A skill task management method and device based on geographic location

Abstract

The application discloses a skill task management method and device based on geographical position, and relates to the technical field of computer data processing. The application obtains structured target data of a user skill task and generates a task graph with sequential dependency relationship according to the structured target data, then associates geographical interest points and execution constraint conditions with task nodes to generate a planning snapshot, subsequently determines a current task node by using a real-time position point stream and a current task progress in an execution stage and performs multi-factor verification, and finally calculates an experience value based on a verification result and updates a ledger, a level and a task progress. The application organically integrates task planning, position authenticity verification and growth incentives into a closed loop process, so that each link in the task execution process is constrained by a planning snapshot in advance and cross-verified by position data afterwards, thereby significantly improving the reliability of skill task management.

Description

Technical Field

[0001] This invention relates to the field of computer data processing technology, and in particular to a method and apparatus for managing skills and tasks based on geographic location. Background Technology

[0002] With the rapid development of mobile internet, IoT, and location services technologies, skills task planning and management are showing broad application prospects in education and training, professional certification, and personal growth. Traditional skills learning often relies on users creating their own plans and manually recording their progress, lacking systematic task decomposition and tracking methods. In recent years, various task management software, online learning platforms, and gamified incentive tools have emerged, attempting to improve users' learning motivation and execution efficiency through mechanisms such as task lists, progress reminders, and point levels. Meanwhile, the maturity of technologies such as geofencing and location check-in has made it technically possible to link task execution with real-world locations, enabling skills task management to evolve towards online-offline integration and behavioral trajectory tracking. The market demand for intelligent solutions that can achieve closed-loop management of the entire task process is growing daily.

[0003] Existing skill task management methods have several shortcomings. First, most methods only provide static task lists or to-do items, requiring users to manually set tasks, select execution locations and times, which is cumbersome and the planning quality depends on the user's experience, resulting in a high barrier to entry and unreasonable planning. Second, existing solutions do not support users freely inputting vague and unstructured growth goals through natural language, and cannot automatically understand user intentions and transform them into executable, structured task sequences. Third, even if some methods introduce location check-in functions, their task planning and execution phases are still disconnected. Task plans are usually manually set by users or generated based on fixed templates, without considering the user's real-time geographical location, resulting in planned offline tasks that may be far from the user's usual location, leading to poor execution feasibility. Finally, existing solutions typically only determine whether the user has arrived at a certain location during task execution, lacking comprehensive verification of multiple dimensions such as dwell time, task execution order, movement speed, and simulated positioning. Furthermore, the planned content can be arbitrarily modified before execution, lacking an immutable and fixed benchmark, making it difficult to prevent cheating and resulting in low reliability of task completion records. Summary of the Invention

[0004] This invention provides a geographical location-based skill task management method and apparatus to solve the technical problems of high user threshold, disconnect between planning and geographical location, and difficulty in verifying the authenticity of task execution in the prior art.

[0005] To address the aforementioned technical problems, this invention provides a location-based skill task management method, comprising: Obtain user input information, parse the user input information based on a preset large language model, and obtain structured target data; Based on the structured target data, generate several task nodes and the sequential dependencies between each task node, and construct a task graph based on the several task nodes and the sequential dependencies between each task node; Associate at least one task node in the task graph with a preset geographic point of interest and preset execution constraints to obtain a target task graph, and generate a planning snapshot based on the target task graph; The system acquires real-time location data uploaded by the user and the current task progress, determines the current task node based on the current task progress and the planning snapshot, and performs multi-factor verification on the matching degree between the real-time location data and the current task node based on the planning snapshot to obtain the verification result. The experience value of the current task node is calculated based on the attributes of the current task node and the verification result, and the user's accumulated experience value ledger and user level are updated based on the experience value. The task progress of the user's skill tasks is updated based on the updated accumulated experience value ledger and user level.

[0006] This invention, based on a pre-defined large language model, understands user natural language input and perceives real-time location, enabling the automatic generation of personalized, executable task plans from fuzzy objectives. This significantly lowers the user's barrier to entry and ensures accurate matching between planned tasks and the user's actual reachable geographic resources. By acquiring structured target data of user skill tasks, abstract user needs can be transformed into machine-processable standardized information, laying a data foundation for subsequent automated planning. Based on this, task nodes with sequential dependencies are generated and a task graph is constructed, creating a logical closed loop between tasks, avoiding arbitrariness in task execution, and improving the systematicity and reliability of planning. Next, geographic points of interest and execution constraints are associated with task nodes, and a planning snapshot is generated, solidifying the spatial and temporal constraints of task execution, providing an irrefutable benchmark for subsequent verification, and enhancing management credibility. Furthermore, by acquiring real-time location point streams and current task progress, and based on... By determining the current task node through a planning snapshot, real-time perception and automated progress of task status are achieved, reducing errors caused by human intervention. This allows for multi-factor verification of the matching degree between real-time location points and the current task node, obtaining verification results. This enables cross-verification of the authenticity of user behavior from multiple dimensions, effectively preventing cheating methods such as simulated location and brief passing through. Finally, experience values ​​are calculated based on task node attributes and verification results, and the accumulated experience value ledger, user level, and task progress are updated. This binds growth incentives to behavioral authenticity; only tasks that pass verification can provide experience feedback, thus improving the overall reliability and authenticity of skill task management.

[0007] Furthermore, The process of generating several task nodes and their sequential dependencies based on the structured target data and user location information, and constructing a task graph based on these dependencies, includes: Skill tags, target levels, and time periods are extracted based on the structured target data; The skill tags are matched with a preset skill knowledge base to obtain the corresponding skill development path; Based on the target level and the time period, the skill development path is divided into several task stages; the geographical constraint range is determined based on the user's location information, and one or more task nodes are generated for each task stage based on the geographical constraint range and a preset task template library; Based on the order of the various task stages, establish sequential dependencies between the task nodes, and connect the task nodes according to the sequential dependencies to generate the task graph.

[0008] This invention extracts skill tags, target levels, and time periods from structured target data to accurately capture the core elements and time constraints of user skill improvement, providing parameter basis for personalized task generation. Next, it matches skill tags with a preset skill knowledge base to obtain corresponding skill growth paths, ensuring that the task generation process follows objective skill progression rules and avoids biases from subjective planning. Furthermore, it divides the skill growth path into several task stages based on the target level and time period, decomposing long-term goals into executable and measurable phased tasks, reducing execution complexity. On this basis, it selectively generates task nodes from available resources matching the user's location, based on the geographical constraints determined by the user's location information, achieving deep integration of task planning and user geolocation. Then, based on a preset task template library, it generates one or more task nodes for each task stage, achieving standardization and reusability of task types, improving the efficiency and consistency of task generation. Finally, it establishes sequential dependencies between task nodes according to the arrangement order of each task stage and connects them to generate a task graph, making the pre- and post-task logic clear and providing a reliable benchmark for subsequent sequence verification, thereby enhancing the structured nature of task planning and the logical rigor of execution.

[0009] Furthermore, associating at least one task node in the task graph with a preset geographic point of interest and preset execution constraints to obtain a target task graph, and generating a planning snapshot based on the target task graph, includes: Obtain the task nodes to be associated in the task graph, and determine the task type and geographical constraint range of the task nodes to be associated. Based on the task type and geographical constraints, several candidate geographic points of interest are matched in a preset geographic point of interest database. The matching degree between the task node to be associated and each candidate geographic point of interest is calculated based on a preset matching degree algorithm, and the candidate geographic point of interest with the highest matching degree is associated with the task node to be associated. The target task graph is obtained by associating the nodes with the preset execution constraints, and a planning snapshot is generated based on the target task graph.

[0010] This invention, by acquiring the task type and geographical constraints of the task nodes to be associated in the task graph, clarifies the functional requirements and spatial boundaries of each task for the execution location, providing filtering conditions for geographical matching. By matching several candidate geographic points of interest (POIs) in a pre-defined database based on the task type and geographical constraints, it rapidly narrows down the candidate set from massive location data, improving matching efficiency. Furthermore, based on a pre-defined matching degree algorithm, it calculates the matching degree between the task node to be associated and each candidate POI, associating the highest degree with it, ensuring the task-location binding result has quantitative optimality and avoiding the actual unexecutability problem caused by arbitrary binding. Finally, by associating the task node with the pre-defined execution constraints and generating a planning snapshot, it solidifies behavioral constraints such as time and dwell time, making the task execution conditions more complete, thereby improving the rationality of geographical association and the complete reliability of the planning snapshot.

[0011] Furthermore, the step of associating the target task node with the preset execution constraints to obtain the target task graph, and generating a planning snapshot based on the target task graph, includes: The task node to be associated is bound based on execution constraints, which include the expected execution time window, minimum dwell time, and fence radius. The task map is updated based on the associated geographic points of interest and execution constraints to generate the target task map; The target task graph is serialized to obtain task sequence data; The hash digest of the task sequence data is calculated based on a preset hash algorithm, and an immutable planning snapshot is generated based on the hash digest.

[0012] This invention binds execution constraints to task nodes, giving each task a clear spatiotemporal execution boundary and providing a quantitative basis for subsequent authenticity verification. Based on the bound geographic points of interest and execution constraints, the task graph is updated to generate the target task graph, integrating all constraint information into a unified task graph structure, maintaining data consistency and integrity. The target task graph is then serialized to obtain task sequence data, converting the complex graph structure into a linear format that is easy to store and transmit, reducing data processing complexity. Finally, a hash digest of the task sequence data is calculated based on a preset hash algorithm, generating an immutable planning snapshot. Utilizing the unidirectional and unique characteristics of the hash digest, the planning snapshot cannot be tampered with or forged once generated, providing an immutable verification benchmark for the entire execution phase and significantly improving the authenticity and tamper-proof capabilities of management.

