Intelligent interactive thematic map generation system and method based on large language model

Abstract

The invention discloses an intelligent interactive thematic map generation system and method based on a large language model. Comprising a large language model module, a thematic map template library, an intelligent data source management module, an intelligent cartographic engine, an interactive optimization module, a user interaction configuration module, a configuration library, an external service interface module, a model data quality verification module, a thematic map personalized correction module and the like. A user interactively selects a template, specifies a data source and describes drawing requirements through a natural language, LLM drives an intelligent drawing engine to generate a thematic map, fine adjustment and real-time optimization of the user are supported, and user-friendly, efficient and flexible thematic map making is achieved. Meanwhile, the system has the functions of data quality verification, personalized configuration, interactive history learning and the like, a powerful tool is provided for non-professional users to quickly customize and finely adjust thematic maps, and the intelligent level and adaptability of thematic map making are greatly improved.

Description

Technical Field

[0001] The present invention relates to the fields of geographic information system (GIS), artificial intelligence, and big data visualization technology, and in particular to an intelligent interactive thematic map generation system and method using a large language model (LLM) as a core engine. Background Art

[0002] At present, the existing thematic map production technologies generally have problems such as complex user operations, high professional knowledge requirements, low efficiency in data source selection and processing, and inflexible mapping rule settings.

[0003] Especially when faced with multiple and heterogeneous geographic data sources and users' diverse and dynamic mapping needs, traditional technologies are difficult to respond efficiently and accurately.

[0004] Therefore, it is urgent to develop a new system and method that can deeply integrate artificial intelligence and natural language processing technology, lower the user operation threshold, enhance the interactivity and intelligent response capability of the mapping process, and adapt to the needs of rapid customization and dynamic optimization of thematic maps in a big data environment. Summary of the invention

[0005] The purpose of the present invention is to solve the problems that the existing thematic map production technology in the prior art generally has the disadvantages of complex user operation, high professional knowledge requirements, low efficiency in data source selection and processing, and insufficient flexibility in mapping rule setting, and proposes an intelligent interactive thematic map generation system based on a large language model, specifically, as follows:

[0006] An intelligent interactive thematic map generation system based on a large language model includes the following components:

[0007] a. Large language model module, used to parse natural language instructions input by users, understand user intentions, intelligently generate or adjust mapping rules, and drive mapping software to perform corresponding operations;

[0008] b. Thematic map template library, with various types of thematic map templates preset for users to choose from;

[0009] c. Data source management module, which supports users to specify data sources through dialogue or direct selection, and seamlessly connects with the background's diverse and heterogeneous data resources to achieve intelligent recommendation, automatic matching, one-click import and real-time update of data sources;

[0010] d. The mapping engine, in collaboration with the large language model, generates high-quality thematic maps using automatic mapping algorithms and adaptive layout strategy technologies based on the template selected by the user, the specified data source, and detailed drawing requirements;

[0011] e. The interactive optimization module receives users' requests for refined adjustments to the generated thematic maps, and drives the mapping engine to perform real-time optimization through a large language model to ensure the dynamic adaptability and continuous availability of the thematic maps.

[0012] Furthermore, the data source management module supports the user to specify the data source through a dialogue, specifically including:

[0013] a. The user describes the name, content or update cycle information of the data source in natural language;

[0014] b. The large language model automatically recommends and matches data sources based on background data resources;

[0015] c. After the user confirms the recommended data source, the system imports it with one click.

[0016] Furthermore, the data source management module supports the user to specify the data source by clicking, specifically including:

[0017] a. The user directly selects a specified file or service link in the file system or online resource library as the data source;

[0018] b. The system performs intelligent identification and verification before importing.

[0019] Furthermore, the mapping engine supports users to make fine adjustments to the generated thematic maps, including:

[0020] a. The user proposes a refined adjustment requirement, which includes: changing the legend font, adjusting the color scheme, adding annotations, modifying thematic properties, or updating the data source;

[0021] b. The large language model receives user needs and drives the mapping engine to perform real-time optimization;

[0022] c. The system intelligently handles data updates, changes in thematic associations, or style adjustments to ensure the dynamic adaptability and continuous availability of thematic maps.

