the second duration in the business function set is defined as the second business function; Get the total number of elements in the business function set, which is defined as the total number of functions; Obtain the quantity of the first business function, which is defined as the first function quantity; Obtain the number of the second business function and set it as the second function number; If the first function number ÷ the total number of functions ≥ (the second function number ÷ the total number of functions) × (1 + 50%), it is determined to be the first habit data; If the first function number ÷ the total number of functions < (the second function number ÷ the total number of functions) × (1 + 50%), it is determined to be the second habit data.

5. According to claim 4, a low-code development intelligent auxiliary method based on machine learning is characterized in that: The habit analysis strategy is specifically as follows: Extract the key position of each element in the business function set on the preliminary integration interface and set it as the target position; Obtain all sub-businesses under the target business and define them as target sub-businesses; Obtain all target positions corresponding to each target sub-business in the business function set and define them as target sub-positions; Obtaining the order of each target sub-location corresponding to each target sub-business to form a sub-location order set; Extract the first element and the last element in the sub-position sequence set, and define them as the sub-first element and the sub-last element respectively; Execute the duration analysis strategy for each target sub-business separately.

6. According to claim 5, a low-code development intelligent auxiliary method based on machine learning is characterized in that: The duration analysis strategy is specifically as follows: Get the time point when the target user selects the sub-first element, and set it as the sub-first time point; Get the time point when the target user selects the sub-end element, and set it as the sub-end time point; Calculate the query duration, query duration = sub-end time point - sub-start time point; Obtain all personnel corresponding to the target business in the target department and set them as reference personnel; Obtain the query duration of each reference person and set it as the reference duration; Calculate the judgment time, judgment time = sum of all reference times ÷ number of reference personnel; If the query duration ≤ determination duration × (1 + 3%), the query duration is determined to be the first duration; If the query duration>determination duration×(1+3%), the query duration is determined to be the second duration.

7. According to claim 1, a low-code development intelligent auxiliary method based on machine learning is characterized in that: The data adjustment strategy is specifically as follows: Get the initial integration interface; Get all target sub-businesses; Extracting a sub-location sequence set corresponding to each target sub-business; Execute steps S1 to S13 for each sub-position sequence set respectively; S1. Identify the first sub-element in the sub-position sequence set as the previous sub-element; S2, define the next element in the sub-position sequence set that is adjacent to the previous sub-element as the next sub-element; S3, obtaining the time point at which the target user selects the previous sub-element, and setting it as the previous time point; S4, obtaining the cursor coordinates of the target user and setting them as the target coordinates; S5. Set unit duration; S6. Collect the target coordinates once every unit time until the target user selects the rear sub-element; S7, obtaining the time point at which the target user selects the latter sub-element, and setting it as the latter time point; S8, the previous time point is used as the starting time point, and the later time point is used as the ending time point to form an analysis period; S9, obtaining the number of each target coordinate within the analysis period, and setting it as the analysis number; S10, extracting target coordinates with an analysis quantity greater than 1 and defining them as determination coordinates; S11, obtaining the coordinates corresponding to the rear sub-element in the preliminary integration interface, and setting them as reference coordinates; S12, if the number of determined coordinates = 1, and the determined coordinates = reference coordinates, the latter sub-element is identified as the former sub-element, and steps S2-S13 are repeated until all elements in the sub-position sequence set have their corresponding coordinates determined; S13, if the number of determination coordinates is greater than 1, extract all determination coordinates that are different from the reference coordinates and define them as analysis coordinates; Obtain the time point corresponding to each analysis coordinate and set it as the analysis time point; Calculate the analysis time corresponding to each analysis coordinate, analysis time = analysis time point - previous time point; The analysis coordinate corresponding to the analysis duration with the smallest value is extracted and identified as the adjustment coordinate; Change the reference coordinates corresponding to the rear sub-element to the adjusted coordinates; The latter sub-element is identified as the former sub-element, and steps S2-S13 are repeatedly executed until the corresponding coordinates of all elements in the sub-position sequence set are determined.

8. A system for executing the low-code development intelligent assistance method based on machine learning as described in claim 1, characterized in that: include: Collection module: used to collect target database, target department data and habit data; Analysis module: used to analyze the target users' usage habits of the preliminary integration interface and form analysis results; Integration module: used to form a preliminary integration interface and adjust the user interface; Adjustment module: used to judge and adjust the preliminary integration interface according to the analysis results.

Images