A VR-based socialized horticulture intervention and evaluation system

Abstract

This invention discloses a socialized horticulture intervention and assessment system based on VR technology, specifically relating to the field of auxiliary psychotherapy. The system includes: a label combination space construction module that establishes a three-dimensional coordinate space based on the patient's cognitive, emotional, and behavioral dimensions to form psychological label combination codes; a communication willingness assessment module that calculates the communication willingness score between two collaborating patients; a label graph construction module that constructs a label fit graph based on the communication willingness score to identify conflicting and complementary label relationships; a group reconstruction module that performs patient isolation and collaborative group reconstruction based on the spatial distribution of conflicting labels; a task performance assessment module that combines different collaborative structures and psychological label inputs to construct a task performance assessment model; and an intervention strategy configuration module that, based on the group structure and task node candidate allocation strategy, selects the optimal intervention strategy configuration scheme through the task response parameters output by the model, achieving optimal matching of task structures under individual label guidance.

Description

Technical Field

[0001] This invention relates to the field of assistive psychotherapy technology, and more specifically, to a socialized horticultural intervention and assessment system based on VR technology. Background Technology

[0002] In existing virtual social gardening intervention systems, users are typically grouped according to interests, age, frequency of operation, or random system strategies to participate in social gardening activities such as co-cultivating plants and collaborative landscaping. However, this grouping method overlooks a crucial fact: users participating in gardening interventions often have different types of psychological problems, such as anxiety, depression, avoidant personality, and social phobia. These problems exhibit significant differences in their psychological mechanisms, directly impacting individual behavior and intervention response during social collaboration. More seriously, existing systems fail to identify the potential "complementary" or "conflicting" relationships between these psychological differences, leading to problems such as cooperation barriers, psychological resistance, or ineffective intervention during actual collaboration. Simultaneously, the system lacks a continuous tracking and feedback mechanism for collaboration effectiveness, failing to dynamically optimize team structure based on recovery performance during collaboration, resulting in low group intervention efficiency and a poor user experience. Especially for users with severe psychological problems or high social sensitivity, inappropriate collaboration pairings may even exacerbate their psychological burden. Summary of the Invention

[0003] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a socialized horticulture intervention and evaluation system based on VR technology to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A VR-based socialized horticulture intervention and assessment system includes a tag combination space construction module, a communication willingness assessment module, a tag map construction module, a group reconstruction module, a task performance assessment module, and an intervention strategy configuration module, wherein:

[0006] The tag combination space construction module maps the psychological problems of patients into a three-dimensional psychological tag system composed of emotions, cognition and behavior, and constructs a psychological tag combination space;

[0007] The communication willingness assessment module monitors virtual gardening collaborative task records, collects patients' interactive behavior data in collaborative task scenarios based on a sliding window, and calculates communication willingness scores among patients with different psychological label combinations.

[0008] The label mapping module maps patients’ communication willingness scores to a psychological label combination space, constructs a label fit map, and distinguishes between conflicting and complementary labels.

[0009] The grouping reconstruction module isolates the individual patients corresponding to conflicting labels and reconstructs the virtual gardening collaboration grouping;

[0010] The task performance evaluation module extracts the task execution feedback data of individual patients corresponding to complementary labels under various horticultural collaborative task structures, and establishes a task performance evaluation model.

[0011] The intervention strategy configuration module constructs task execution strategies for patients within the virtual gardening collaboration group based on the output results of the task performance evaluation model, and outputs the corresponding intervention strategy configurations.

[0012] In a preferred embodiment, the label combination space construction module maps the psychological problems of the patient group into a three-dimensional psychological label system composed of emotions, cognition, and behavior. The construction of the psychological label combination space specifically includes:

[0013] Obtain psychological problem classification records from the medical records of the patient group, and convert each patient's psychological problem classification record into cognitive score, emotional score and behavioral score;

[0014] For each patient, a single-point psychological vector corresponding to the combination of emotional, cognitive, and behavioral labels is established. The label items corresponding to all single-point psychological vectors are embedded into a three-dimensional spatial structure to establish a psychological label combination space.

[0015] In a preferred embodiment, embedding the label item corresponding to the single-point psychological vector into a three-dimensional spatial structure to establish a psychological label combination space specifically includes:

[0016] Construct a three-dimensional coordinate system with cognitive, emotional, and behavioral dimensions as axes, embed the three types of labels in the single-point psychological vector into the corresponding axes and fix the polarity direction;

[0017] Perform a three-axis mapping operation on the single-point psychological vectors of all patients to generate the coordinates of the single-point psychological vectors of the patients in a three-dimensional coordinate system;

[0018] Based on the positive and negative combinations of three-axis coordinates, the three-dimensional psychological space is divided into eight non-overlapping quadrant regions, and each quadrant is numbered.

[0019] Obtain the quadrant number to which the patient's coordinates belong and mark it as the encoding value of the patient's psychological label combination.