[0023] Furthermore, the system also includes the following components:

[0024] a. User interaction configuration module, used to interact with users on thematic map style requirements, obtain and store system general mapping parameter configuration and personalized parameter configuration;

[0025] b. Configuration library, used to store system general mapping parameter configuration, system personalized mapping parameter configuration and generated thematic maps;

[0026] c. External service interface module, used to receive thematic map generation requests from external systems and obtain model data and thematic map types;

[0027] d. A model data quality verification module, which is used to perform quality inspection on the acquired model data and provide the model data that meets the quality requirements to the automatic mapping module;

[0028] e. Thematic map personalization correction module is used to configure thematic map system mapping personalization parameters and optimize the thematic map mapping effect when the external system determines that the current thematic map does not meet the requirements, until the external system determines that the current thematic map meets the requirements, store the system mapping personalization parameter configuration and the generated thematic map, and output the generated thematic map.

[0029] Furthermore, the thematic map personalized correction module includes:

[0030] a. The unit for judging whether the thematic map meets the requirements is used for applying the thematic map to the external system and judging whether the current thematic map meets the requirements;

[0031] b. Personalized parameter configuration unit, which is used to open the current thematic map through the configuration tool when the external system determines that the current thematic map does not meet the requirements, configure the personalized mapping parameters of the thematic map system, and optimize the thematic map mapping effect;

[0032] c. The generated thematic map output unit is used to store the system mapping personalized parameter configuration and the generated thematic map to the configuration library and output the generated thematic map when the external system determines that the current thematic map meets the requirements.

[0033] Furthermore, the system also includes:

[0034] a. Interaction history database, used to store the interaction history of thematic map style requirements, data source selection, and drawing requirements described by users in natural language;

[0035] b. The learning module is used to learn user preferences and mapping habits based on historical interaction data, and optimize the ability of the large language model to understand and generate mapping rules.

[0036] The present invention also provides a method for generating an intelligent interactive thematic map based on a large language model, comprising the following steps:

[0037] S1: The user selects a suitable thematic map type in the thematic map template library;

[0038] S2: The user specifies the data source through dialogue or direct selection;

[0039] S3: The user describes the drawing requirements in natural language, including: the classification rules of the hierarchical thematic map, the number of point representatives of the point density map, the equipment attributes of the distribution network thematic map, and the analysis indicators of the ecological monitoring report;

[0040] S4: The large language model drives the mapping engine to generate thematic maps, using automatic mapping algorithms and adaptive layout strategy technology to ensure the high quality and high standards of thematic maps;

[0041] S5: Users request fine-tuning of the generated thematic maps. The large language model drives the mapping engine to perform real-time optimization. The system intelligently handles data updates and dynamic scenarios of demand changes until the user is satisfied.

[0042] Furthermore, in the process of generating the thematic map by the mapping engine in step S4, the following steps are also included:

[0043] a. The automatic mapping module automatically generates thematic maps based on the data source, thematic map type, and system general mapping parameter configuration and system personalized mapping parameter configuration specified by the user;

[0044] b. The model data quality verification module performs quality inspection on the acquired model data and provides the model data that meets the quality requirements to the automatic drawing module;

[0045] c. If the model data does not meet the quality requirements, the model data quality verification module generates the model data verification failure results and a list of model data quality issues, and provides feedback to the user;

[0046] d. If the generated thematic map does not meet the user's requirements, the thematic map personalization correction module configures the personalized mapping parameters of the thematic map system and optimizes the thematic map mapping effect until the user's requirements are met.

[0047] Furthermore, the specific steps for fine-tuning in S5 are as follows:

[0048] a. Users describe their adjustment requirements through natural language, such as adding annotations to specific areas, changing the display mode of certain thematic attributes, adjusting the transparency or overlay order of layers;

[0049] b. The large language model understands user needs and generates corresponding mapping rule adjustment instructions;

[0050] c. The mapping engine updates the thematic map in real time according to the adjustment instructions, and also updates the personalized configuration information saved in the system for subsequent reuse or adjustment.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] The intelligent interactive thematic map generation system and method based on a large language model provided by the present invention has the following significant advantages:

[0053] User-friendly: It provides a rich variety of templates to choose from, natural language interactive data sources, and detailed descriptions of drawing requirements, which greatly reduces the user's professional skill requirements and improves the user experience.

[0054] Efficient and intelligent: Integrates a large language model to achieve intelligent recommendation, automatic matching and one-click import of data sources, intelligent generation and dynamic adjustment of mapping rules, and greatly improves data processing and mapping efficiency.

[0055] Flexible adaptation: Supports multiple data source types, thematic map types, and real-time interactive adjustments by users, ensuring that thematic maps can quickly adapt to dynamic scenarios such as data updates and changes in demand.

[0056] High quality standards: Automatic mapping algorithms, adaptive layout strategies, intelligent annotation technologies, etc. are used to ensure the high quality and high standards of thematic maps generated, in line with professional mapping specifications and user aesthetic needs.

