Low-code development platform based on AI drive

Abstract

The invention relates to the technical field of computer software development, in particular to an AI drive-based low-code development platform, which comprises an AI intelligent form module and an AI recommendation field module which are connected with each other, the AI intelligent form module comprises a demand analysis unit, a table structure automatic generation unit and a table specification verification unit, the AI recommendation field module comprises a data characteristic analysis unit, a field type matching unit and a field attribute optimization unit which are connected in sequence, and the table structure automatic generation unit is connected with the demand analysis unit and the data characteristic analysis unit. The field attribute optimization unit is connected with the table specification verification unit. The form generation accuracy is high, and the response speed is high.

Description

Technical Field

[0001] This invention relates to the field of computer software development technology, and more specifically to an AI-driven low-code development platform. Background Technology

[0002] As enterprises accelerate their digital transformation, low-code development platforms have become an important tool for SMEs to build business systems due to their ability to simplify development processes and shorten development cycles. However, in practical applications, significant bottlenecks still exist in data structure design (such as database table creation), field type selection, and the application of business knowledge. First, non-technical personnel (such as business specialists) often lack database design knowledge, leading to redundant table structures and mismatched field types. Second, when technical personnel manually design table structures, they need to repeatedly consult industry standards, resulting in low efficiency. Simultaneously, employees frequently need to consult professionals for knowledge support during business operations, leading to high communication costs, delayed responses, and hindering business progress.

[0003] In related technologies, while existing low-code development platforms support visual form creation, the table structure requires manual definition of field names, types, and relationships, and cannot automatically generate designs that conform to best practices based on user needs. Field type selection relies on user experience (such as whether to set "phone number" as text or numeric), which can easily lead to data storage anomalies or low query efficiency. Business consultation often relies on pre-set help documents or human customer service, which cannot provide accurate assistance by combining enterprise-specific knowledge bases with real-time business scenarios, and the response speed is slow (average waiting time exceeds 30 minutes), making it difficult to meet the needs of enterprises for efficient development and business decision-making. Summary of the Invention

[0004] The present invention aims to provide an AI-driven low-code development platform with high accuracy and fast response speed in form generation.

[0005] To address the aforementioned technical problems, this invention provides the following technical solution: This invention provides an AI-driven low-code development platform, comprising: an AI intelligent form module and an AI recommended field module interconnected; the AI ​​intelligent form module includes a requirement parsing unit, a table structure automatic generation unit, and a table specification verification unit; the AI ​​recommended field module includes a data characteristic analysis unit, a field type matching unit, and a field attribute optimization unit connected in sequence; the table structure automatic generation unit is connected to the requirement parsing unit and the data characteristic analysis unit respectively; and the field attribute optimization unit is connected to the table specification verification unit.

[0006] By adopting the above technical solution, when a user initiates a form creation request, the user inputs their natural language request via voice or text. The request parsing unit receives the user's natural language request and converts it into structured data. The table structure automatic generation unit outputs a preliminary SQL table model based on the structured data. The data characteristic analysis unit performs further analysis, and the field type matching unit matches and recommends more precise field types based on the analysis results. The field attribute optimization unit recommends and optimizes field attributes. The table specification verification unit verifies the optimized table structure and provides modification suggestions, finally outputting the form. In the above process, the design of the AI ​​intelligent form module can effectively reduce the form design error rate and improve response speed, while the AI ​​recommended field module can effectively improve the accuracy of form generation.

[0007] Optionally, the requirement parsing unit has a built-in text and / or speech recognition interface, which can receive user natural language requirements and convert them into structured data; the table structure automatic generation unit stores table structure generation algorithms based on industry standards and can output a preliminary SQL table creation; the table standard verification unit integrates static detection scripts, which can scan for non-standard items in the table structure.

[0008] Optionally, the data characteristic analysis unit has a built-in data format recognition algorithm for analyzing the length, format, and business purpose of field data; the field type matching unit is used to store field type mapping logic and output the appropriate database field type; the field attribute optimization unit has a built-in attribute configuration rule library for recommending primary key, NOT NULL, index, and default value attributes.

[0009] Optionally, the SQL table prototype includes field names, initial data types, and field relationships.

[0010] Optionally, the non-standard items include redundant fields and type mismatches.

[0011] Optionally, an AI-driven low-code development platform further includes an AI consultation assistant module, which includes a decision support unit, an enterprise knowledge base integration unit, and a real-time consultation interaction unit, all connected to the decision support unit.

[0012] By adopting the above technical solution, when users want to make adjustments, they only need to fine-tune the table structure according to the suggestions of the table specification verification unit or their own business needs.

[0013] Optionally, the enterprise knowledge base integration unit supports importing PDF and / or Word and / or Excel format documents, which can be converted into structured knowledge through OCR recognition and NLP word segmentation; the real-time consultation interaction unit has a built-in text and / or voice interaction interface, which allows users to initiate consultations through input boxes or voice; the decision support unit stores business rule reasoning algorithms, which can output decision suggestions in combination with real-time business data.