[0020] In a preferred embodiment, the communication willingness assessment module monitors virtual gardening collaborative task records, collects patient interaction behavior data in collaborative task scenarios based on a sliding window, and calculates communication willingness scores among patients with different psychological label combinations, specifically including:

[0021] During the execution of a virtual gardening task, interactive task nodes for two-person collaboration are identified, and the sequence of interactive behaviors is extracted based on the timestamps of the patient's initiation and response to interactions within the task nodes.

[0022] Define an equal-width sliding time window, extract the patient's interaction behavior fragments window by window in the interaction behavior sequence, and count the cumulative number of initiated and responded interactions;

[0023] Within each sliding time window, the cumulative length of delayed response behavior and the proportion of asynchronous response behavior among patients are counted to establish an interactive response breakage index.

[0024] Construct a communication willingness scoring function based on weighted integration, and generate a communication willingness score based on the cumulative number of interactions and the interaction response break index.

[0025] In a preferred embodiment, the label mapping module maps the patient's communication willingness score to a psychological label combination space, constructs a label fit map, and distinguishes between conflicting and complementary labels, specifically including:

[0026] Obtain the communication willingness scores of patients in two-person collaborative interactions at each interactive task node, and extract the encoded values ​​of the two patients in the psychological label combination space;

[0027] Based on the coded values, the communication willingness scores are grouped according to quadrant regions. The mean of the communication willingness scores within each quadrant region group is calculated, and a psychological label group suitability score matrix is ​​constructed.

[0028] Using the psychological label combinations in the fit rating matrix as nodes and the relationships between psychological label combinations as connecting edges, a psychological label fit graph structure is established, and fit ratings are labeled on the connecting edges.

[0029] Set threshold ranges for determining conflict and complementarity, and classify different combinations of psychological labels into conflict labels and complementary labels based on the threshold range in which the fit score falls.

[0030] In a preferred embodiment, the group reconstruction module isolates patient individuals corresponding to conflicting tags and reconstructs virtual gardening collaboration groups, specifically including:

[0031] Retrieve conflicting label pairs already identified in the psychological label fit map and locate the corresponding patient psychological label combination code;

[0032] Extract patient individuals with conflicting label combination codes from the current collaboration group, and remove the corresponding patient individuals' task participation history and interaction trajectory;

[0033] Based on the distribution of conflict labels in each collaborative group, a conflict-minimizing sort is performed to determine the priority sequence of individual patients to be isolated.

[0034] Prioritize the isolation of patients and remove them from the original collaborative groups to create a new list of virtual horticulture collaborative group combinations.

[0035] In a preferred embodiment, the task performance evaluation module extracts task execution feedback data of individual patients corresponding to complementary labels under various horticultural collaborative task structures, and establishes a task performance evaluation model, specifically including:

[0036] The complete operation chain of the completed historical virtual gardening collaborative tasks is broken down into process nodes according to task stages and interaction types, and a set of numbered task structure units is constructed.

[0037] In the psychological label combination space, extract the single-point psychological vector composed of the cognitive score, emotional score and behavioral score of each patient in each reconstructed virtual gardening collaboration group;

[0038] The system collects feedback data from the execution logs of each task node within the task structure unit set, obtains the executor of the task node, and integrates the single-point psychological vector of the executor with the response parameters of the task node into a task behavior sample set.

[0039] The response parameters of the task nodes are obtained from the feedback data in the execution log, including the task progress rate, operation accuracy, and collaboration synchronization rate.

[0040] The task behavior sample set is used as the training sample set, and the patient's single-point psychological vector and task node number are used as input items, while the response parameters of the task node are used as output items to train the task performance evaluation model.

[0041] In a preferred embodiment, the intervention strategy configuration module constructs task strategies for patients within the virtual horticulture collaboration group based on the output results of the task performance evaluation model, and the output of the corresponding intervention strategy configuration specifically includes:

[0042] Regularly update the medical records of the patient group. When the number of patients increases or the single-point psychological vector coordinates of the psychological label combination space show limit changes, update the label adaptability map and reconstruct the virtual gardening collaboration group.

[0043] In each virtual gardening collaboration group, a task node allocation strategy table for all patients performing tasks is pre-generated, and the patient's single-point psychological vector and task node allocation strategy are input into the task performance evaluation model.

[0044] The output of the task performance evaluation model is normalized and converted into a task performance score through weighted fusion. The task node allocation strategy corresponding to the highest task performance score is selected as the intervention strategy configuration.