[0057] In summary, the present invention provides an innovative, user-friendly, efficient and intelligent thematic map generation solution, which is suitable for multiple industries and application scenarios such as electricity, environmental protection, and remote sensing. It provides a powerful tool for non-professional users to quickly customize and fine-tune thematic maps, and greatly improves the intelligence level and adaptability of thematic map production. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The drawings described herein are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations. In the drawings:

[0059] Figure 1 This is a schematic diagram of the architecture of the intelligent interactive thematic map generation system based on the large language model of the present invention;

[0060] Figure 2 Schematic diagram I of the user interaction interface of the present invention;

[0061] Figure 3 Schematic diagram II of the user interaction interface of the present invention;

[0062] Figure 4 Schematic diagram III of the user interaction interface of the present invention;

[0063] Figure 5 It is a schematic diagram of the intelligent data source management process of the present invention;

[0064] Figure 6 A schematic diagram of the process of generating and adjusting the intelligent mapping rules of the present invention. DETAILED DESCRIPTION

[0065] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0066] Embodiment 1: To achieve the purpose of convenient operation, high efficiency in data source selection and processing, and flexible mapping rule setting, this embodiment provides an intelligent interactive thematic map generation system based on a large language model, which specifically includes the following system architecture parts:

[0067] a. Large Language Model Module (LLM): Integrates a large language model with deep natural language understanding and generation capabilities to parse natural language instructions input by users, understand user intentions, intelligently generate or adjust mapping rules, and drive mapping software to perform corresponding operations;

[0068] b. Thematic map template library: preset various types of thematic map templates, such as independent value thematic map, graded thematic map, point density map, statistical thematic map, distribution network thematic map, ecological monitoring report map, etc., covering multiple industry application fields such as electricity, environmental protection, remote sensing, etc., for users to choose according to their needs;

[0069] c. Data source management module: supports users to specify data sources through dialogue or direct selection, such as Shapefile, MDB file, Geodatabase, remote sensing image, online service interface, etc.

[0070] It also seamlessly connects with the diverse and heterogeneous data resources in the background (such as GeoDatabase, thematic database, online API, etc.) to achieve intelligent recommendation, automatic matching, one-click import and real-time update of data sources.

[0071] d. Mapping engine: It works closely with the large language model to automatically generate high-quality and high-standard thematic maps based on the template selected by the user, the specified data source and detailed drawing requirements, combined with automatic mapping algorithms, adaptive layout strategies, intelligent annotation technology, etc. At the same time, it supports users to make real-time interactive adjustments to the generated thematic maps, such as personalized corrections, layer management, style switching, etc.

[0072] e. Interactive optimization module: Receive users' requests for refined adjustments to the generated thematic maps (such as changing legend fonts, adjusting color schemes, adding annotations, modifying thematic properties, updating data sources, etc.), and drive the mapping engine to perform real-time optimization through a large language model to ensure the dynamic adaptability and continuous availability of thematic maps.

[0073] In addition, during the optimization process, the system can intelligently identify and handle issues such as data conflicts, quality verification, and thematic correlation analysis to ensure the accuracy and consistency of thematic maps.

[0074] It should also be noted that: the data source management module has two modes, as follows:

[0075] (1) Support users to specify data sources through dialogue, including:

[0076] a. The user describes the name, content or update cycle information of the data source in natural language;

[0077] b. The large language model automatically recommends and matches data sources based on background data resources;

[0078] c. After the user confirms the recommended data source, the system imports it with one click.

[0079] (2) Support users to specify data sources by clicking, including:

[0080] a. The user selects a specified file or service link in the file system or online resource library as the data source

[0081] b. The system performs intelligent identification and verification before importing.

[0082] It should also be noted that the mapping engine supports users to make fine adjustments to the generated thematic maps in the following specific ways:

[0083] a. The user proposes a refined adjustment requirement, which includes: changing the legend font, adjusting the color scheme, adding annotations, modifying thematic properties, or updating the data source;

[0084] b. The large language model receives user needs and drives the mapping engine to perform real-time optimization;

[0085] c. The system intelligently handles data updates, changes in thematic associations, or style adjustments to ensure the dynamic adaptability and continuous availability of thematic maps.