[0013] Furthermore, the step of obtaining the real-time location point stream uploaded by the user and the current task progress, and determining the current task node based on the current task progress and the planning snapshot, includes: Get the real-time location data stream uploaded by the user and the current task progress; Extract the sequence of task nodes and the sequential dependencies between each task node from the planning snapshot; Based on the current task progress, determine the set of completed task nodes, and based on the sequential dependency relationship and the set of completed task nodes, determine the task nodes to be executed; The first fence range of the geographic points of interest associated with the task node to be executed is obtained based on the planning snapshot; when the location coordinates in the real-time location point stream are detected to enter the first fence range, the task node to be executed is determined as the current task node.

[0014] This invention dynamically captures users' real-time location behavior and task completion status by acquiring real-time location point streams uploaded by users and current task progress, providing real-time data support for task advancement. It extracts task node sequences and sequential dependencies between task nodes from planning snapshots, enabling the system to accurately grasp the logical constraints between tasks and provide a basis for determining the next task to be executed. Next, it determines the set of completed task nodes based on the current task progress and determines the task node to be executed based on sequential dependencies and the completed set, achieving automatic inference of task progress and avoiding errors or cheating that might be introduced by manual selection by the user. Furthermore, it obtains the fence range of geographic points of interest associated with the task node to be executed based on the planning snapshot, and determines the task node to be executed as the current task node when real-time location coordinates are detected entering the fence. This ensures that the activation of the current task is entirely triggered by location rather than subjective user declaration, thereby ensuring the correctness of the task execution order and the authenticity of the location, and improving the automation and reliability of task advancement.

[0015] Furthermore, the multi-factor verification of the matching degree between the real-time location point stream and the current task node based on the planning snapshot to obtain the verification result includes: Extract the geographic points of interest data associated with the current task node and the sequential position of the current task node in the task node sequence from the planning snapshot; Based on the real-time location point stream, the user's location coordinates, timestamps, positioning accuracy, and simulated positioning markers at each time point are obtained; Based on the geographic point of interest data and the sequential location, multi-factor verification is performed on the location coordinates, timestamp, positioning accuracy, and simulated positioning marker to obtain the verification results.

[0016] This invention provides a dual reference from a fixed benchmark for multi-factor verification by extracting the geographic points of interest data associated with the current task node and their sequential position in the task node sequence from the planning snapshot. This enables cross-validation of user behavior, making it difficult for any single-dimensional anomaly to pass the overall verification. This significantly increases the probability of identifying cheating behavior and enhances the ability to ensure the authenticity of task execution results.

[0017] Furthermore, the geographic point of interest data includes target location coordinates, target fence radius, and target minimum dwell time; the multi-factor verification of the location coordinates, timestamp, positioning accuracy, and simulated positioning marker based on the geographic point of interest data and the sequential location, to obtain the verification result, includes: A simulated positioning test is performed based on the simulated positioning marker. If the simulated positioning marker is detected as true, the simulated positioning test result is "fail". The moving speed is calculated based on the distance and time difference between adjacent coordinates in the real-time location point stream. If the moving speed exceeds a preset speed threshold, a speed anomaly marker is generated. Based on the sequential position of the current task node and the sequential dependencies in the planning snapshot, check whether the current task node has completed all the preceding task nodes. If not, generate an order anomaly flag. Based on the location coordinates in the real-time location point stream and the target location coordinates, the continuous dwell time of the user within the radius of the target fence is calculated. If the continuous dwell time is lower than the target minimum dwell time, an insufficient dwell time marker is generated. Based on the expected execution time window of the current task node and the timestamp of the real-time location point stream, if the user's actual execution time does not fall within the expected execution time window, a time deviation marker is generated. Based on the simulated positioning detection results, velocity anomaly markers, sequence anomaly markers, insufficient dwell time markers, and time deviation markers, the verification results are generated, and the verification results include an authenticity score.

[0018] This invention detects locations based on simulated positioning markers; if a marker is genuine, the attempt is immediately rejected, thus cutting off the path to location spoofing using simulated positioning software at the source, providing a hard anti-spoofing effect. By calculating the movement speed based on the distance and time difference between adjacent coordinates and comparing it with a speed threshold, it can identify illogical behaviors such as ultra-high-speed movement, preventing cheating methods that bypass dwell detection through rapid point jumps. Furthermore, it verifies the completion of preceding tasks based on the sequential position of the current task node and the sequential dependencies in the planning snapshot, ensuring that the logical order of task execution cannot be reversed and preventing skipping critical steps to directly complete subsequent tasks. The system detects user behavior by analyzing their continuous dwell time within the target fence radius and comparing it with the minimum dwell time. This distinguishes between brief passes and genuine stays, ensuring that users invest sufficient time at the task location. By comparing the actual execution time with the expected execution time window and generating time deviation markers, it identifies abnormal execution behavior that deliberately avoids the specified time period. By integrating the above multiple detection results and markers to generate a verification result including an authenticity score, the system quantifies multi-factor judgment into a comprehensive score, making the verification result both comprehensive and comparable, thereby comprehensively improving the accuracy and reliability of task authenticity judgment.

[0019] Furthermore, the step of calculating the experience value of the current task node based on its attributes and the verification result, and updating the user's accumulated experience value ledger and user level based on the experience value, includes: When the verification result indicates that the current task node passes the multi-factor verification, the task type, difficulty coefficient, and location level of the geographic point of interest of the current task node are extracted from the planning snapshot. A base experience value is determined based on the task type, and the base experience value is weighted according to the difficulty coefficient to obtain a first experience value. The location coefficient is determined based on the location level, and a second experience value is calculated based on the preset planning bonus coefficient, the location coefficient, and the first experience value. The target experience value is obtained by multiplying the second experience value by the authenticity score; and the user's accumulated experience value ledger and user level are updated based on the target experience value.

[0020] This invention triggers experience value calculation only when the verification result shows that the current task node passes multi-factor validation, achieving a strong binding between experience distribution and actual behavior, and eliminating the possibility of obtaining growth incentives through fake tasks. It extracts the task type, difficulty coefficient, and location level of the current task node from the planning snapshot, providing multi-dimensional value assessment parameters for experience value calculation, ensuring that the incentive weight of different tasks matches their actual effort. Then, a base experience value is determined based on the task type and weighted according to the difficulty coefficient to obtain the first experience value, so that higher difficulty tasks receive more experience, reflecting the fairness of the incentive. Subsequently, a location coefficient is determined based on the location level and combined with the planning bonus coefficient and the first experience value to calculate the second experience value, giving extra bonuses to high-quality locations and planned tasks, further strengthening users' willingness to perform according to the plan in high-quality locations. Finally, the second experience value is multiplied by the authenticity score to obtain the target experience value, updating the accumulated experience value ledger and user level, so that the authenticity of the behavior directly determines the final experience value; the lower the authenticity score, the less experience is obtained, effectively increasing users' motivation to consciously abide by the rules of actual execution, and indirectly enhancing the reliability of the entire skill task management.

[0021] Secondly, the present invention provides a geographical location-based skill task management device, comprising: a data acquisition module, a task map construction module, a planning snapshot generation module, a verification module, an experience calculation module, and a progress update module; The data acquisition module is used to acquire user input information, parse the user input information based on a preset large language model, and obtain structured target data. The task graph construction module is used to generate several task nodes and sequential dependencies between each task node based on the structured target data, and to construct a task graph based on the several task nodes and sequential dependencies between each task node. The planning snapshot generation module is used to associate at least one task node in the task map with a preset geographic point of interest and preset execution constraints to obtain a target task map, and generate a planning snapshot based on the target task map. The verification module is used to obtain the real-time location point stream and current task progress uploaded by the user, determine the current task node based on the current task progress and the planning snapshot, and perform multi-factor verification on the matching degree between the real-time location point stream and the current task node based on the planning snapshot to obtain the verification result. The experience calculation module is used to calculate the experience value of the current task node based on the attributes of the current task node and the verification result, and update the user's accumulated experience value ledger and user level based on the experience value. The progress update module is used to update the task progress of the user skill task based on the updated accumulated experience value ledger and user level.

[0022] Furthermore, the task graph construction module is used to generate several task nodes and sequential dependencies between each task node based on the structured target data, and to construct a task graph based on the several task nodes and sequential dependencies between each task node, including: Skill tags, target levels, and time periods are extracted based on the structured target data; The skill tags are matched with a preset skill knowledge base to obtain the corresponding skill development path; Based on the target level and the time period, the skill development path is divided into several task stages; One or more task nodes are generated for each task stage based on a preset task template library. Based on the order of the various task stages, establish sequential dependencies between the task nodes, and connect the task nodes according to the sequential dependencies to generate the task graph. Attached Figure Description

[0023] Figure 1 A schematic flowchart of a location-based skill task management method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a geolocation-based skill task management device provided in an embodiment of the present invention. Detailed Implementation

[0024] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0025] The terms "first" and "second," etc., in the specification, claims, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0026] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0027] Example 1 See Figure 1 , Figure 1 This is a schematic flowchart illustrating a location-based skill task management method according to an embodiment of the present invention. The embodiment of the present invention provides a location-based skill task management method, including steps 101 to 105, as detailed below: Step 101: Obtain user input information, parse the user input information based on a preset large language model, and obtain structured target data; Step 102: Obtain user location information, generate several task nodes and sequential dependencies between each task node based on the structured target data and user location information, and construct a task graph based on the several task nodes and sequential dependencies between each task node; Step 103: Associate at least one task node in the task graph with a preset geographic point of interest and preset execution constraints to obtain a target task graph, and generate a planning snapshot based on the target task graph; Step 104: Obtain the real-time location point stream uploaded by the user and the current task progress; determine the current task node based on the current task progress and the planning snapshot; and perform multi-factor verification on the matching degree between the real-time location point stream and the current task node based on the planning snapshot to obtain the verification result. Step 105: Calculate the experience value of the current task node based on the attributes of the current task node and the verification result, and update the user's accumulated experience value ledger and user level based on the experience value; Step 106: Update the task progress of the user skill task based on the updated accumulated experience value ledger and user level.

[0028] In this embodiment, the user inputs their skill development goals through a terminal device, and the system performs semantic parsing and field extraction on the input to generate structured target data.