[0045] The technical effects and advantages of the VR-based socialized horticulture intervention and evaluation system of this invention are as follows:

[0046] By constructing a three-dimensional psychological label combination space composed of emotional, cognitive, and behavioral labels, a structured model of patients' psychological states is achieved, enhancing the accuracy of subsequent individual difference identification. By introducing an interactive behavior data collection mechanism into horticultural collaborative tasks, combined with a sliding window method for dynamic assessment, this approach effectively improves the objectivity and timeliness of communication willingness evaluation without relying on traditional questionnaires. The constructed label fit map can accurately distinguish between conflicting and complementary labels, providing a stable data foundation for collaborative group optimization. Isolation and grouping of patients with conflicting labels avoids behavioral disorders caused by the superposition of psychological conflicts during collaboration. Furthermore, an evaluation model built using task performance feedback data makes task execution results quantifiable and comparable, improving the targeting and interpretability of intervention strategy generation. Finally, the model output is applied to task strategy allocation, effectively constructing a correspondence between individual labels and task structures, thereby achieving optimal configuration of collaborative task matching among patients and enhancing the overall flexibility of the intervention. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the structure of a socialized horticulture intervention and evaluation system based on VR technology according to the present invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0049] Example 1

[0050] Figure 1 This invention presents a socialized horticulture intervention and evaluation system based on VR technology, comprising a tag combination space construction module, a communication willingness assessment module, a tag map construction module, a group reconstruction module, a task performance assessment module, and an intervention strategy configuration module, wherein:

[0051] The tag combination space construction module maps the psychological problems of patients into a three-dimensional psychological tag system composed of emotions, cognition and behavior, and constructs a psychological tag combination space;

[0052] The communication willingness assessment module monitors virtual gardening collaborative task records, collects patients' interactive behavior data in collaborative task scenarios based on a sliding window, and calculates communication willingness scores among patients with different psychological label combinations.

[0053] The label mapping module maps patients’ communication willingness scores to a psychological label combination space, constructs a label fit map, and distinguishes between conflicting and complementary labels.

[0054] The grouping reconstruction module isolates the individual patients corresponding to conflicting labels and reconstructs the virtual gardening collaboration grouping;

[0055] The task performance evaluation module extracts the task execution feedback data of individual patients corresponding to complementary labels under various horticultural collaborative task structures, and establishes a task performance evaluation model.

[0056] The intervention strategy configuration module constructs task execution strategies for patients within the virtual gardening collaboration group based on the output results of the task performance evaluation model, and outputs the corresponding intervention strategy configurations.

[0057] The label combination space construction module maps the psychological problems of the patient group into a three-dimensional psychological label system composed of emotions, cognition and behavior, and constructs a psychological label combination space.

[0058] The system retrieves historical psychological problem classification records from the patient's existing medical records. These records originate from disease tags generated by psychiatric clinical diagnoses, rehabilitation plan tracking forms, and auxiliary psychological assessment systems. The recorded information covers the categories of psychological problems that have been explicitly identified or observed over a long period, such as anxiety disorders, depressive disorders, impulse control disorders, cognitive impairments, avoidant personality disorder, and social impairments. To achieve unified calculation and assessment, this step introduces a structured transformation rule base, establishing qualitative mapping relationships between various psychological problem tags and three psychological dimensions—cognition, emotion, and behavior. This transformation rule base is set by mental health professionals and models the common impacts of various common psychological problems on clinical manifestations and behavioral intervention responses. For example, anxiety disorders are identified as significantly affecting the emotional dimension, while also exhibiting withdrawal behavior in the behavioral dimension; impulse control disorders primarily affect the behavioral dimension, and may also cause decision-making delays in the cognitive dimension; cognitive impairments are directly mapped to the cognitive dimension, while potentially having a weak impact on the emotional dimension. For each psychological problem label, the system sets quantitative scores across three dimensions, ranging from -5 to +5. Positive values ​​indicate an enhancing or activating effect on the dimension, negative values ​​indicate an inhibiting or weakening effect, and 0 indicates no significant effect. For example, anxiety disorder typically transforms into an emotional dimension of +4, a behavioral dimension of -2, and a cognitive dimension of +1. After completing the dimensional transformation for all patient psychological problem labels, the mapped values ​​of each label are summed within each dimension to generate a three-dimensional psychological value combination for each patient. This combination is expressed as a standard vector consisting of cognitive, emotional, and behavioral scores, serving as input identifiers for patients in subsequent task structure matching and grouping analysis. To prevent extreme data distribution, the system sets amplitude limits after each dimensional transformation; dimensional values ​​exceeding ±5 are truncated to ±5, ensuring all data fall within a reasonable statistical range.

[0059] Using a three-dimensional psychological score vector as the basic data input, a three-dimensional spatial coordinate system is constructed with cognitive, emotional, and behavioral dimensions as axes. The spatial structure of the coordinate system adopts a Cartesian coordinate configuration, where the X-axis corresponds to the cognitive dimension, the Y-axis to the emotional dimension, and the Z-axis to the behavioral dimension. Each coordinate axis is set with its origin at 0. Positive and negative directions represent the positive or negative effects on that psychological dimension, specifically defined as follows: a positive direction for the cognitive dimension represents cognitive activity and clear logic, while a negative direction represents cognitive stagnation and difficulty in judgment; a positive direction for the emotional dimension represents a state of heightened or stable emotions, while a negative direction represents a tendency towards depression and anxiety; a positive direction for the behavioral dimension represents outward behavior and positive expression, while a negative direction represents withdrawal and indifference. All patients' three-dimensional psychological score vectors are mapped to this three-dimensional coordinate system in a single-point manner. To ensure comparability between values ​​of different dimensions, the system first performs linear normalization on the original dimension values, uniformly mapping the range of -5 to +5 to a standardized interval of -1 to +1, ensuring that each dimension has an equal weight in its influence on the coordinate projection. For example, a patient's original psychological score is cognition +3, emotion -2, and behavior +1. After normalization, the scores are cognition +0.6, emotion -0.4, and behavior +0.2, respectively. The coordinates of this single-point psychological vector in the three-dimensional psychological space are (0.6, -0.4, 0.2).