[0086] Embodiment 2: In addition to the components in Embodiment 1, this system also includes the following components:

[0087] a. User interaction configuration module, used to interact with users on thematic map style requirements, obtain and store system general mapping parameter configuration and personalized parameter configuration;

[0088] b. Configuration library, used to store system general mapping parameter configuration, system personalized mapping parameter configuration and generated thematic maps;

[0089] c. External service interface module, used to receive thematic map generation requests from external systems and obtain model data and thematic map types;

[0090] d. A model data quality verification module, which is used to perform quality inspection on the acquired model data and provide the model data that meets the quality requirements to the automatic mapping module;

[0091] e. Thematic map personalized correction module, which is used to configure the personalized parameters of thematic map system when the external system determines that the current thematic map does not meet the requirements, optimize the thematic map effect, until the external system determines that the current thematic map meets the requirements, store the system personalized parameter configuration and the generated thematic map, and output the generated thematic map;

[0092] f. Interaction history database, used to store the interaction history of thematic map style requirements, data source selection and drawing requirements described by users in natural language;

[0093] g. The learning module is used to learn user preferences and mapping habits based on historical interaction data, and optimize the ability of the large language model to understand and generate mapping rules.

[0094] It should be noted that the personalized correction module of thematic maps includes:

[0095] a. The unit for judging whether the thematic map meets the requirements is used for applying the thematic map to the external system and judging whether the current thematic map meets the requirements;

[0096] b. Personalized parameter configuration unit, which is used to open the current thematic map through the configuration tool when the external system determines that the current thematic map does not meet the requirements, configure the personalized mapping parameters of the thematic map system, and optimize the thematic map mapping effect;

[0097] c. The generated thematic map output unit is used to store the system mapping personalized parameter configuration and the generated thematic map to the configuration library and output the generated thematic map when the external system determines that the current thematic map meets the requirements.

[0098] Embodiment 3:

[0099] See also Figure 1-Figure 6 This embodiment specifically provides a method for generating an intelligent interactive thematic map based on a large language model, comprising the following steps:

[0100] S1: The user selects a suitable thematic map type in the thematic map template library;

[0101] S2: The user specifies the data source through dialogue or direct selection;

[0102] S3: The user describes the drawing requirements in natural language, including: the classification rules of the hierarchical thematic map, the number of point representatives of the point density map, the equipment attributes of the distribution network thematic map, and the analysis indicators of the ecological monitoring report;

[0103] S4: The large language model drives the mapping engine to generate thematic maps, using automatic mapping algorithms and adaptive layout strategy technology to ensure the high quality and high standards of thematic maps;

[0104] S5: Users request fine-tuning of the generated thematic maps. The large language model drives the mapping engine to perform real-time optimization. The system intelligently handles data updates and dynamic scenarios of demand changes until the user is satisfied.

[0105] Furthermore, in the process of generating the thematic map by the mapping engine in step S4, the following steps are also included:

[0106] a. The automatic mapping module automatically generates thematic maps based on the data source, thematic map type, and system general mapping parameter configuration and system personalized mapping parameter configuration specified by the user;

[0107] b. The model data quality verification module performs quality inspection on the acquired model data and provides the model data that meets the quality requirements to the automatic drawing module;

[0108] c. If the model data does not meet the quality requirements, the model data quality verification module generates the model data verification failure results and a list of model data quality issues, and provides feedback to the user;

[0109] d. If the generated thematic map does not meet the user's requirements, the thematic map personalization correction module configures the personalized mapping parameters of the thematic map system and optimizes the thematic map mapping effect until the user's requirements are met.

[0110] Furthermore, the specific steps for fine-tuning in S5 are as follows:

[0111] a. Users describe their adjustment requirements through natural language, such as adding annotations to specific areas, changing the display mode of certain thematic attributes, adjusting the transparency or overlay order of layers;

[0112] b. The large language model understands user needs and generates corresponding mapping rule adjustment instructions;

[0113] c. The mapping engine updates the thematic map in real time according to the adjustment instructions, and also updates the personalized configuration information saved in the system for subsequent reuse or adjustment.

[0114] Specifically, the above steps are described in detail as follows:

[0115] 1. Template selection:

[0116] Users can select the appropriate thematic map type from the rich thematic map template library provided by the system. The template library contains a variety of preset thematic map styles, layouts and analysis functions, such as geographic distribution maps, time series maps, heat maps, pie charts, bar charts, line charts, etc., covering geographic information, demographics, economic analysis, environmental monitoring, traffic planning and other fields. Users can browse and select the most suitable template intuitively through the system interface according to actual research needs or report topics. For example, if users need to analyze the population density distribution of various regions in a city, they can select the "Population Distribution Heat Map" template.

[0117] 2. Intelligent data source specification:

[0118] The data source can be specified in two different ways.

[0119] (1)Dialogue mode specification:

[0120] The user describes the name, content, update cycle and other information of the data source through natural language, and LLM automatically recommends and matches the data source based on the background data resources. After the user confirms, the data source can be imported with one click. The specific operations are as follows:

[0121] Describe the data source: Users use natural language to describe the name, content, update cycle and other key information of the required data source. For example: "I want to use the 2022 provincial capital city air pollution index data. The data source is the national city air quality report every month, which is updated once a month."