[0014] In summary, the present invention has at least the following beneficial technical effects: 1. Technical Results: Through the AI-powered intelligent form module, table structure design time has been reduced from the traditional 2 hours to 5 minutes, and the design error rate (such as mismatched field types or missing primary keys) has been reduced from 35% to below 3%; the AI-powered recommended field module has reduced field storage usage by 40% and improved data query efficiency by 25% (e.g., the response time for "order table" queries has been reduced from 0.8 seconds to 0.6 seconds); the AI-powered consultation assistant module has reduced business consultation response time from 30 minutes to within 10 seconds, with a consultation accuracy rate of over 92%, significantly reducing the cost of manual consultation.

[0015] 2. Economic benefits: Enterprises do not need to hire additional professional database designers, reducing labor costs by 30%; business consultation time is reduced by 80%, and employee work efficiency is increased by 40%.

[0016] 3. Social impact: Lowering the barrier to low-code development, enabling non-technical personnel to independently complete the design of standardized table structures, promoting the digital transformation of SMEs; AI consulting assistants help employees quickly master business knowledge, improve the overall business capabilities of enterprises, and promote the popularization and application of digital technology in various industries. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the structure of an AI-driven low-code development platform.

[0018] Figure labeling: 1. AI Intelligent Form Module; 11. Requirement Analysis Unit; 12. Automatic Table Structure Generation Unit; 13. Table Standard Validation Unit; 2. AI Recommended Field Module; 21. Data Characteristic Analysis Unit; 22. Field Type Matching Unit; 23. Field Attribute Optimization Unit; 3. AI Consultation Assistant Module; 31. Enterprise Knowledge Base Integration Unit; 32. Real-time Consultation Interaction Unit; 33. Decision Support Unit. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0020] This invention provides an AI-driven low-code development platform.

[0021] refer to Figure 1 An AI-driven low-code development platform includes: an interconnected AI smart form module 1 and an AI recommendation field module 2.

[0022] The AI-powered intelligent form module 1 includes a requirement parsing unit 11, an automatic table structure generation unit 12, and a table specification verification unit 13. The requirement parsing unit 11 is connected to the automatic table structure generation unit 12, the automatic table structure generation unit 12 is connected to the input end of the AI ​​recommended field module 2, and the output end of the AI ​​recommended field module 2 is connected to the table specification verification unit 13.

[0023] The requirement parsing unit 11 has a built-in text and / or speech recognition interface (integrating iFlytek Speech SDK), which can receive users' natural language requirements and convert them into structured data; the table structure automatic generation unit 12 stores table structure generation algorithms based on industry standards, which can output a preliminary SQL table creation form containing field names, initial data types, and field relationships; the table standard verification unit 13 integrates a static detection script, which can scan for non-standard items such as redundant fields and type mismatches in the table structure.

[0024] AI-recommended field module 2 includes a data characteristic analysis unit 21, a field type matching unit 22, and a field attribute optimization unit 23, which are connected in sequence. The data characteristic analysis unit 21 is connected to the table structure automatic generation unit 12, and the field attribute optimization unit 23 is connected to the table specification verification unit 13.

[0025] The data characteristic analysis unit 21 has a built-in data format recognition algorithm (supporting regular expression matching and data range detection) that can analyze the length, format, and business purpose of field data; the field type matching unit 22 stores field type mapping logic (such as "amount data containing decimals → DECIMAL type") and can output the appropriate database field type; the field attribute optimization unit 23 has a built-in attribute configuration rule library that can recommend attributes such as primary key, NOT NULL, index, and default value.

[0026] An AI-driven low-code development platform also includes: AI Consulting Assistant Module 3.

[0027] The AI ​​consultation assistant module 3 includes an enterprise knowledge base integration unit 31, a real-time consultation interaction unit 32, and a decision support unit 33. The enterprise knowledge base integration unit 31 and the real-time consultation interaction unit 32 are respectively connected to the decision support unit 33.

[0028] The enterprise knowledge base integration unit 31 supports importing PDF / Word / Excel format documents and can convert them into structured knowledge through OCR recognition and NLP word segmentation; the real-time consultation interaction unit 32 has built-in text and / or voice interaction interfaces, allowing users to initiate consultations through input boxes or voice; the decision support unit 33 stores business rule reasoning algorithms and can output decision suggestions in combination with real-time business data.

[0029] The implementation principle of an AI-driven low-code development platform according to an embodiment of this application is as follows: When a user initiates a form creation request, the user inputs their request in natural language via voice or text, for example: "I need to create a form for customer information registration, including fields such as name, contact information, order amount, and purchase date."

[0030] First, the AI ​​intelligent form module 1 performs preliminary processing. Specifically, the requirement parsing unit 11 receives the user's natural language requirement and uses the built-in text / speech recognition interface (integrated with iFlytek Voice SDK) to convert it into structured data, identifying the required fields for the form as "Name", "Contact Information", "Order Amount", and "Purchase Date". The table structure automatic generation unit 12, based on the structured data output by the requirement parsing unit 11, uses a table structure generation algorithm to output a preliminary SQL table structure containing field names, initial data types (such as VARCHAR, DECIMAL, DATE, etc.), and field relationships.