[0060] The space is divided into eight non-overlapping quadrants. The division is determined by the combination of positive and negative values ​​for each coordinate axis. For example, if the cognitive value is positive, the emotional value is negative, and the behavioral value is positive, then the coordinate point belongs to the "positive-negative-positive" quadrant. To unify the numbering system, this embodiment uses a three-bit binary numbering method, defining positive as 1 and negative as 0. Each dimension's positive-negative combination is converted into three binary bits, forming eight categories from 000 to 111. For example, "negative-negative-positive" corresponds to number 010, i.e., quadrant 2; "positive-positive-positive" corresponds to number 111, i.e., quadrant 8. This quadrant number becomes the psychological label combination code value for each patient. The system binds each patient's psychological vector coordinate point with its corresponding quadrant number to generate a coding index and establishes a psychological label combination space, including the patient identification code, three-dimensional coordinates, quadrant number, and label combination code.

[0061] The communication willingness assessment module monitors virtual gardening collaborative task records, collects patients' interactive behavior data in collaborative task scenarios based on a sliding window, and calculates communication willingness scores among patients with different psychological label combinations.

[0062] The system monitors collaborative tasks within a virtual gardening system. The platform offers various pre-defined task structures, some of which include clearly defined two-person collaborative sub-tasks, such as alternating planting, synchronized pruning, and reciprocal transport. These task nodes have identifiable collaborative behavior entry points, which the system defines as two-person interactive task nodes. The identification criteria are: two participants must interact with each other in time, and this interaction must have temporal dependencies and a completion interlocking mechanism within the task module. The system retrieves operational log data from the virtual gardening platform to extract behavioral records during the collaborative task. This information includes the operation object number, execution timestamp, behavior type label (e.g., tool call, planting confirmation, task progress), and operator number. Within the identified two-person interactive task nodes, the system selects the first interactive action initiated by each patient within the task node as an "initiated interaction" event and marks the first valid feedback action of the responding party as a "response interaction" event. A sequence of interactive behaviors is defined as a chain of initiation-response events between a pair of patients within the same task node. The system assembles all initiation-response behavior pairs into an interaction behavior sequence according to timestamp order, and assigns a patient pair number and task node identifier to each sequence.

[0063] A sliding time window mechanism is introduced on the constructed interaction behavior sequence. The sliding time window is defined based on the average cycle of the task rhythm. In this embodiment, referring to the median operation time of three common collaborative tasks, the length of a single time window is set to 30 seconds, and the window sliding step is set to 15 seconds. That is, a time period of 30 seconds is extracted every 15 seconds in the sequence to form a continuous sliding window queue of equal width. This setting can cover the behavior distribution characteristics throughout the task process, while taking into account both computational efficiency and data integrity. For each sliding time window, the interaction behavior segment extraction operation is performed to retrieve the initiator-response interaction record group completed within the time period, and the initiator and responder of the interaction are marked respectively. Within each time window, the system counts the number of interaction initiations and responses completed by the patient in the current window, and records the time interval between interaction pairs. For behavior pairs with response delays exceeding the task-set threshold (e.g., the response should be completed within 3 seconds under normal task rhythm), the delay in seconds is recorded as the length of the delayed response behavior. If an interaction behavior does not receive a response within the same time window, the behavior is marked as an asynchronous response behavior.

[0064] Within a sliding time window, the following indicators are calculated for each patient: first, the cumulative number of completed interactions (including the total number of initiations and responses); second, the cumulative duration of delayed responses; and third, the proportion of asynchronous responses among all interactions. For example, within a certain time window, patient A and patient B completed 8 interactions, of which 3 were delayed responses, accumulating a delay of 9 seconds, and 2 responses were in the wrong order. Therefore, the delay duration within that window can be calculated as 9 seconds, and the proportion of asynchronous responses is 25%.