[0122] LLM recommendation matching: The system's built-in LLM searches multiple data warehouses connected to the backend (such as public data platforms, statistical agency interfaces, internal databases, etc.) based on user descriptions, identifies and recommends qualified data sources. LLM uses natural language understanding technology to parse user descriptions, extract keywords (such as "provincial capital city air pollution index", "2022", "air quality report", "monthly update"), and accurately matches these keywords in the backend data resources.

[0123] User confirms import: The system displays a list of recommended data sources to the user, including detailed information such as data source name, description, update date, data range, etc. After the user verifies that everything is correct, click "Confirm Import", and the system automatically completes operations such as data source authorization verification, data capture, cleaning and conversion, and loads the required data into the thematic map project.

[0124] (2) Specify by clicking:

[0125] The user directly selects the specified file, service link, etc. in the file system or online resource library as the data source, and the system performs intelligent identification and verification before importing. The specific operations are as follows:

[0126] Select data file / link: Users can upload locally stored data files in CSV, Excel, JSON, GeoJSON, etc. formats by dragging or selecting files in the "Upload Data" area of ​​the system interface. Alternatively, users can paste API interface links of online data services, such as RESTfulAPI, WFS services, etc.

[0127] Intelligent identification and verification: The system automatically detects the format and content structure of the uploaded file, or parses the data format returned by the API interface. For geospatial data, the system also checks information such as coordinate system and geometry type. For unstructured data, the system uses natural language processing technology to parse text content and extract key fields. The system will perform preliminary verification of data integrity and validity, such as checking for missing values, outliers, duplicate values, etc.

[0128] Import and conversion: After verification, the system imports the data file or API interface data into the thematic map project, and performs necessary data conversion and format adaptation according to the template requirements, such as field renaming, data type conversion, geocoding, etc.

[0129] 3. Smart drawing requirements description:

[0130] Thematic maps usually include different types. The specific drawing description is similar to the following according to the type of thematic map.

[0131] (1) Hierarchical thematic map: Users describe the number of levels, the basis for grading (such as attribute value interval, natural index, statistical threshold, etc.) and the interval division rules (equal width, proportional, natural distribution, etc.) in natural language, and LLM generates the corresponding mapping rules. The specific operations are as follows:

[0132] Describe the grading rules: Users use natural language to describe the detailed requirements of the grading thematic map. For example: "Divide the map into five levels according to the GDP value of each province. The first level is the top 10% with the highest GDP, the second level is the second highest 20%, and so on. The color changes from dark green to light green."

[0133] LLM understands and generates rules: LLM understands user descriptions, extracts key information (such as the number of classifications "five levels", the basis for classification "GDP value", the interval division rule "equal proportion", and the color gradient "dark green to light green"), and generates corresponding mapping rules based on this information. The rules include classification threshold calculation formulas, color mapping tables, etc.

[0134] Rule application and visualization: The system calculates and classifies the data according to the generated mapping rules, assigns appropriate levels and colors to each province, and displays them visually on the map.

[0135] (2) Point density map: The user defines the actual number of objects (such as population, GDP value, vegetation coverage, etc.) or logical meaning (such as the number of events, number of monitoring sites, etc.) represented by each point, and LLM generates point density calculation rules based on this. The specific operations are as follows:

[0136] Define the meaning of point density: The user describes the actual object or logical meaning represented by each point. For example: "Each point represents a monitoring site, and the size of the point represents the number of pollution events recorded by the site in the past year."

[0137] LLM generates calculation rules: LLM understands the user description, determines the relationship between the point size and the actual number of objects (such as linear, logarithmic, square root, etc.), and generates point density calculation rules. The rules include point size calculation formulas, maximum and minimum point size limits, etc.

[0138] Calculate and draw point density: The system calculates the data according to the generated calculation rules, generates the appropriate point size for each monitoring site, and draws the point density map on the map.

[0139] (3) Other types of thematic maps: Users describe the customized requirements of specific thematic maps in natural language, such as the equipment attributes of the distribution network wiring diagram, the analysis indicators of the ecological monitoring report, etc., and LLM converts them into corresponding mapping rules. The specific operation is similar to the above. Users use natural language to describe the detailed requirements of the thematic map, LLM understands and generates the corresponding mapping rules, and the system processes data and displays them visually according to the rules.