[0029] In this embodiment, the system automatically generates a task graph containing multiple task nodes based on the skill tags, target levels, and time periods in the structured target data, and establishes sequential dependencies for each task node that conform to the skill progression logic. For task nodes that need to be executed offline, the system matches suitable location entities from the geographic point of interest database and configures execution constraints for the task node, including the expected execution time window, minimum dwell time, and fence radius. After user confirmation, the system hashes and solidifies the complete planning data, including the task graph, geographic associations, and constraints, to form an immutable planning snapshot. During the execution phase, the terminal continuously collects and uploads the user's real-time location stream. Based on the currently recorded task progress and the dependencies in the planning snapshot, the system automatically identifies the next task node to be executed. When it detects that the user's location coordinates have entered the fence range of the geographic point of interest bound to the node, it determines the node as the current task node.

[0030] Subsequently, the geographic points of interest (POIs) and sequential location information of the current task node are extracted from the planning snapshot. This data is then combined with simulated location markers, movement speed, dwell time, completion status of preceding tasks, and the validity of execution time from the real-time location stream for multi-factor comprehensive verification, generating a verification result including an authenticity score. If the verification result indicates success, the system calculates a basic experience value based on the task type, difficulty level, and location level of the bound POIs for the current task node. This experience value is then weighted according to the authenticity score in the verification result to obtain a final experience value, which is added to the user's accumulated experience value ledger in real time. The user's level is then updated based on the accumulated value.

[0031] Finally, the system updates the task progress based on the updated level and ledger status, marks the current task node as completed, and unlocks the next task node according to the sequential dependencies in the planning snapshot. When the user's level or accumulated experience value reaches the preset advancement threshold, the planning of the next stage of skill growth tasks is automatically triggered.

[0032] This invention can be widely applied to vocational skills training platforms. The system automatically plans learning paths based on the certification goals selected by trainees and ensures the authenticity of learning behavior through on-site check-in and residence verification of offline training and practical operations. Only trainees who genuinely complete the practical components can obtain corresponding experience points and unlock subsequent advanced courses. It can also be applied to personal growth management. Users can set personal skill goals such as fitness instructor certification or language proficiency exams. The system breaks down these goals into specific daily or weekly tasks, using location verification to encourage users to complete training or learning in real venues, preventing fraudulent check-ins. Simultaneously, experience points and a level mechanism incentivize users to continuously execute their plans. This invention is also applicable to internal employee skills enhancement scenarios within enterprises. Human resources departments can customize growth tasks for employees based on job competency models. Forced geographical location binding and residence time verification ensure that employees actually participate in offline training or practice. Assessment results are automatically recorded as employee growth records, providing objective evidence for promotion evaluations.

[0033] As a specific example of an embodiment of the present invention, a user inputs unstructured growth goal text into an application on their mobile terminal (such as a smartphone), for example: "I want to obtain a basic barista certification within 3 months. Please help me plan, preferably with practical training at a coffee training school near my home (Chaoyang District)." After the system obtains the text, it calls a pre-trained Large Language Model (LLM) as an artificial intelligence semantic parsing model. The model performs intent recognition and entity extraction on the text, outputting structured data: skill tag = "coffee making", target level = "basic", time period = "90 days", and geographical preference = "Chaoyang District".

[0034] Subsequently, the system automatically generates multiple task nodes based on this structured data, including online learning of basic theory, in-store latte art practice training, and taking certification exams. It also establishes sequential dependencies between task nodes based on skill progression logic, such as requiring theoretical learning to be completed before practical training can begin, thus constructing a complete task graph.

[0035] As a specific example of this invention, the "coffee making" skill tag is first matched with a preset skill knowledge base to obtain a standard growth path: basic theory, latte art practice, and certification exam. Then, based on the "90 days" and "beginner" goals, this path is divided into three task stages. Next, based on the geographical constraints determined in the previous step, the system filters a list of candidate POIs located in Chaoyang District that provide coffee training services from a preset task template library and POI database. Finally, the system calls an artificial intelligence task planning model, taking the task stage, available POI list, and time period as input. This model, through optimization algorithms that consider factors such as location distance, venue level, and user time window, generates specific task nodes for the user: for example, task node A (online learning of basic coffee theory, no geographical constraints), task node B (attending a latte art practice class at "XX Coffee Academy," associated with the academy's POI), and task node C (attending the beginner exam at "YY Certification Center," associated with the center's POI). The model also establishes sequential dependencies: A must be completed to unlock B, and B must be completed to unlock C, thus constructing a complete task graph.

[0036] Next, pre-defined geographic points of interest are associated with task nodes that require offline execution. For example, the latte art practice task is associated with the coordinates of a vocational training college. Execution constraints are then configured for this task, including an expected execution time window on a weekend evening, a minimum dwell time of no less than sixty minutes, and a geofence with a radius of one hundred meters. After the user confirms the above task map, associations, and constraints, they are serialized and a hash digest is calculated to generate an immutable planning snapshot.

[0037] During the execution phase, users continuously upload real-time location data via their mobile devices. The system parses the next task node from the planning snapshot based on the current task progress and identifies the current task node when the user's location coordinates enter the geofence bound to that task node. Subsequently, the system performs multi-factor verification on the match between the real-time location data and the task node, such as checking simulated location markers, calculating movement speed, calculating dwell time within the geofence, verifying the completion status of previous tasks, and comparing execution time windows, to arrive at a comprehensive verification result. If the verification passes, the system calculates the experience points corresponding to this task based on the task type, difficulty level, and location level, and adds them to the user's accumulated experience points ledger. Simultaneously, the user's level is updated according to a preset threshold. Finally, the system automatically advances the task progress based on the updated level and ledger, marking completed task nodes as complete and unlocking the next task node. If the level reaches the advancement condition, the generation of the next stage of the skill growth task map is triggered.

[0038] In this embodiment, by acquiring structured target data of user skill tasks, the abstract needs of users can be transformed into standardized information that can be processed by machines, laying a data foundation for subsequent automated planning. Based on this, task nodes with sequential dependencies are generated and a task graph is constructed, creating a logical closed loop between tasks, avoiding arbitrariness in task execution, and improving the systematicity and reliability of planning. Next, geographic points of interest and execution constraints are associated with task nodes, and a planning snapshot is generated, solidifying the spatial and temporal constraints of task execution, providing an irrefutable benchmark for subsequent verification, and enhancing the credibility of management. Furthermore, by acquiring real-time location point streams and current task progress... Based on the planning snapshot, the system determines the current task node, enabling real-time perception and automated progress of task status, reducing errors caused by human intervention. It then performs multi-factor verification of the matching degree between real-time location points and the current task node, obtaining verification results. This allows for cross-verification of the authenticity of user behavior from multiple dimensions, effectively preventing cheating methods such as simulated location and brief passing through. Finally, based on task node attributes and verification results, experience values ​​are calculated and the accumulated experience value ledger, user level, and task progress are updated. This binds growth incentives to behavioral authenticity; only tasks that pass verification can provide experience feedback, thus improving the overall reliability and authenticity of skill task management.

[0039] In this embodiment, the step of generating several task nodes and their sequential dependencies based on the structured target data and user location information, and constructing a task graph based on the several task nodes and their sequential dependencies, includes: Skill tags, target levels, and time periods are extracted based on the structured target data; The skill tags are matched with a preset skill knowledge base to obtain the corresponding skill development path; Based on the target level and the time period, the skill development path is divided into several task stages; the geographical constraint range is determined based on the user's location information, and one or more task nodes are generated for each task stage based on the geographical constraint range and a preset task template library; Based on the order of the various task stages, establish sequential dependencies between the task nodes, and connect the task nodes according to the sequential dependencies to generate the task graph.

[0040] In this embodiment, the system first performs semantic parsing and field normalization on the user-input natural language target or structured form data, extracting at least one skill tag, one target level, and one time period parameter. Then, the system uses the extracted skill tag as an index to perform matching retrieval in a pre-built skill knowledge base. This knowledge base stores standardized growth paths for different skill domains in a directed graph or tree structure, with each path consisting of multiple ability nodes arranged in ascending order of difficulty. Based on the matching results, the system obtains the complete skill growth path corresponding to the user's skill tag.

[0041] In this embodiment, the structured target data includes at least the following parameters: skill tags characterizing the specific skill type the user expects to learn or improve; target levels identifying the skill level or certification level the user hopes to achieve; and time periods defining the planned total duration for completing the skill task. Furthermore, the structured target data may include: geographical range parameters constraining the acceptable spatial area for the user when performing offline tasks; budget range parameters reflecting the range of economic costs the user is willing to pay to complete the skill task; user preference parameters recording the user's personalized preferences regarding task execution time, location type, etc.; certificate import information linking relevant certification data already obtained or targeted by the user; and historical level status inheriting the user's existing growth level or accumulated experience value ledger information.

[0042] In this embodiment, the system determines the terminal capability node that needs to be reached in the path based on the target level, and performs time axis mapping on the path according to the time period, dividing the sequence of capability nodes from the starting node to the ending node of the skill growth path into several consecutive task stages, with each stage corresponding to a time sub-interval and capability subset.

[0043] In this embodiment, for each divided task stage, the system calls a preset task template library. The template library has preset parameterizable task templates according to stage type, such as cognition stage, practice stage, certification stage, etc. The system selects the matching template from the template library according to the attributes of the current stage and instantiates it to generate one or more task nodes. Each task node includes task type, default execution parameters and configurable constraint fields.

[0044] In this embodiment, the system establishes sequential dependencies for all generated task nodes based on the order of each task stage on the timeline and the dependency logic of capability nodes within each stage. This ensures that all preceding task nodes of a subsequent task node must be completed before it can be activated. The system then connects these task nodes in a directed acyclic graph according to their dependencies, generating a structured task graph that can be used for subsequent geographic association and constraint configuration.

[0045] As a specific example of an embodiment of the present invention, the system performs semantic parsing on the natural language target or templated selection input by the user, and extracts skill tags representing skill direction such as coffee making, target level representing the expected degree of achievement such as beginner, and target period representing the execution time of the plan such as ninety days.