[0065] After completing the statistical analysis of behavioral indicators at the sliding time window level, an interaction response breakage index is constructed to characterize the quality of interaction, and a communication willingness scoring function is further developed. The interaction response breakage index uses two statistics as core inputs: the cumulative duration of delayed response behavior and the proportion of asynchronous response behavior. To unify the weighting of the two dimensions, the system normalizes both values ​​separately, setting the maximum delay duration to 15 seconds and the maximum asynchronous response proportion to 1.0. After normalization, a 9-second delayed response becomes 0.6, while the asynchronous proportion remains unchanged at 0.25. The interaction response breakage index is defined as the weighted sum of these two normalized values. To prevent a single abnormal event from excessively amplifying the breakage, a response elasticity factor is introduced to adjust the asynchronous response item exponentially, setting its exponent to 1.2. The final breakage index is then expressed as the weighted sum of the two values. The weighting coefficients can be specifically set according to the task type and the patient's label quadrant combination; for example, asynchronous behavior is given higher weight in two-person collaborative tasks. A communication willingness scoring function was further constructed. This function uses the cumulative number of interactions as a positive factor for the degree of proactive communication and the interaction response breakage index as a negative factor. The overall function structure is: Communication Willingness Score = Baseline Activity Score × (1 - Response Breakage Factor). The baseline activity score is determined by the cumulative number of interactions per unit time, with a maximum value of 10. The breakage factor has a maximum value not exceeding 1. Taking the aforementioned patient pair as an example, if the number of interactions within a certain time window is 8 (activity score is 8 / 10) and the response breakage is 0.62, then the communication willingness score is 8 × (1 - 0.62) = 3.04.

[0066] The label mapping module maps patients’ communication willingness scores to a psychological label combination space, constructs a label fit map, and distinguishes between conflicting labels and complementary labels.

[0067] The encoded values ​​of the two patients in the collaborative pair in the psychological label combination space are extracted. This label combination space is constructed based on a three-dimensional label system, comprising cognitive, emotional, and behavioral dimensions. In this three-axis coordinate system, the space is divided into eight non-overlapping quadrants according to the positive or negative attributes of the scores. Each patient's psychological label combination can be uniquely mapped to an encoded value in a specific quadrant. The communication willingness scores of each collaborative patient pair are categorized according to their corresponding two sets of encoded value combinations (e.g., 001-010, 001-111, etc.). That is, all scores with label combinations of 001-010 are grouped into one group, scores of 001-111 into another group, and so on, forming a score grouping table between label combinations. Statistical analysis is performed on the score data of each group, and the average communication willingness score between the label combination pairs is calculated as the suitability index of the combination. If multiple score records exist for a certain combination, the arithmetic mean of all scores should be used. Outliers can be removed using a simple quantile filtering method. In this way, a psychological label combination suitability score matrix is ​​constructed. Each entry in the matrix represents the average fit score between a set of mental label coding combinations.

[0068] The resulting matrix uses the initiator's label combination as the behavior and the responder's label combination as the column. Each intersection item in the two-dimensional matrix represents the suitability score of the corresponding label combination. For example, the second row and third column of the matrix (corresponding to the binary identifier and subtracting 1) might represent an average communication willingness score of 0.61 for the 001-010 combination, reflecting the degree of label fit between patients coded with 001 and those coded with 010.

[0069] Assuming actual collaborative behavior exists between psychological label combinations, connecting edges are constructed for all combinations whose rating values ​​appear in the rating matrix. Each edge connects a pair of different label encoding nodes, and its corresponding rating value is embedded in the graph structure as the edge weight information. The directionality of the edges can be omitted, i.e., it is considered an undirected graph, as the suitability of the label combinations can be considered symmetrical. A threshold range for suitability rating is set, and the rating values ​​are normalized to the 0-1 range. Conflict thresholds and complementarity thresholds are set to 0.4 and 0.7, respectively; that is, ratings below 0.4 are conflicting label combinations, ratings above 0.7 are complementary label combinations, and those in between are neutral combinations. The thresholds are derived from prior statistical experience, such as statistically analyzing the 25th and 75th percentiles of all rating distributions, or manually preset based on the distribution of patients' psychological problems. The rating value on each connecting edge is compared with the above threshold ranges, and the edges are classified into conflicting, complementary, or neutral edges according to the interval, and marked with different symbols or colors for subsequent analysis and processing. In the graph, label pairs pointed to by conflicting edges will be marked as unsuitable for co-op, while label pairs pointed to by complementary edges will be marked as cooperative recommendations.

[0070] The group reconstruction module isolates the individual patients corresponding to the conflicting labels and reconstructs the virtual gardening collaboration group.

[0071] In the constructed psychological label adaptation graph, the system first traverses all connecting edges, retrieving label combination edges identified as conflict relationships. Each edge identified as a conflict relationship has a unique bidirectional psychological label combination code. Based on the node codes of the conflict label pairs, the system sequentially matches the psychological label combination codes of all enrolled patient individuals in the current virtual gardening collaboration groups, and filters out group records that simultaneously contain conflict combination codes. The system then locates the patient individual ID corresponding to the aforementioned conflict combination. To avoid the intervention analysis being influenced by historical behavior, the system retrieves the task execution records, two-person collaboration node interaction data, and collaborative operation logs of this type of patient in the current and previous gardening task cycles, and performs a clearing operation on their interaction behavior trajectory and task participation history. This process is completed by a unified trajectory and record stripping engine to prevent the participation history of conflicting patients from interfering with the subsequent strategy training model.