[0140] 4. Large Language Model (LLM) drives intelligent mapping:

[0141] Based on the template selected by the user, the intelligently matched data source and the generated mapping rules, the LLM-driven mapping engine uses automatic mapping algorithms, adaptive layout strategies and other technologies to generate high-quality thematic maps. At the same time, the system automatically records the user's personalized configuration parameters for subsequent reuse or adjustment. The specific process is as follows:

[0142] (1)Integrated data and rules:

[0143] Template integration: The system first extracts the basic geographic information (such as base map, geographic coordinate system, layer structure, etc.), default style (such as color scheme, symbol library, font setting, etc.), interactive functions (such as zooming, roaming, query, etc.) and analysis models or algorithms of specific thematic map types (such as hot spot analysis, buffer analysis, network analysis, etc.) contained in the template selected by the user. These template information constitute the basic framework and visual specifications of the mapping task;

[0144] Data source fusion:

[0145] The system integrates the intelligently matched data sources to ensure that data from different sources and formats can be superimposed and analyzed in a unified geographic coordinate system. This includes preprocessing steps such as data format conversion, coordinate system conversion, spatial index establishment, and data cleaning (removing invalid values, filling missing values, and processing outliers). For time series data, the system also needs to align and integrate the time dimension;

[0146] Cartographic rule fusion:

[0147] The system merges the various mapping rules (such as grading rules, point density calculation rules, symbolization rules, color mapping rules, etc.) generated by the user through natural language description with the rules preset in the template. In the case of conflicts or inconsistencies, the system gives priority to the rules specified by the user. The system also ensures that all rules fully correspond to the fields, attributes, units and other information of the data source, preparing for the subsequent execution of the mapping engine;

[0148] Mapping task description generation:

[0149] The system encapsulates the integrated template information, data source, and mapping rules to form a complete mapping task description. This description includes all the input data, operation instructions, parameter settings, expected outputs, etc. required for mapping, and is well structured and executable, making it easy for LLM to understand and convert.

[0150] (2) LLM-driven drawing engine:

[0151] Task description analysis:

[0152] LLM receives the mapping task description generated by the system, parses it using natural language understanding technology, and extracts the key elements of the task (such as data source ID, data table name, field name, mapping rule ID, parameter value, etc.). LLM ensures that the semantics of the task description is fully understood and converted into an internal representation.

[0153] Instruction set generation:

[0154] Based on the understanding of the mapping task description, LLM converts it into an instruction set that can be directly executed by the mapping engine. The instruction set includes a series of operation instructions, such as data loading, data filtering, spatial aggregation, spatial analysis, symbolization, layout adjustment, etc. Each instruction contains detailed information such as the operation type, operation object, and operation parameters, and follows the instruction format and syntax specified by the mapping engine.

[0155] Automatic mapping algorithm execution:

[0156] After the mapping engine receives the instruction set generated by LLM, it executes it in sequence. The automatic mapping algorithm mainly includes the following steps:

[0157] Step 1: Data aggregation: Aggregate the original data spatially or by attributes, such as summarizing population by administrative division, calculating average pollution value by distance range, etc. The aggregation results serve as the basis for symbolization and layout in the next step.

[0158] Step 2: Spatial analysis: According to the mapping requirements, the spatial relationship analysis of the data is performed, such as calculating the distance between two points, generating buffers, finding the nearest facilities, building network paths, etc. The results of spatial analysis provide in-depth spatial insights for thematic maps.

[0159] Step 3: Symbolization: According to the mapping rules, the aggregated or analyzed data is converted into visual symbols (such as color, shape, size, texture, etc.) and mapped to the corresponding positions on the map. During the symbolization process, the system considers factors such as the numerical range, classification level, and visual weight of the data to ensure the beauty of the symbols and the effectiveness of information transmission.

[0160] (3) Adaptive layout strategy execution:

[0161] While generating thematic maps, the mapping engine dynamically adjusts the chart layout according to data characteristics (such as data volume, distribution range, spatial complexity, etc.), screen size (such as resolution, aspect ratio, available window size, etc.), and other user preferences (such as zoom level, rotation angle, layer visibility, etc.). Adaptive layout strategies ensure that thematic maps can maintain good readability and aesthetic effects in different display environments, such as automatically adjusting the size and position of legends, optimizing label layout to avoid overlap, and adapting to high-resolution screens or small mobile screens.

[0162] (4) Generate thematic maps:

[0163] The mapping engine generates high-quality thematic maps according to LLM instructions, including geographic basemaps, thematic layers, legends, labels, annotations and other elements. The system automatically records the user's personalized configuration parameters, such as data sources, mapping rules, color schemes, font styles, etc., to facilitate subsequent reuse or adjustment by users.