[0046] The system matches extracted skill tags with a pre-set skill knowledge base, which stores standardized progression paths for different skills from beginner to advanced. For example, coffee making skills include three stages: basic understanding, practical training, and certification assessment. The system determines the required depth of development based on the target level and adapts the standard path to a timeframe based on the target period, dividing the skill development path into several task stages. For example, the first stage is a 30-day basic understanding stage, the second stage is a 40-day skill practice stage, and the third stage is a 20-day certification preparation stage. For each task stage, the system accesses a pre-set task template library, which stores standardized task descriptions and parameter templates categorized by task type. The system instantiates one or more specific task nodes from the template library based on the attributes of the current stage; for example, generating offline training course tasks and practical operation assessment tasks in the skill practice stage.

[0047] Finally, based on the inherent sequential order of each task stage, the system establishes clear sequential dependencies for all generated task nodes, ensuring that task nodes in subsequent stages can only be unlocked after the task nodes in the preceding stages are completed. These task nodes are then connected by directed edges according to their dependencies, thus forming a complete task graph. This task graph provides a structured execution framework for subsequent geographic point of interest binding and behavior verification.

[0048] In this embodiment, by extracting skill tags, target levels, and time periods from structured target data, the core elements and time constraints of user skill improvement can be accurately captured, providing parameter basis for personalized task generation. Next, the skill tags are matched with a preset skill knowledge base to obtain the corresponding skill growth path, ensuring that the task generation process follows objective skill progression rules and avoids biases from subjective planning. Furthermore, the skill growth path is divided into several task stages based on the target level and time period, decomposing long-term goals into executable and measurable phased tasks, reducing execution complexity. Then, based on a preset task template library, one or more task nodes are generated for each task stage, achieving standardization and reusability of task types and improving the efficiency and consistency of task generation. Finally, the sequential dependencies between task nodes are established according to the arrangement order of each task stage, and a task graph is generated, making the pre- and post-task logic clear and providing a reliable benchmark for subsequent sequence verification, thereby enhancing the structured nature of task planning and the logical rigor of execution.

[0049] In this embodiment, associating at least one task node in the task graph with a preset geographic point of interest and preset execution constraints to obtain a target task graph, and generating a planning snapshot based on the target task graph, includes: Obtain the task nodes to be associated in the task graph, and determine the task type and geographical constraint range of the task nodes to be associated. Based on the task type and geographical constraints, several candidate geographic points of interest are matched in a preset geographic point of interest database. The matching degree between the task node to be associated and each candidate geographic point of interest is calculated based on a preset matching degree algorithm, and the candidate geographic point of interest with the highest matching degree is associated with the task node to be associated. The target task graph is obtained by associating the nodes with the preset execution constraints, and a planning snapshot is generated based on the target task graph.

[0050] In this embodiment, the task nodes to be associated in the task graph are traversed first. The task nodes to be associated are the task nodes in the task graph that have not yet been associated. Based on the task type field of each task node and the user's preset geographical preference parameters, the task type attributes of the node, such as online learning, offline training, certification exam, etc., and the geographical constraints, such as search radius, city or region boundary, are determined.

[0051] Subsequently, the system uses task type as the primary key for filtering and geographical constraints as the spatial filtering condition to perform a structured query in a pre-set database of geographic points of interest. Each record in this database contains multi-dimensional attributes such as the unique identifier of the point of interest, latitude and longitude coordinates, category label set, business hours interval, venue rating, and user reviews. The system then selects several candidate geographic points of interest that simultaneously meet the requirements of task type matching and spatial location falling within the constraints.

[0052] For each candidate point, the system invokes a preset matching algorithm. This algorithm calculates the spatial distance score, label semantic similarity score, time window overlap score, and location level normalization score between the task node and the candidate point. These scores are then weighted and summed according to preset weights to obtain a comprehensive matching score. The system selects the geographic point of interest with the highest comprehensive matching score from all candidate points and establishes a binding relationship between this point of interest and the task node to be associated.

[0053] In this embodiment, by obtaining the task type and geographical constraint range of the task nodes to be associated in the task graph, the functional requirements and spatial boundaries of each task for the execution location can be clarified, providing filtering conditions for geographical matching. By matching several candidate geographic points of interest in a preset geographic point of interest database according to the task type and geographical constraint range, the candidate set can be quickly narrowed down from massive location data, improving matching efficiency. Then, based on a preset matching degree algorithm, the matching degree between the task node to be associated and each candidate geographic point of interest is calculated, and the highest matching degree is associated, so that the binding result between the task and the location has quantitative optimality, avoiding the actual unexecutability problem caused by arbitrary binding. Finally, by associating the task node to be associated with the preset execution constraints and generating a planning snapshot, behavioral constraints such as time and dwell time are solidified, making the task execution conditions more complete, thereby improving the rationality of geographical association and the completeness and reliability of the planning snapshot.

[0054] In this embodiment, the step of associating the target task node with the preset execution constraints to obtain the target task graph, and generating a planning snapshot based on the target task graph, includes: The task node to be associated is bound based on execution constraints, which include the expected execution time window, minimum dwell time, and fence radius. The task map is updated based on the associated geographic points of interest and execution constraints to generate the target task map; The target task graph is serialized to obtain task sequence data; The hash digest of the task sequence data is calculated based on a preset hash algorithm, and an immutable planning snapshot is generated based on the hash digest.

[0055] In this embodiment, preset execution constraints corresponding to the task node type are further obtained from the constraint template library, including the expected execution time window, minimum dwell time, and geofence radius, and these constraints are associated and bound to the same task node. After completing the above processing of all task nodes to be associated, the system determines the updated task graph as the target task graph, and performs serialization encapsulation and hash digest calculation on the target task graph to generate an immutable and fixed planning snapshot.

[0056] As a specific example of an embodiment of the present invention, taking the latte art practice task node in the coffee making skill task as an example, the task node is first obtained and its task type is determined to be an offline skills training course. At the same time, the geographical constraint range is read from the user's structured target data as an area within a radius of five kilometers centered on the user's home address.

[0057] The system searches for points of interest (POIs) within a 5-kilometer radius of the user's location, using either the Gaode Maps API or its own training institution database, and filters out three candidate training institutions. For the first institution, the system calculates its distance score from the user's home, the match between its tags and latte art practice, the overlap between its business hours and the user's preferred weekend evening hours, and the institution's user rating on the platform. A weighted average of these factors yields a match score of 85. Similarly, the second institution scores 92 points, and the third institution scores 78 points. The system then associates the second institution (with the highest score of 92) with the latte art practice task node.

[0058] Subsequently, the system loads preset execution constraints for the task node, including setting the expected execution time window to 6 PM to 10 PM on weekend evenings, setting the minimum dwell time to 60 minutes, and establishing a geofence with a radius of 50 meters centered on the organization's coordinates. The system replaces the corresponding node in the original task map with the task node bound to the geographic point of interest and execution constraints, generating a target task map. Finally, the system serializes the target task map and calculates its SHA256 hash digest, forming an immutable planning snapshot. Subsequent execution phases will use this snapshot as a benchmark for authenticity verification.

[0059] In this embodiment, by binding execution constraints with task nodes to be associated, each task has a clear spatiotemporal execution boundary, providing a quantitative basis for subsequent authenticity verification. The task graph is then updated based on the task nodes bound with geographic points of interest and execution constraints to generate the target task graph, integrating all constraint information into a unified task graph structure and maintaining data consistency and integrity. The target task graph is then serialized to obtain task sequence data, converting the complex graph structure into a linear format that is easy to store and transmit, reducing data processing complexity. Finally, a hash digest of the task sequence data is calculated based on a preset hash algorithm, generating an immutable planning snapshot. Utilizing the unidirectional and unique characteristics of the hash digest, the planning snapshot cannot be tampered with or forged once generated, thus providing an immutable verification benchmark for the entire execution phase, significantly improving the authenticity and anti-tampering capabilities of management.

[0060] In this embodiment, obtaining the real-time location stream uploaded by the user and the current task progress, and determining the current task node based on the current task progress and the planning snapshot, includes: Get the real-time location data stream uploaded by the user and the current task progress; Extract the sequence of task nodes and the sequential dependencies between each task node from the planning snapshot; Based on the current task progress, determine the set of completed task nodes, and based on the sequential dependency relationship and the set of completed task nodes, determine the task nodes to be executed; The first fence range of the geographic points of interest associated with the task node to be executed is obtained based on the planning snapshot; when the location coordinates in the real-time location point stream are detected to enter the first fence range, the task node to be executed is determined as the current task node.

[0061] In this embodiment, the system continuously receives real-time location point streams uploaded by users through the data interface of the mobile terminal. Each location point contains at least latitude and longitude coordinates, timestamps, and positioning accuracy information. At the same time, the system maintains a current task progress variable that records the user's completed task indicators.

[0062] In this embodiment, the system deserializes and extracts the complete sequence of task nodes and the sequential dependencies between nodes stored as directed edges from the fixed-storage planning snapshot. The system reads the list of completed task identifiers in the current task progress and matches them with the nodes in the task node sequence to generate a set of completed task nodes. The system traverses the task node sequence, and for each task node not marked as completed, checks whether all its predecessor dependent task nodes already exist in the set of completed task nodes, and determines the first task node that meets this condition as the task node to be executed.

[0063] In this embodiment, the system reads the geographic points of interest (POIs) bound to the task node to be executed from the planning snapshot, extracts the preset fence radius, and generates a first fence range centered on the latitude and longitude coordinates of the POI. The system continuously performs geometric determination between the latest location coordinates in the real-time location point stream and the first fence range. When it detects that the location coordinates fall inside the fence boundary, it marks the task node to be executed and determines it as the current task node, triggering the subsequent multi-factor validation process.