[0072] Based on the confirmed conflicting patients, to avoid unnecessary large-scale group restructuring, the system further calculates the conflict label distribution density. Here, conflict distribution density refers to the coverage breadth and node clustering degree of a specific conflict label pair in the current grouping system. Using grouping units as the granularity, the frequency of conflict label pairs appearing together in each collaborative group is evaluated, and the frequency is sorted in descending order as a reference for isolation priority. Adopting the conflict minimization principle, a priority isolation ranking queue is constructed based on the conflict distribution density and the task participation ratio in the collaborative group. This ranking queue prioritizes removing individuals who exhibit conflicting interactions in multiple collaborative tasks and have relatively low individual task participation, in order to reduce disruption to the task chain after group splitting. Task participation is calculated using the product of the number of task nodes completed by the patient in the previous cycle and the percentage of task contribution as a reference indicator; the numerical scoring result is provided by the task performance evaluation model.

[0073] After identifying priority isolation targets, these patients are systematically removed from their original collaborative groups and marked as unassigned. Next, the task is restructured based on complementary label groups that do not yet have conflicting relationships in the current system. Unassigned patients are matched with available psychological label combinations (if no complementary label patient is matched, the patient is placed in a separate group). Referring to the operation records and interaction rhythm within the original task groups, several new collaborative group combination lists are generated.

[0074] The task performance evaluation module extracts the task execution feedback data of individual patients corresponding to complementary labels under various horticultural collaborative task structures, and establishes a task performance evaluation model.

[0075] The data from completed virtual gardening collaborative tasks is structured. This task data originates from operation logs and interaction records within the system, covering multiple stages such as task initiation, execution, interaction, and completion. To ensure the task structure is indexable and assignable, the entire operation chain is broken down into task stages (such as planting preparation, plant configuration, maintenance interaction, and result confirmation) based on the organizational logic of the task content. Furthermore, task nodes are categorized into "independent execution nodes" and "collaborative interaction nodes" based on whether they have explicit interaction requirements. Each separated node is assigned a unique number, forming a standardized "task structure unit set."

[0076] After numbering the task structure units, the execution feedback data for each task node is extracted from the task logs, focusing on the following three types of parameters: First, task progress rate, which is the time record from task start to task node completion, obtainable through the difference in timestamps in the task node logs; second, operation accuracy, which is the system's statistical analysis of the error rate or deviation ratio in each operation, based on a comparison between the operation record and the expected operation; third, collaboration synchronization rate, applicable only to collaborative interaction nodes, defined as the ratio of the average time difference between each participant completing their assigned task to the total time of the collaborative node, used to measure the degree of synchronization and cooperation. Each feedback data item must be bound to its executor identity, thereby combining the patient's single-point psychological vector with the corresponding task node number and feedback parameters within the same data unit to form a structured "task behavior sample". To ensure sample diversity and representativeness, at least 80% of the execution records of all historical node types should be collected, and a sample set should be constructed based on the task node number. All sample data must undergo data integrity verification, removing records with missing feedback items or incomplete patient labels to ensure a consistent and accurate data foundation for model training.

[0077] The completed task behavior sample set is input into the task performance evaluation model training process. The model is designed as a multi-input multi-output mapping model. Its input consists of the patient's single-point psychological vector three-dimensional coordinates (cognition, emotion, behavior) and task node numbers, which are jointly encoded using embedding vectors. The output consists of three response parameters: task progress rate, operational accuracy, and collaborative synchronization rate. Model training can be implemented using a multilayer perceptron network (MLP) or a gradient boosting tree (GBDT) structure, depending on the sample size, dimensionality, and degree of nonlinearity. Five-fold cross-validation should be performed during model training, and an early stopping strategy should be implemented to prevent overfitting. For parameter settings, the initial learning rate can be set to 0.01, the maximum number of training epochs is 200, and training is terminated if the validation set loss does not decrease for 10 consecutive epochs. In model accuracy validation, target error thresholds should be set for each response parameter, for example, task progress rate error no higher than ±10%, operational accuracy error no higher than ±5%, and collaborative synchronization rate deviation no higher than ±0.1. After training, the model is saved in a deployable form.

[0078] The intervention strategy configuration module constructs task strategies for patients within the virtual horticulture collaboration group based on the output results of the task performance evaluation model, and outputs the corresponding intervention strategy configuration.

[0079] The label space and label map are updated based on the dynamic changes in the medical records of patient groups. A patient record update monitoring system is established in the system database. Registered medical records are periodically traversed, the number of newly added patients is counted, and the cognitive, emotional, and behavioral scores of existing patients are resampled and compared. If any patient's score dimension changes beyond a set drift threshold (e.g., emotional score changes by more than ±1.5 points), it is determined that their single-point psychological vector coordinates have shifted. Subsequently, the updated single-point psychological vectors of all patients are remapped into the three-dimensional psychological label combination space. The symbols of each vector on the three axes constitute a combination logic (positive and negative polarity combination), and the region is divided according to a predetermined eight-quadrant numbering system, for example, positive-negative-positive corresponds to the 3rd quadrant. When a patient's coordinate change causes a shift in their quadrant, it is marked as a "label quadrant change," simultaneously triggering a local reconstruction of the label adaptation map. During graph reconstruction, based on the defined communication willingness rating mapping logic, a new graph structure is constructed with psychological label combinations as nodes and suitability ratings as connection edge weights, and the judgment boundaries of conflict labels and complementary labels are refreshed.