[0164] In summary, through the above process of integrating data and rules and LLM-driven mapping engine, the system effectively transforms user selection, data source, and mapping rules into visualization results, realizes intelligent and automatic generation of thematic maps, and ensures their good performance in various display environments. During the whole process, users only need to describe their needs in natural language without in-depth understanding of complex GIS technology and mapping processes, which greatly improves the efficiency of thematic map production and user experience.

[0165] 5. Interactive Optimization Iteration:

[0166] When users make detailed adjustments to the generated thematic maps, LLM receives the information and drives the mapping engine to perform real-time optimization. During the optimization process, the system intelligently handles complex situations such as data updates, thematic association changes, and style adjustments to ensure the dynamic adaptability and continuous availability of thematic maps. The specific operations are as follows:

[0167] (1) Users raise refined requirements: After viewing the generated thematic maps, users may raise refined adjustment requirements, such as changing the color scheme, adding new data fields, adjusting the layer order, zooming in on a specific area, etc. Users convey their requirements to the system through natural language or graphical interface operations.

[0168] (2) LLM understands and converts requirements: LLM understands the detailed requirements put forward by users and converts them into operational instructions that can be executed by the mapping engine.

[0169] (3) Real-time optimization and update: After receiving the LLM conversion instruction, the mapping engine immediately optimizes and updates the thematic map. When dealing with complex situations such as data updates, changes in thematic associations, and style adjustments, the system uses intelligent algorithms to automatically identify data changes, maintain the consistency of thematic associations, and adapt to different screen sizes and resolutions to ensure that the thematic map always maintains the best display effect.

[0170] (4) User feedback and loop optimization: Users review the optimized thematic maps. If they are not satisfied, they can continue to propose adjustment requirements. The system will enter the next round of optimization iteration until the user is satisfied. Throughout the process, the system continuously records the user's interaction history and preferences to provide personalized suggestions and support for future mapping tasks.

[0171] In summary, the present invention provides an innovative, user-friendly, efficient and intelligent thematic map generation solution, which integrates a large language model, realizes intelligent recommendation, automatic matching and one-click import of data sources, intelligent generation and dynamic adjustment of mapping rules, and greatly improves data processing and mapping efficiency; integrates a large language model, realizes intelligent recommendation, automatic matching and one-click import of data sources, intelligent generation and dynamic adjustment of mapping rules, and greatly improves data processing and mapping efficiency; uses automatic mapping algorithms, adaptive layout strategies, intelligent annotation technologies, etc. to ensure high quality and high standards of generated thematic maps, which meet professional mapping specifications and user aesthetic needs; is suitable for multiple industries and application scenarios such as electricity, environmental protection, and remote sensing, and provides a powerful tool for non-professional users to quickly customize and fine-tune thematic maps, and greatly improves the intelligence level and adaptability of thematic map production.

[0172] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any equivalent replacement or change made by any technician in this field according to the technical solution and concept of the present invention should be covered within the protection scope of the present invention.

Claims

1. An intelligent interactive thematic map generation system based on a large language model, characterized in that: It includes the following components: a. Large language model module, used to parse natural language instructions input by users, understand user intentions, intelligently generate or adjust mapping rules, and drive mapping software to perform corresponding operations; b. Thematic map template library, with various types of thematic map templates preset for users to choose from; c. Data source management module, which supports users to specify data sources through dialogue or direct selection, and seamlessly connects with multi-source and heterogeneous data resources in the background to achieve intelligent recommendation, automatic matching, one-click import and real-time update of data sources; d. The mapping engine, in collaboration with the large language model, generates high-quality thematic maps using automatic mapping algorithms and adaptive layout strategy technologies based on the template selected by the user, the specified data source, and detailed drawing requirements; e. The interactive optimization module receives users' requests for refined adjustments to the generated thematic maps, and drives the mapping engine to perform real-time optimization through a large language model to ensure the dynamic adaptability and continuous availability of the thematic maps.

2. According to claim 1, the intelligent interactive thematic map generation system based on a large language model is characterized in that: The data source management module supports users to specify data sources through dialogue, specifically including: a. The user describes the name, content or update cycle information of the data source in natural language; b. The large language model automatically recommends and matches data sources based on background data resources; c. After the user confirms the recommended data source, the system imports it with one click.

3. The intelligent interactive thematic map generation system based on a large language model according to claim 2 is characterized in that: The data source management module supports users to specify data sources by clicking, specifically including: a. The user directly selects a specified file or service link in the file system or online resource library as the data source; b. The system performs intelligent identification and verification before importing.