[0064] As a specific example of an embodiment of the present invention, taking the latte art practice task node in a coffee-making skills task as an example, the user carries a smartphone with the application installed. The system receives the GPS location point stream uploaded by the phone in real time. Each location point records latitude and longitude coordinates, timestamp, and positioning accuracy value. The system also records the user's task progress status, where the theoretical learning task node in the basic cognition stage has been marked as completed. The system reconstructs the complete task node sequence from the planning snapshot: node A is online learning of basic coffee theory, node B is attending latte art practice training in a store, node C is taking the primary certification exam, and the sequential dependency relationship between node A and node B, and node B and node C. The system reads the current task progress and knows that node A is completed, while nodes B and C are not completed. It then checks that node B's predecessor, node A, already exists in the set of completed task nodes, and therefore determines node B as the task node to be executed. The system obtains the latitude and longitude coordinates of a vocational training college bound to node B and its fence radius of fifty meters from the planning snapshot, and generates a circular geofence with a radius of fifty meters centered on these coordinates. When the user actually arrives near the training academy, the system detects that the location coordinates reported by the mobile phone have entered the 50-meter fence range. Then, it automatically identifies node B as the task node that the user is currently executing, without the user having to manually click to start the task, thereby initiating a multi-factor verification of the location authenticity of the task node.

[0065] In this embodiment, by acquiring the real-time location point stream uploaded by the user and the current task progress, the system can dynamically capture the user's real-time location behavior and task completion status, providing real-time data support for task advancement. Furthermore, by extracting the task node sequence and the sequential dependencies between each task node from the planning snapshot, the system can accurately grasp the logical constraints between tasks, providing a basis for determining the next task to be executed. Next, based on the current task progress, the system determines the set of completed task nodes, and based on the sequential dependencies and the completed set, determines the task node to be executed, achieving automatic inference of task progress and avoiding errors or cheating that might be introduced by manual selection by the user. Then, based on the planning snapshot, the system obtains the fence range of the geographic points of interest associated with the task node to be executed, and when the real-time location coordinates are detected to enter the fence, the task node to be executed is determined as the current task node. This ensures that the activation of the current task is entirely triggered by location rather than by the user's subjective declaration, thereby ensuring the correctness of the task execution order and the authenticity of the location, improving the automation and reliability of task advancement.

[0066] In this embodiment, the step of performing multi-factor verification on the matching degree between the real-time location point stream and the current task node based on the planning snapshot to obtain the verification result includes: Extract the geographic points of interest data associated with the current task node and the sequential position of the current task node in the task node sequence from the planning snapshot; Based on the real-time location point stream, the user's location coordinates, timestamps, positioning accuracy, and simulated positioning markers at each time point are obtained; Based on the geographic point of interest data and the sequential location, multi-factor verification is performed on the location coordinates, timestamp, positioning accuracy, and simulated positioning marker to obtain the verification results.

[0067] In this embodiment, the complete data structure corresponding to the current task node is obtained by deserialization from the solidified planning snapshot, and two key verification criteria are extracted from it: the first is the geographic point of interest data associated with the task node, specifically including the target location coordinates of the point of interest, the preset fence radius, the minimum dwell time, the expected execution time window, and the location level; the second is the sequential position index of the task node in the overall task node sequence and the record of its related sequential dependencies.

[0068] Simultaneously, the system acquires real-time location data uploaded by users during the execution period in batches through the mobile terminal's positioning interface. Each location point is parsed to extract latitude and longitude coordinates, acquisition timestamp, horizontal positioning accuracy, and underlying simulation positioning marker fields such as isMock or provider type. The system aligns the static baseline data extracted from the planning snapshot with the dynamic sensing data extracted from the real-time location data stream. It then performs spatiotemporal matching of the location data stream with the time window and geofence of the task node according to chronological order. This provides a standardized input data structure for subsequent simulation positioning detection, speed anomaly detection, sequence compliance detection, dwell time detection, and time deviation detection, thereby enabling systematic multi-factor verification of the matching degree between the real-time location data stream and the current task node.

[0069] In this embodiment, by extracting the geographic points of interest data associated with the current task node and its sequential position in the task node sequence from the planning snapshot, a dual reference from the fixed benchmark is provided for multi-factor verification, realizing cross-validation of user behavior. Any single-dimensional anomaly is difficult to pass the overall verification, thereby significantly improving the probability of cheating behavior being identified and enhancing the ability to ensure the authenticity of task execution results.

[0070] In this embodiment, the geographic point of interest (POI) data includes target location coordinates, target fence radius, and target minimum dwell time; the multi-factor verification of the location coordinates, timestamp, positioning accuracy, and simulated positioning marker based on the POI data and the sequential location to obtain the verification result includes: A simulated positioning test is performed based on the simulated positioning marker. If the simulated positioning marker is detected as true, the simulated positioning test result is "fail". The moving speed is calculated based on the distance and time difference between adjacent coordinates in the real-time location point stream. If the moving speed exceeds a preset speed threshold, a speed anomaly marker is generated. Based on the sequential position of the current task node and the sequential dependencies in the planning snapshot, check whether the current task node has completed all the preceding task nodes. If not, generate an order anomaly flag. Based on the location coordinates in the real-time location point stream and the target location coordinates, the continuous dwell time of the user within the radius of the target fence is calculated. If the continuous dwell time is lower than the target minimum dwell time, an insufficient dwell time marker is generated. Based on the expected execution time window of the current task node and the timestamp of the real-time location point stream, if the user's actual execution time does not fall within the expected execution time window, a time deviation marker is generated. Based on the simulated positioning detection results, velocity anomaly markers, sequence anomaly markers, insufficient dwell time markers, and time deviation markers, the verification results are generated, and the verification results include an authenticity score.

[0071] In this embodiment, the simulated positioning tag field carried by each location point in the real-time location point stream is first read. If the tag value is true or the positioning provider type belongs to the test simulator category, the simulated positioning detection result is directly determined to be unsuccessful. This result can be used as a hard rejection condition.

[0072] In this embodiment, two consecutive adjacent location points in the real-time location point stream are traversed, and the instantaneous movement speed is calculated by dividing the spherical distance between the two points by the time difference. This speed is then compared with a preset speed threshold. The speed threshold is configured with different values ​​according to the task type, such as the upper limit for walking mode or the upper limit for motor vehicle mode. When the instantaneous speed exceeds the threshold, a speed anomaly marker is generated and the time period of the anomaly is recorded.

[0073] In this embodiment, the sequential position index of the current task node and its list of predecessor dependent node identifiers are read from the planning snapshot. Each predecessor dependent node is traversed and verified to see if it exists in the set of completed task nodes. If any predecessor dependent node is not completed, an order anomaly flag is generated.

[0074] In this embodiment, all points whose location coordinates are less than or equal to the target location coordinates are selected from the real-time location point stream. After sorting by timestamp, subsequences that are continuously located within the fence are identified. The start and end time differences of each subsequence are accumulated to obtain the continuous dwell time. If the dwell time is lower than the target minimum dwell time, an insufficient dwell time marker is generated.

[0075] In this embodiment, the start and end times of the expected execution time window of the current task node are extracted. It is then checked whether the actual execution interval formed by the timestamp of the first entry into the fence and the timestamp of the last exit from the fence in the real-time location point stream intersects with the window. If there is no intersection or the intersection duration is less than a preset ratio, a time deviation marker is generated.

[0076] In this embodiment, the simulated positioning detection result is used as a veto item. If it passes, the speed abnormality mark, sequence abnormality mark, insufficient dwell time mark, and time deviation mark are used as deduction items. The comprehensive deduction is calculated according to the preset weight allocation to generate an authenticity score, which is output as the final verification result.

[0077] In this embodiment, detection is performed based on simulated positioning markers. If the marker is genuine, the system directly determines that the operation fails, cutting off the path of location spoofing using simulated positioning software at the source, thus providing a hard anti-spoofing effect. By calculating the movement speed based on the distance and time difference between adjacent coordinates and comparing it with a speed threshold, illogical behaviors such as ultra-high-speed movement can be identified, preventing cheating methods such as bypassing dwell detection by quickly jumping points. Furthermore, the completion of preceding tasks is verified based on the sequential position of the current task node and the sequential dependencies in the planning snapshot, ensuring that the logical order of task execution cannot be reversed and preventing skipping key steps and directly completing the task. The system analyzes user behavior to continue tasks. By statistically analyzing the continuous dwell time of users within the target fence radius and comparing it with the minimum dwell time, it can distinguish between brief passages and genuine stays, ensuring that users do indeed invest sufficient time at the task location. By comparing the actual execution time with the expected execution time window and generating time deviation markers, it can detect abnormal execution behavior that deliberately avoids the specified time period. By integrating the above multiple detection results and markers to generate a verification result including an authenticity score, the multi-factor judgment is quantified into a comprehensive score, making the verification result both comprehensive and comparable, thereby comprehensively improving the accuracy and reliability of task authenticity judgment.

[0078] In this embodiment, the step of calculating the experience value of the current task node based on its attributes and the verification result, and updating the user's accumulated experience value ledger and user level based on the experience value, includes: When the verification result indicates that the current task node passes the multi-factor verification, the task type, difficulty coefficient, and location level of the geographic point of interest of the current task node are extracted from the planning snapshot. A base experience value is determined based on the task type, and the base experience value is weighted according to the difficulty coefficient to obtain a first experience value. The location coefficient is determined based on the location level, and a second experience value is calculated based on the preset planning bonus coefficient, the location coefficient, and the first experience value. The target experience value is obtained by multiplying the second experience value by the authenticity score; and the user's accumulated experience value ledger and user level are updated based on the target experience value.

[0079] In this embodiment, the current task node is deemed to have passed verification and the experience value calculation is triggered only when the authenticity score reaches the preset pass threshold and no hard rejection condition is triggered.

[0080] In this embodiment, the task type field, difficulty coefficient value, and location level field of the associated geographic point of interest for the current task node are read from the planning snapshot. The task type distinguishes different categories such as online learning, offline training, and certification exams. The difficulty coefficient reflects the level of advancement of the task in the skill development path, and the location level represents the authority or facility level of the execution location. A preset basic experience value mapping table is queried based on the task type to obtain the corresponding basic experience value. This basic experience value is then multiplied by the difficulty coefficient; the higher the difficulty coefficient, the larger the weighted first experience value, thus reflecting the positive impact of task difficulty on growth value.