[0080] After updating the collaborative grouping map, a pre-generation operation is performed on the task allocation strategy for the groups to which complementary label combinations belong in the current label fit map. Specifically, the method involves: first, statistically analyzing each task node in the historical task structure unit set to determine the total number of task nodes for the horticulture task in the current period (e.g., 15 nodes); then, combining this with the number of patients in the current group (e.g., 5), generating different task node allocation schemes through permutations and combinations. For each patient-task node matching combination, the mapping relationship between task number and executor number is recorded, and strategy table entries are generated. Subsequently, the task node allocation strategy for each group and the standardized psychological vectors of each patient within the group are used as joint inputs and uniformly input into the task performance evaluation model. This model, trained with historical execution feedback data in the previous steps, has the ability to predict the patient's completion speed, accuracy, and collaborative performance in specific task nodes.

[0081] The task performance evaluation model generates corresponding predicted outputs for each input patient psychological vector and task node assignment combination. The output includes three response parameters for each task node: task progress rate, operational accuracy, and collaborative synchronization rate. To improve the uniformity of comparison, the response parameters of all strategy combinations are normalized. Taking the min-max normalization method as an example, the output values ​​of each strategy under different parameters are compressed to the [0,1] interval. Then, a weighting function is constructed based on the system's preset parameter weights to perform a fusion calculation on the three types of parameters for each strategy, generating the task performance score for that strategy. The weights are set based on task characteristics; for example, in task types emphasizing collaboration, the collaborative synchronization rate weight can be set to 0.5, and the operational accuracy and progress rate weights can be set to 0.25 respectively. Finally, the strategy with the highest score is selected from all candidate strategies, and its task node-patient matching relationship entries are extracted as the final intervention strategy configuration result for this round of virtual gardening collaboration.

[0082] In this embodiment, the virtual gardening task environment is fully implemented using virtual reality (VR) technology. Patients participate in gardening tasks through a wearable head-mounted display and motion-capture-enabled controllers. The system constructs an immersive virtual greenhouse space scene, embedding several interactive gardening work units, such as virtual planting pots, soil manipulation platforms, watering troughs, and pruning areas. These are all generated by a 3D modeling system (e.g., Unity3D or Unreal Engine) and include an event response mechanism that records operation trajectories and timestamps. Specific interactive tasks, such as sowing, watering, fertilizing, pruning, and collaborative planting, are presented in a node-based task flow. The system can dynamically adjust the task progress map based on the completion status of task nodes. The spatial positioning module integrated into the VR system collects the patient's hand movement trajectories and posture changes during operation. Combined with the head-mounted display's gaze tracking function, it records the patient's task response time during the task process, thereby achieving accurate collection of task node behavior data. All of the patient's task behaviors are mapped in real-time to system log data for use by other system modules.

[0083] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0084] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0085] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0086] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0087] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0088] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0089] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0090] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0091] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0092] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A socialized horticultural intervention and evaluation system based on VR technology, characterized in that, It includes modules for building a label combination space, assessing communication willingness, building a label graph, reconstructing grouping, evaluating task performance, and configuring intervention strategies. The tag combination space construction module maps the psychological problems of patients into a three-dimensional psychological tag system composed of emotions, cognition and behavior, and constructs a psychological tag combination space; The communication willingness assessment module monitors virtual gardening collaborative task records, collects patients' interactive behavior data in collaborative task scenarios based on a sliding window, and calculates communication willingness scores among patients with different psychological label combinations. The label mapping module maps patients’ communication willingness scores to a psychological label combination space, constructs a label fit map, and distinguishes between conflicting and complementary labels. The grouping reconstruction module isolates the individual patients corresponding to conflicting labels and reconstructs the virtual gardening collaboration grouping; The task performance evaluation module extracts the task execution feedback data of individual patients corresponding to complementary labels under various horticultural collaborative task structures, and establishes a task performance evaluation model. The intervention strategy configuration module constructs task strategies for patients in the virtual gardening collaboration group based on the output results of the task performance evaluation model, and outputs the corresponding intervention strategy configuration. The label combination space construction module maps the psychological problems of the patient group into a three-dimensional psychological label system composed of emotions, cognition, and behavior. The construction of the psychological label combination space specifically includes: Obtain psychological problem classification records from the medical records of the patient group, and convert each patient's psychological problem classification record into cognitive score, emotional score and behavioral score; For each patient, a single-point psychological vector corresponding to the combination of emotional, cognitive and behavioral labels is established. All the label items corresponding to the single-point psychological vectors are embedded into a three-dimensional spatial structure to establish a psychological label combination space. The step of embedding the labels corresponding to all single-point psychological vectors into a three-dimensional spatial structure to establish a psychological label combination space specifically includes: Construct a three-dimensional coordinate system with cognitive, emotional, and behavioral dimensions as axes, embed the three types of labels in the single-point psychological vector into the corresponding axes and fix the polarity direction; Perform a three-axis mapping operation on the single-point psychological vectors of all patients to generate the coordinates of the single-point psychological vectors of the patients in a three-dimensional coordinate system; Based on the positive and negative combinations of three-axis coordinates, the three-dimensional psychological space is divided into eight non-overlapping quadrant regions, and each quadrant is numbered. Obtain the quadrant number to which the patient's coordinates belong and mark it as the encoding value of the patient's psychological label combination; The task performance evaluation module extracts task execution feedback data of individual patients corresponding to complementary labels under various horticultural collaborative task structures, and establishes a task performance evaluation model, specifically including: The complete operation chain of the completed historical virtual gardening collaborative tasks is broken down into process nodes according to task stages and interaction types, and a set of numbered task structure units is constructed. In the psychological label combination space, extract the single-point psychological vector composed of the cognitive score, emotional score and behavioral score of each patient in each reconstructed virtual gardening collaboration group; The system collects feedback data from the execution logs of each task node within the task structure unit set, obtains the executor of the task node, and integrates the single-point psychological vector of the executor with the response parameters of the task node into a task behavior sample set. The response parameters of the task nodes are obtained from the feedback data in the execution log, including the task progress rate, operation accuracy, and collaboration synchronization rate. The task behavior sample set is used as the training sample set, and the patient's single-point psychological vector and task node number are used as input items, while the response parameters of the task node are used as output items to train the task performance evaluation model.