4. The intelligent interactive thematic map generation system based on a large language model according to claim 3 is characterized in that: The mapping engine supports users to make fine adjustments to the generated thematic maps, including: a. The user proposes a refined adjustment requirement, which includes: changing the legend font, adjusting the color scheme, adding annotations, modifying thematic properties, or updating the data source; b. The large language model receives user needs and drives the mapping engine to perform real-time optimization; c. The system intelligently handles data updates, changes in thematic associations, or style adjustments to ensure the dynamic adaptability and continuous availability of thematic maps.

5. The intelligent interactive thematic map generation system based on a large language model according to claim 4, characterized in that: The system also includes the following components: a. User interaction configuration module, used to interact with users on thematic map style requirements, obtain and store system general mapping parameter configuration and personalized parameter configuration; b. Configuration library, used to store system general mapping parameter configuration, system personalized mapping parameter configuration and generated thematic maps; c. External service interface module, used to receive thematic map generation requests from external systems and obtain model data and thematic map types; d. A model data quality verification module, which is used to perform quality inspection on the acquired model data and provide the model data that meets the quality requirements to the automatic mapping module; e. Thematic map personalization correction module is used to configure thematic map system mapping personalization parameters and optimize the thematic map mapping effect when the external system determines that the current thematic map does not meet the requirements, until the external system determines that the current thematic map meets the requirements, store the system mapping personalization parameter configuration and the generated thematic map, and output the generated thematic map.

6. The intelligent interactive thematic map generation system based on a large language model according to claim 5, characterized in that: The thematic map personalized correction module includes: a. The unit for judging whether the thematic map meets the requirements is used for applying the thematic map to the external system and judging whether the current thematic map meets the requirements; b. Personalized parameter configuration unit, which is used to open the current thematic map through the configuration tool when the external system determines that the current thematic map does not meet the requirements, configure the personalized mapping parameters of the thematic map system, and optimize the thematic map mapping effect; c. The generated thematic map output unit is used to store the system mapping personalized parameter configuration and the generated thematic map to the configuration library and output the generated thematic map when the external system determines that the current thematic map meets the requirements.

7. The intelligent interactive thematic map generation system based on a large language model according to claim 6, characterized in that: The system also includes: a. Interaction history database, used to store the interaction history of thematic map style requirements, data source selection, and drawing requirements described by users in natural language; b. The learning module is used to learn user preferences and mapping habits based on historical interaction data, and optimize the ability of the large language model to understand and generate mapping rules.

8. The method for generating an intelligent interactive thematic map based on a large language model according to claim 7, characterized in that: The following steps are involved: S1: The user selects a suitable thematic map type in the thematic map template library; S2: The user specifies the data source through dialogue or direct selection; S3: The user describes the drawing requirements in natural language, including: the classification rules of the hierarchical thematic map, the number of point representatives of the point density map, the equipment attributes of the distribution network thematic map, and the analysis indicators of the ecological monitoring report; S4: The large language model drives the mapping engine to generate thematic maps, using automatic mapping algorithms and adaptive layout strategy technology to ensure the high quality and high standards of thematic maps; S5: Users request fine-tuning of the generated thematic maps. The large language model drives the mapping engine to perform real-time optimization. The system intelligently handles data updates and dynamic scenarios of demand changes until the user is satisfied.

9. The method for generating an intelligent interactive thematic map based on a large language model according to claim 8, characterized in that: In the process of generating the thematic map by the mapping engine in step S4, the following steps are also included: a. The automatic mapping module automatically generates thematic maps based on the data source, thematic map type, and system general mapping parameter configuration and system personalized mapping parameter configuration specified by the user; b. The model data quality verification module performs quality inspection on the acquired model data and provides the model data that meets the quality requirements to the automatic drawing module; c. If the model data does not meet the quality requirements, the model data quality verification module generates the model data verification failure results and a list of model data quality issues, and provides feedback to the user; d. If the generated thematic map does not meet the user's requirements, the thematic map personalization correction module configures the personalized mapping parameters of the thematic map system and optimizes the thematic map mapping effect until the user's requirements are met.

10. The method for generating an intelligent interactive thematic map based on a large language model according to claim 8 or 9, characterized in that: The specific steps for fine-tuning in S5 are as follows: a. Users describe their adjustment requirements through natural language, such as adding annotations to specific areas, changing the display mode of certain thematic attributes, adjusting the transparency or overlay order of layers; b. The large language model understands user needs and generates corresponding mapping rule adjustment instructions; c. The mapping engine updates the thematic map in real time according to the adjustment instructions, and also updates the personalized configuration information saved in the system for subsequent reuse or adjustment.

Images