[0081] In this embodiment, the corresponding venue coefficient is matched from a preset venue coefficient table according to the venue level. A higher venue coefficient corresponds to a better venue. At the same time, the planning bonus coefficient is read from the planning snapshot to distinguish whether the current task belongs to the planned task confirmed by the user. Then, the first experience value is multiplied by the product of the venue coefficient and the planning bonus coefficient to obtain the second experience value, so that the behavior of choosing a high-quality execution location and following the planned path will be given additional incentives.

[0082] In this embodiment, the authenticity score is further read and multiplied by the second experience value to obtain the final target experience value. The higher the score, the closer the target experience value is to the second experience value; if the score is too low, the experience value is deducted proportionally. The system adds the target experience value to the user's accumulated experience value ledger, generates an updated accumulated experience value, and determines and updates the user's current level based on the range of the accumulated value in the preset level threshold table. This completes the full-link calculation from task attributes, execution location, planning compliance, and behavioral authenticity to experience value accumulation and level growth.

[0083] In this embodiment, experience value calculation is triggered only when the verification result shows that the current task node passes multi-factor validation, achieving a strong binding between experience distribution and actual behavior, and eliminating the possibility of obtaining growth incentives through fake tasks. Furthermore, the task type, difficulty coefficient, and location level of the current task node are extracted from the planning snapshot, providing multi-dimensional value assessment parameters for experience value calculation, ensuring that the incentive weight of different tasks matches their actual effort. Next, a basic experience value is determined based on the task type and weighted according to the difficulty coefficient to obtain the first experience value, ensuring that higher-difficulty tasks yield more experience, reflecting the fairness of the incentive. Subsequently, a location coefficient is determined based on the location level and combined with the planning bonus coefficient and the first experience value to calculate the second experience value, giving extra bonuses to high-quality locations and planned tasks, further strengthening users' willingness to execute according to the plan in high-quality locations. Finally, the second experience value is multiplied by the authenticity score to obtain the target experience value, updating the accumulated experience value ledger and user level, so that the authenticity of the behavior directly determines the final experience value; the lower the authenticity score, the less experience is gained, effectively increasing users' motivation to consciously abide by the rules of actual execution, indirectly enhancing the reliability of the entire skill task management.

[0084] Example 2 In practical applications, while adaptable to general scenarios across all offline venues, existing solutions lack a more reliable hardware-level verification solution that merchants / organizations with compliant operations and precise customer acquisition needs can choose independently. Furthermore, the current solution lacks a linkage mechanism between verification reliability and user growth incentives. It cannot guide users to choose higher-reliability verification methods through differentiated incentive rules, nor can it incentivize offline merchants to deploy reliable verification devices. Ultimately, this results in inconsistent task verification reliability and the inability to achieve a closed-loop incentive system between users and merchants through technical means. This embodiment also achieves dual-mode compatibility between basic mobile phone location verification and robust hardware verification using dynamic QR code check-in machines, meeting the differentiated deployment needs of different merchants. It establishes a technical linkage mechanism between verification reliability and growth incentives, guiding users to choose higher-reliability verification methods through differentiated experience value bonus rules, while simultaneously incentivizing offline merchants to deploy reliable verification devices. This achieves a technical closed-loop system of tiered task verification reliability and two-way incentives between users and merchants, further improving the reliability and implementability of skills-based task management.

[0085] This invention provides another location-based skill task management method, including: acquiring user input information, parsing the user input information based on a preset large language model, and obtaining structured target data; Obtain user location information, generate several task nodes and sequential dependencies between each task node based on the structured target data and user location information, and construct a task graph based on the several task nodes and sequential dependencies between each task node; Associate at least one task node in the task graph with a preset geographic point of interest and preset execution constraints to obtain a target task graph, and generate a planning snapshot based on the target task graph; The system acquires real-time location data uploaded by the user and the current task progress, determines the current task node based on the current task progress and the planning snapshot, and performs multi-factor verification on the matching degree between the real-time location data and the current task node based on the planning snapshot to obtain the verification result. The experience value of the current task node is calculated based on the attributes of the current task node and the verification result, and the user's accumulated experience value ledger and user level are updated based on the experience value. The task progress of the user's skill tasks is updated based on the updated accumulated experience value ledger and user level.

[0086] In this embodiment, at least one task node in the task graph is associated with a preset geographic point of interest (POI) and preset execution constraints to obtain a target task graph. The POI is marked with verification mode configuration information. The verification modes include a basic mobile phone positioning verification mode and a dynamic QR code check-in machine hardware strong verification mode. The POI corresponding to the hardware strong verification mode is pre-bound to a corresponding dynamic QR code check-in machine. The execution constraints include verification rules corresponding to the verification mode and an experience value bonus coefficient specific to the hardware strong verification mode. An immutable planning snapshot is generated based on the target task graph.

[0087] In this embodiment, the real-time location stream uploaded by the user and the current task progress are acquired. Based on the current task progress and the planning snapshot, the current task node is determined. According to the verification mode corresponding to the current task node, the corresponding verification process is executed: if it is the basic mobile phone positioning verification mode, the matching degree between the real-time location stream and the current task node is verified by multiple factors based on the planning snapshot to obtain the verification result; if it is the dynamic QR code check-in machine hardware strong verification mode, the check-in data generated by the user scanning the dynamic QR code check-in machine bound to the geographic interest point corresponding to the current task node through the terminal is acquired, combined with the real-time location stream, and multi-factor verification is performed based on the planning snapshot to obtain the verification result; then, based on the attributes of the current task node, the verification result, and the corresponding verification mode, the experience value of the current task node is calculated. Where the hardware strong verification mode passes the verification, the experience value is calculated according to the preset experience value bonus coefficient; and the user's accumulated experience value ledger and user level are updated based on the final experience value; the task progress of the user's skill task is updated based on the updated accumulated experience value ledger and user level.

[0088] In this embodiment: the recommended verification mode is determined based on the task type and difficulty level of the task node; for basic learning tasks and low-difficulty task nodes, the mobile phone location basic verification mode is recommended by default; for authentication tasks, high-difficulty tasks, and high-weight tasks, the dynamic QR code check-in machine hardware strong verification mode is recommended by default; when matching corresponding geographic points of interest, geographic points of interest matching the verification mode are displayed first, and for geographic points of interest in the hardware strong verification mode, the corresponding experience value bonus coefficient is simultaneously marked.

[0089] In this embodiment, locations that have not deployed dynamic QR code check-in machines can only configure the basic mobile phone location verification mode; locations that have completed the binding and qualification certification of dynamic QR code check-in machine devices can configure the hardware strong verification mode, and can independently adjust the experience value bonus coefficient within a preset range. The adjusted coefficient is synchronized to the cloud platform and the planning snapshot.

[0090] In this embodiment, Scenario A is an example of a small community bookstore without any deployed hardware. Its administrator logs into the merchant management terminal, and the system automatically detects the lack of a bound attendance machine, therefore only enabling the "Basic Mobile Location Verification Mode" configuration option. The merchant can set basic constraints such as fence radius and expected business hours window, and confirm activation. The system marks the location's geographic points of interest as basic verification mode. Scenario B is a merchant requiring strong hardware verification: a chain coffee training academy ("XX Coffee Academy"), which purchased and deployed dynamically changing QR code attendance machines to improve the authenticity of student visits and obtain accurate traffic data. The administrator executes the following process: Power on the time clock and connect it to the network. By scanning the QR code on the back of the device through the management terminal and entering the venue qualification documents, the cloud-based pre-binding and qualification authentication of the time clock's unique device identifier and the venue's geographic point of interest identifier is completed.

[0091] Select "Hardware Strong Verification Mode" on the management terminal and set the QR code transformation cycle to 15 seconds.

[0092] Within the range allowed by the platform (e.g., 1.0 to 2.0), the experience value bonus coefficient is set to 1.5, and this coefficient is synchronized to the cloud platform in real time.

[0093] Configure the types of tasks that the attendance machine can handle, such as high-weight tasks like "Latte Art Practice Class" and "Graduation Assessment".

[0094] In this embodiment, if the current task node is configured to be dual-mode compatible, the user can choose the verification mode independently; if the hardware strong verification mode is selected and the verification is passed, the bonus experience value is calculated; if the mobile phone positioning basic verification mode is selected and the verification is passed, the basic experience value is calculated and no bonus is enjoyed.

[0095] The user inputs "I want to obtain a junior barista certificate in 3 months and practice at a location near my home with bonus points" via natural language. The system's AI planning model automatically recommends a hardware-strong verification mode based on the type of the task node "Latte Art Practice Class" (certification, high difficulty). When matching POIs, the system prioritizes displaying "XX Coffee Academy" with deployed check-in machines, prominently noting "Hardware-strong verification, experience points +50%" on the user's interface. The user can choose: if they select this bonus location, the task node is bound to the hardware-strong verification mode; if they select a regular coffee shop without a check-in machine, it is bound to the basic mobile location verification mode with no bonus. This example assumes the user selects a bonus location.

[0096] Once the user arrives at the task location, the system detects that the location has entered the fence and automatically performs multi-factor verification based on location point flow, including simulated positioning, speed, dwell time, sequence, and time window, to generate an authenticity score.

[0097] In this embodiment, the QR code generation and synchronization rules of the dynamically changing QR code attendance machine include: the attendance machine has a built-in tamper-proof security encryption chip and a unique device identifier; the unique device identifier and the unique identifier of the corresponding geographic point of interest are pre-bound and authenticated in the cloud; the attendance machine generates a single-use dynamic encrypted QR code based on the unique device identifier, the current timestamp, the bound geographic point of interest identifier, the currently acceptable task node identifier, and a random number seed using the national cryptographic SM3 hash algorithm, with the conversion period configurable from 5 to 60 seconds; the attendance machine synchronizes the original encryption factor of each generated QR code to the cloud platform through an encrypted communication link; the original encryption factor automatically becomes invalid and cannot be reused after the QR code expires.