2. The socialized horticulture intervention and evaluation system based on VR technology according to claim 1, characterized in that, The communication willingness assessment module monitors virtual gardening collaborative task records, collects patients' interactive behavior data in collaborative task scenarios based on a sliding window, and calculates communication willingness scores among patients with different psychological label combinations, specifically including: During the execution of a virtual gardening task, interactive task nodes for two-person collaboration are identified, and the sequence of interactive behaviors is extracted based on the timestamps of the patient's initiation and response to interactions within the task nodes. Define an equal-width sliding time window, extract the patient's interaction behavior fragments window by window in the interaction behavior sequence, and count the cumulative number of initiated and responded interactions; Within each sliding time window, the cumulative length of delayed response behavior and the proportion of asynchronous response behavior among patients are counted to establish an interactive response breakage index. Construct a communication willingness scoring function based on weighted integration, and generate a communication willingness score based on the cumulative number of interactions and the interaction response break index.

3. The socialized horticulture intervention and evaluation system based on VR technology according to claim 1, characterized in that, The label mapping module maps patients' communication willingness scores to a psychological label combination space, constructs a label fit map, and distinguishes between conflicting and complementary labels, specifically including: Obtain the communication willingness scores of patients in two-person collaborative interactions at each interactive task node, and extract the encoded values ​​of the two patients in the psychological label combination space; Based on the coded values, the communication willingness scores are grouped according to quadrant regions. The mean of the communication willingness scores within each quadrant region group is calculated, and a psychological label group suitability score matrix is ​​constructed. Using the psychological label combinations in the fit rating matrix as nodes and the relationships between psychological label combinations as connecting edges, a psychological label fit graph structure is established, and fit ratings are labeled on the connecting edges. Set threshold ranges for determining conflict and complementarity, and classify different combinations of psychological labels into conflict labels and complementary labels based on the threshold range in which the fit score falls.

4. The socialized horticulture intervention and evaluation system based on VR technology according to claim 1, characterized in that, The group reconstruction module isolates patient individuals corresponding to conflicting tags and reconstructs virtual gardening collaboration groups, specifically including: Retrieve conflicting label pairs already identified in the psychological label fit map and locate the corresponding patient psychological label combination code; Extract patient individuals with conflicting label combination codes from the current collaboration group, and remove the corresponding patient individuals' task participation history and interaction trajectory; Based on the distribution of conflict labels in each collaborative group, a conflict-minimizing sort is performed to determine the priority sequence of individual patients to be isolated. Prioritize the isolation of patients and remove them from the original collaborative groups to create a new list of virtual horticulture collaborative group combinations.

5. A socialized horticulture intervention and evaluation system based on VR technology according to claim 1, characterized in that, The intervention strategy configuration module constructs task strategies for patients within the virtual horticulture collaboration group based on the output results of the task performance evaluation model, and the corresponding intervention strategy configurations output include: Regularly update the medical records of the patient group. When the number of patients increases or the single-point psychological vector coordinates of the psychological label combination space show limit changes, update the label adaptability map and reconstruct the virtual gardening collaboration group. In each virtual gardening collaboration group, a task node allocation strategy table for all patients performing tasks is pre-generated, and the patient's single-point psychological vector and task node allocation strategy are input into the task performance evaluation model. The output of the task performance evaluation model is normalized and converted into a task performance score through weighted fusion. The task node allocation strategy corresponding to the highest task performance score is selected as the intervention strategy configuration.

Images