[0031] Next, the AI-recommended field module 2 optimizes the fields. Specifically, the data characteristic analysis unit 21 further analyzes the fields output by the table structure automatic generation unit 12, such as identifying that the "order amount" field should contain decimals and its length should be within a certain range. The field type matching unit 22, based on the analysis results of the data characteristic analysis unit 21, matches and recommends more precise field types, such as recommending the "order amount" field type as DECIMAL to ensure data accuracy and storage efficiency. The field attribute optimization unit 23 recommends and optimizes field attributes, such as setting the "name" and "contact information" fields as primary key candidates (or setting them as NOT NULL according to business needs), adding an index to the "purchase date" field to improve query efficiency, and setting a default value of 0 for the "order amount" field, etc.

[0032] Then, the form is validated and adjusted through the table standardization verification unit 13 in the AI ​​intelligent form module 1. Specifically, the table standardization verification unit 13 verifies the table structure optimized by the AI ​​recommended field module 2, checks for non-standard items such as redundant fields and type mismatches, and provides modification suggestions.

[0033] When users want to make adjustments, they only need to make minor adjustments to the table structure according to the suggestions of Table Specification Validation Unit 13 or their own business needs, such as adding or deleting certain fields, adjusting field types or attributes, etc.

[0034] During this process, the AI ​​Consulting Assistant module 3 provides decision support. Specifically, if users or administrators have questions about form design, field selection, or business rules, they can import relevant enterprise documents (such as data dictionaries, business specifications, etc.) into the enterprise knowledge base integration unit 31. Users can also consult in real-time through the real-time consultation interaction unit 32, initiating the consultation simply by using an input box or voice, for example, asking, "Should the purchase date field be set as a required field?". The decision support unit 33 combines structured knowledge from the enterprise knowledge base with real-time business data, using business rule reasoning algorithms to provide decision suggestions, such as the answer: "According to our business specifications, the purchase date field should be set as a required field to track order time."

[0035] Finally, based on the decision suggestions provided by the AI ​​consultation assistant module 3, the user then uses the form specification verification unit 13 to perform final confirmation of the form. After confirmation, the user can use the form for customer information registration or other business scenarios to achieve efficient data collection and management.

[0036] Through this interconnected workflow, the three modules collaborate to complete the entire process from user needs analysis, form design optimization, decision support to final form confirmation and use, improving the accuracy and efficiency of form design.

[0037] The above description of the embodiments is only used to provide a detailed introduction to the technical solution of the present invention. However, the description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention, and should not be construed as a limitation of the present invention. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be covered within the protection scope of the present invention.

Claims

1. An AI-driven low-code development platform, characterized in that, include: The AI ​​intelligent form module (1) and the AI ​​recommended field module (2) are interconnected. The AI ​​intelligent form module (1) includes a requirement parsing unit (11), a table structure automatic generation unit (12), and a table specification verification unit (13). The AI ​​recommended field module (2) includes a data characteristic analysis unit (21), a field type matching unit (22), and a field attribute optimization unit (23) connected in sequence. The table structure automatic generation unit (12) is connected to the requirement parsing unit (11) and the data characteristic analysis unit (21) respectively. The field attribute optimization unit (23) is connected to the table specification verification unit (13).

2. The AI-driven low-code development platform as described in claim 1, characterized in that, The requirement parsing unit (11) has a built-in text and / or speech recognition interface, which can receive users' natural language requirements and convert them into structured data; the table structure automatic generation unit (12) stores a table structure generation algorithm based on industry standards, which can output a preliminary SQL table creation; the table standard verification unit (13) integrates a static detection script, which can scan for non-standard items in the table structure.

3. The AI-driven low-code development platform as described in claim 2, characterized in that, The data characteristic analysis unit (21) has a built-in data format recognition algorithm for analyzing the length, format, and business purpose of field data; the field type matching unit (22) is used to store field type mapping logic and output the appropriate database field type; the field attribute optimization unit (23) has a built-in attribute configuration rule library for recommending primary key, non-null, index, and default value attributes.

4. The AI-driven low-code development platform as described in claim 2, characterized in that, The initial SQL table creation prototype includes field names, initial data types, and field relationships.

5. The AI-driven low-code development platform as described in claim 2, characterized in that, The non-standard items include redundant fields and type mismatches.

6. The AI-driven low-code development platform as described in claim 1, characterized in that, Also includes: AI Consulting Assistant Module (3) includes a decision support unit (33), and an enterprise knowledge base integration unit (31) and a real-time consultation interaction unit (32) respectively connected to the decision support unit (33).

7. The AI-driven low-code development platform as described in claim 6, characterized in that, The enterprise knowledge base integration unit (31) supports the import of PDF and / or Word and / or Excel format documents, and can be converted into structured knowledge through OCR recognition and NLP word segmentation; the real-time consultation interaction unit (32) has a built-in text and / or voice interaction interface, which supports users to initiate consultation through input boxes or voice. The decision support unit (33) stores the business rule reasoning algorithm and can output decision suggestions in combination with real-time business data.

Images