[0098] In this embodiment, the multi-factor verification process of the hardware strong verification mode includes: acquiring the check-in data uploaded by the user terminal, wherein the check-in data includes an encrypted string obtained by scanning a dynamically changing QR code check-in machine, a unique user identifier, a current task node identifier, a scanning timestamp, and synchronously collected real-time location data; a first verification step: cross-verifying the encrypted string with the original encryption factors pre-stored in the cloud and generated by the corresponding check-in machine within the scanning time window; if the verification does not match, a verification failure result is directly generated; a second verification step: verifying whether the geographic point of interest identifier bound to the check-in machine is consistent with the geographic point of interest identifier associated with the current task node; if they are inconsistent, a verification failure result is directly generated; a third verification step: based on the synchronously collected real-time location point stream, completing simulated positioning detection, movement speed verification, fence dwell time verification, pre-task completion status verification, and execution time window verification, and generating an authenticity score; based on the verification results of the above three steps, generating a final verification result.

[0099] As a specific example of an embodiment of the present invention, after a user arrives at "XX Coffee Academy," the system detects that the location has entered a 50-meter fence and prompts the user to check in. At this time, the user opens the APP to scan the dynamic QR code that is refreshed in real time on the screen of the check-in machine. The check-in machine generates a new QR code every 15 seconds. The QR code is generated using the national cryptographic SM3 algorithm based on the device ID, current timestamp, location POI identifier, currently available task node ID, and random number seed. The user terminal uploads the encrypted string obtained from the scan, the user ID, the task node ID, the scan timestamp, and the synchronously collected real-time location data to the cloud. The cloud platform performs a three-step verification: First step: compare the encrypted string with the original encryption factor generated by the check-in machine within the scanning time window and pre-stored in the cloud. If a match is found, the verification is successful. Second step: verify that the POI identifier bound to the check-in machine is consistent with the POI identifier associated with the current task node. The third step: Based on the real-time location data collected synchronously, complete the simulated location detection (passed), movement speed verification (passed), dwell time within the fence (the user has stayed in the store for 40 minutes, exceeding the minimum requirement of 30 minutes), completion status of the pre-requisite tasks (the basic theory course has been completed), and execution time window (falling within the scheduled 14:00-16:00), generating an authenticity score of 95 points.

[0100] In this embodiment, calculating the experience value of the current task node based on its attributes, verification results, and corresponding verification mode includes: when the verification result is passed, extracting the task type and difficulty coefficient of the current task node from the planning snapshot to determine the basic experience value; if the current verification mode is the mobile phone positioning basic verification mode, multiplying the basic experience value by the authenticity score to obtain the final experience value; if the current verification mode is the hardware strong verification mode, multiplying the basic experience value, the authenticity score, and the experience value bonus coefficient to obtain the bonus final experience value.

[0101] The system extracts a base experience value of 100 from the planning snapshot for the current task node. Under strong hardware verification mode, the final experience value = base experience value × authenticity score × experience value bonus coefficient = 100 × 0.95 × 1.5 = 142.5. The system adds 142.5 to the user's accumulated experience value ledger, the user level increases from Lv.2 to Lv.3, the task progress is updated to "Latte Art Practice Class" completed, and the next task node "Elementary Certification Exam" is unlocked.

[0102] In this embodiment, after a user completes the task check-in and passes verification through hardware strong verification mode, the cloud platform synchronously pushes the user's in-store data, task completion data, and user tag data to the merchant management terminal of the corresponding location, completing the traffic conversion loop. For example, the cloud platform automatically generates an in-store record: user ID (anonymized), in-store time, stay duration, completed task node "Latte Art Practice Class", and user skill tag "Coffee Making". This record is pushed to the merchant management terminal of "XX Coffee Academy" in real time, where merchants can view data such as the number of students visiting the store today, course completion rate, and user profiles for business analysis and targeted marketing (such as pushing advanced course coupons to the user), completing the traffic conversion loop.

[0103] Example 3 Please refer to Figure 2 , Figure 2 A schematic diagram of a geolocation-based skill task management device provided in an embodiment of the present invention includes: a data acquisition module 201, a task map construction module 202, a planning snapshot generation module 203, a verification module 204, an experience calculation module 205, and a progress update module 206; The data acquisition module 201 is used to acquire structured target data of user skill tasks; The task graph construction module 202 is used to generate a number of task nodes and sequential dependencies between each task node based on the structured target data, and to construct a task graph based on the number of task nodes and sequential dependencies between each task node. The planning snapshot generation module 203 is used to associate at least one task node in the task map with a preset geographic point of interest and preset execution constraints to obtain a target task map, and generate a planning snapshot based on the target task map. The verification module 204 is used to obtain the real-time location point stream and current task progress uploaded by the user, determine the current task node based on the current task progress and the planning snapshot, and perform multi-factor verification on the matching degree between the real-time location point stream and the current task node based on the planning snapshot to obtain the verification result. The experience calculation module 205 is used to calculate the experience value of the current task node based on the attributes of the current task node and the verification result, and update the user's accumulated experience value ledger and user level based on the experience value. The progress update module 206 is used to update the task progress of the user skill task based on the updated accumulated experience value ledger and user level.

[0104] In this embodiment, the task graph construction module is used to generate a plurality of task nodes and sequential dependencies between each task node based on the structured target data, and to construct a task graph based on the plurality of task nodes and sequential dependencies between each task node, including: Skill tags, target levels, and time periods are extracted based on the structured target data; The skill tags are matched with a preset skill knowledge base to obtain the corresponding skill development path; Based on the target level and the time period, the skill development path is divided into several task stages; One or more task nodes are generated for each task stage based on a preset task template library. Based on the order of the various task stages, establish sequential dependencies between the task nodes, and connect the task nodes according to the sequential dependencies to generate the task graph.

[0105] In this embodiment, the planning snapshot generation module is used to associate at least one task node in the task map with a preset geographic point of interest and preset execution constraints to obtain a target task map, and to generate a planning snapshot based on the target task map, including: Obtain the task nodes to be associated in the task graph, and determine the task type and geographical constraint range of the task nodes to be associated. Based on the task type and geographical constraints, several candidate geographic points of interest are matched in a preset geographic point of interest database. The matching degree between the task node to be associated and each candidate geographic point of interest is calculated based on a preset matching degree algorithm, and the candidate geographic point of interest with the highest matching degree is associated with the task node to be associated. The target task graph is obtained by associating the nodes with the preset execution constraints, and a planning snapshot is generated based on the target task graph.

[0106] In this embodiment, the planning snapshot generation module is used to associate the task nodes to be associated with the preset execution constraints to obtain a target task graph, and generate a planning snapshot based on the target task graph, including: The task node to be associated is bound based on execution constraints, which include the expected execution time window, minimum dwell time, and fence radius. The task map is updated based on the associated geographic points of interest and execution constraints to generate the target task map; The target task graph is serialized to obtain task sequence data; The hash digest of the task sequence data is calculated based on a preset hash algorithm, and an immutable planning snapshot is generated based on the hash digest.

[0107] In this embodiment, the verification module is used to acquire the real-time location point stream uploaded by the user and the current task progress, and to determine the current task node based on the current task progress and the planning snapshot, including: Get the real-time location data stream uploaded by the user and the current task progress; Extract the sequence of task nodes and the sequential dependencies between each task node from the planning snapshot; Based on the current task progress, determine the set of completed task nodes, and based on the sequential dependency relationship and the set of completed task nodes, determine the task nodes to be executed; The first fence range of the geographic points of interest associated with the task node to be executed is obtained based on the planning snapshot; when the location coordinates in the real-time location point stream are detected to enter the first fence range, the task node to be executed is determined as the current task node.

[0108] In this embodiment, the verification module is used to perform multi-factor verification on the matching degree between the real-time location point stream and the current task node based on the planning snapshot, and obtain the verification result, including: Extract the geographic points of interest data associated with the current task node and the sequential position of the current task node in the task node sequence from the planning snapshot; Based on the real-time location point stream, the user's location coordinates, timestamps, positioning accuracy, and simulated positioning markers at each time point are obtained; Based on the geographic point of interest data and the sequential location, multi-factor verification is performed on the location coordinates, timestamp, positioning accuracy, and simulated positioning marker to obtain the verification results.

[0109] In this embodiment, the geographic point of interest (POI) data includes target location coordinates, target fence radius, and target minimum dwell time; the verification module is used to perform multi-factor verification on the location coordinates, timestamp, positioning accuracy, and simulated positioning marker based on the POI data and the sequential location, and obtain verification results, including: A simulated positioning test is performed based on the simulated positioning marker. If the simulated positioning marker is detected as true, the simulated positioning test result is "fail". The moving speed is calculated based on the distance and time difference between adjacent coordinates in the real-time location point stream. If the moving speed exceeds a preset speed threshold, a speed anomaly marker is generated. Based on the sequential position of the current task node and the sequential dependencies in the planning snapshot, check whether the current task node has completed all the preceding task nodes. If not, generate an order anomaly flag. Based on the location coordinates in the real-time location point stream and the target location coordinates, the continuous dwell time of the user within the radius of the target fence is calculated. If the continuous dwell time is lower than the target minimum dwell time, an insufficient dwell time marker is generated. Based on the expected execution time window of the current task node and the timestamp of the real-time location point stream, if the user's actual execution time does not fall within the expected execution time window, a time deviation marker is generated. Based on the simulated positioning detection results, velocity anomaly markers, sequence anomaly markers, insufficient dwell time markers, and time deviation markers, the verification results are generated, and the verification results include an authenticity score.

[0110] In this embodiment, the experience calculation module is used to calculate the experience value of the current task node based on the attributes of the current task node and the verification result, and to update the user's accumulated experience value ledger and user level based on the experience value, including: When the verification result indicates that the current task node passes the multi-factor verification, the task type, difficulty coefficient, and location level of the geographic point of interest of the current task node are extracted from the planning snapshot. A base experience value is determined based on the task type, and the base experience value is weighted according to the difficulty coefficient to obtain a first experience value. The location coefficient is determined based on the location level, and a second experience value is calculated based on the preset planning bonus coefficient, the location coefficient, and the first experience value. The target experience value is obtained by multiplying the second experience value by the authenticity score; and the user's accumulated experience value ledger and user level are updated based on the target experience value.

[0111] The present invention provides a computer program product, including a computer program or instructions, which, when executed by a communication device, implement the method described therein.

[0112] In this embodiment of the invention, a processing device is also provided, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the above-described location-based skill task management method.

[0113] In this embodiment of the invention, a computer-readable storage medium is also provided, which includes a stored computer program, wherein the computer program controls the device where the computer-readable storage medium is located to execute the above-described location-based skill task management method when it is running.

[0114] For example, a computer program can be divided into one or more modules, one or more of which are stored in memory and executed by a processor to perform the present invention. The one or more modules can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in a processing device.

[0115] The processing device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The processing device may include, but is not limited to, a processor, memory, and a display. Those skilled in the art will understand that the above components are merely examples of the processing device and do not constitute a limitation on the processing device. It may include more or fewer components than the specified components, or a combination of certain components, or different components. For example, the processing device may also include input / output devices, network access devices, buses, etc.

[0116] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the processing device, connecting all parts of the processing device through various interfaces and lines.

[0117] Memory can be used to store computer programs and / or modules. The processor performs various functions of the processing device by running or executing the computer programs and / or modules stored in the memory, and by accessing data stored in the memory. Memory can mainly include a program storage area and a data storage area. The program storage area can store the operating system, at least one application program required for a function (such as sound playback, text conversion, etc.), etc.; the data storage area can store data created based on the use of the mobile phone (such as audio data, text message data, etc.). In addition, memory can include high-speed random access memory, and can also include non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital cards (SD cards), flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0118] The module for managing location-based skills tasks, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. Those skilled in the art can understand and implement this without any inventive effort.

[0119] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A location-based skills task management method, characterized in that, include: Obtain user input information, parse the user input information based on a preset large language model, and obtain structured target data; Obtain user location information, generate several task nodes and sequential dependencies between each task node based on the structured target data and user location information, and construct a task graph based on the several task nodes and sequential dependencies between each task node; Associate at least one task node in the task graph with a preset geographic point of interest and preset execution constraints to obtain a target task graph, and generate a planning snapshot based on the target task graph; Obtain the real-time location point stream uploaded by the user and the current task progress, and determine the current task node based on the current task progress and the planning snapshot; Based on the planning snapshot, a multi-factor verification is performed on the matching degree between the real-time location point stream and the current task node to obtain the verification result; The experience value of the current task node is calculated based on the attributes of the current task node and the verification result, and the user's accumulated experience value ledger and user level are updated based on the experience value. The task progress of the user's skill tasks is updated based on the updated accumulated experience value ledger and user level.

2. The location-based skill task management method as described in claim 1, characterized in that, The process of generating several task nodes and their sequential dependencies based on the structured target data and user location information, and constructing a task graph based on these dependencies, includes: Skill tags, target levels, and time periods are extracted based on the structured target data; The skill tags are matched with a preset skill knowledge base to obtain the corresponding skill development path; Based on the target level and the time period, the skill development path is divided into several task stages; the geographical constraint range is determined based on the user's location information, and one or more task nodes are generated for each task stage based on the geographical constraint range and a preset task template library; Based on the order of the various task stages, establish sequential dependencies between the task nodes, and connect the task nodes according to the sequential dependencies to generate the task graph.

3. The location-based skill task management method as described in claim 2, characterized in that, The step of associating at least one task node in the task map with a preset geographic point of interest and preset execution constraints to obtain a target task map, and generating a planning snapshot based on the target task map, includes: Obtain the task nodes to be associated in the task graph, and determine the task type and geographical constraint range of the task nodes to be associated. Based on the task type and geographical constraints, several candidate geographic points of interest are matched in a preset geographic point of interest database. The matching degree between the task node to be associated and each candidate geographic point of interest is calculated based on a preset matching degree algorithm, and the candidate geographic point of interest with the highest matching degree is associated with the task node to be associated. The target task graph is obtained by associating the nodes with the preset execution constraints, and a planning snapshot is generated based on the target task graph.

4. The location-based skill task management method as described in claim 3, characterized in that, The step of associating the target task node with the preset execution constraints to obtain the target task graph, and generating a planning snapshot based on the target task graph, includes: The task node to be associated is bound based on execution constraints, which include the expected execution time window, minimum dwell time, and fence radius. The task map is updated based on the associated geographic points of interest and execution constraints to generate the target task map; The target task graph is serialized to obtain task sequence data; The hash digest of the task sequence data is calculated based on a preset hash algorithm, and an immutable planning snapshot is generated based on the hash digest.

5. The location-based skill task management method as described in claim 4, characterized in that, The step of obtaining the real-time location stream uploaded by the user and the current task progress, and determining the current task node based on the current task progress and the planning snapshot, includes: Get the real-time location data stream uploaded by the user and the current task progress; Extract the sequence of task nodes and the sequential dependencies between each task node from the planning snapshot; Based on the current task progress, determine the set of completed task nodes, and based on the sequential dependency relationship and the set of completed task nodes, determine the task nodes to be executed; The first fence range of the geographic points of interest associated with the task node to be executed is obtained based on the planning snapshot; when the location coordinates in the real-time location point stream are detected to enter the first fence range, the task node to be executed is determined as the current task node.

6. The location-based skill task management method as described in claim 5, characterized in that, The process of performing multi-factor verification on the matching degree between the real-time location point stream and the current task node based on the planning snapshot to obtain verification results includes: Extract the geographic points of interest data associated with the current task node and the sequential position of the current task node in the task node sequence from the planning snapshot; Based on the real-time location point stream, the user's location coordinates, timestamps, positioning accuracy, and simulated positioning markers at each time point are obtained; Based on the geographic point of interest data and the sequential location, multi-factor verification is performed on the location coordinates, timestamp, positioning accuracy, and simulated positioning marker to obtain the verification results.

7. The location-based skill task management method as described in claim 6, characterized in that, The geographic point of interest data includes the target location coordinates, the target fence radius, and the target minimum dwell time. The method involves performing multi-factor verification on the location coordinates, timestamp, positioning accuracy, and simulated positioning markers based on the geographic point of interest data and the sequential location, to obtain verification results, including: A simulated positioning test is performed based on the simulated positioning marker. If the simulated positioning marker is detected as true, the simulated positioning test result is "fail". The moving speed is calculated based on the distance and time difference between adjacent coordinates in the real-time location point stream. If the moving speed exceeds a preset speed threshold, a speed anomaly marker is generated. Based on the sequential position of the current task node and the sequential dependencies in the planning snapshot, check whether the current task node has completed all the preceding task nodes. If not, generate an order anomaly flag. Based on the location coordinates in the real-time location point stream and the target location coordinates, the continuous dwell time of the user within the radius of the target fence is calculated. If the continuous dwell time is lower than the target minimum dwell time, an insufficient dwell time marker is generated. Based on the expected execution time window of the current task node and the timestamp of the real-time location point stream, if the user's actual execution time does not fall within the expected execution time window, a time deviation marker is generated. Based on the simulated positioning detection results, velocity anomaly markers, sequence anomaly markers, insufficient dwell time markers, and time deviation markers, the verification results are generated, and the verification results include an authenticity score.

8. The location-based skill task management method as described in claim 7, characterized in that, The step of calculating the experience value of the current task node based on its attributes and the verification result, and updating the user's accumulated experience value ledger and user level based on the experience value, includes: When the verification result indicates that the current task node passes the multi-factor verification, the task type, difficulty coefficient, and location level of the geographic point of interest of the current task node are extracted from the planning snapshot. A base experience value is determined based on the task type, and the base experience value is weighted according to the difficulty coefficient to obtain a first experience value. The location coefficient is determined based on the location level, and a second experience value is calculated based on the preset planning bonus coefficient, the location coefficient, and the first experience value. The target experience value is obtained by multiplying the second experience value by the authenticity score; and the user's accumulated experience value ledger and user level are updated based on the target experience value.

9. A location-based skill task management device, characterized in that, include: The module includes a data acquisition module, a task graph construction module, a planning snapshot generation module, a verification module, an experience calculation module, and a progress update module. The data acquisition module is used to acquire user input information, parse the user input information based on a preset large language model, and obtain structured target data. The task graph construction module is used to generate several task nodes and sequential dependencies between each task node based on the structured target data, and to construct a task graph based on the several task nodes and sequential dependencies between each task node. The planning snapshot generation module is used to associate at least one task node in the task map with a preset geographic point of interest and preset execution constraints to obtain a target task map, and generate a planning snapshot based on the target task map. The verification module is used to obtain the real-time location point stream uploaded by the user and the current task progress, and determine the current task node based on the current task progress and the planning snapshot; Based on the planning snapshot, a multi-factor verification is performed on the matching degree between the real-time location point stream and the current task node to obtain the verification result; The experience calculation module is used to calculate the experience value of the current task node based on the attributes of the current task node and the verification result, and update the user's accumulated experience value ledger and user level based on the experience value. The progress update module is used to update the task progress of the user skill task based on the updated accumulated experience value ledger and user level.

10. A location-based skill task management device as described in claim 9, characterized in that, The task graph construction module is used to generate several task nodes and sequential dependencies between each task node based on the structured target data, and to construct a task graph based on the several task nodes and sequential dependencies between each task node, including: Skill tags, target levels, and time periods are extracted based on the structured target data; The skill tags are matched with a preset skill knowledge base to obtain the corresponding skill development path; Based on the target level and the time period, the skill development path is divided into several task stages; One or more task nodes are generated for each task stage based on a preset task template library. Based on the order of the various task stages, establish sequential dependencies between the task nodes, and connect the task nodes according to the sequential dependencies to generate the task graph.

Images