A procrastination management method and system

Abstract

The application relates to the field of procrastination intervention, and particularly relates to a procrastination management method and system. The method comprises the following steps: acquiring multi-modal user data; determining a procrastination type of a user based on the multi-modal user data; setting a first weight coefficient of a behavior characteristic item and a second weight coefficient of a physiological characteristic item according to the procrastination type; calculating the deviation degree between the measured current value of each characteristic item and the baseline value of the characteristic item, and determining the overall deviation degree based on the deviation degree of each characteristic item and the corresponding weight coefficient; setting a response gain coefficient according to the intervention acceptance degree in the current application scenario; determining an intervention intensity parameter based on the response gain coefficient and the overall deviation degree; and calling an intervention strategy corresponding to the intervention intensity under the current procrastination type based on the intervention intensity parameter. While improving the timeliness and effectiveness of intervention, the negative effects such as the sense of intrusion, psychological resistance or weakening of internal motivation caused by improper intervention are effectively avoided.

Description

Technical Field

[0001] This application relates to the field of procrastination intervention, and more particularly to a procrastination management method and system. Background Technology

[0002] Procrastination, referring to the tendency to avoid performing tasks or making decisions by postponing them, has increasingly become a focus of attention in the fields of psychology, education, and management. Its causes are complex and have long since transcended the scope of simply poor time management or laziness. It has been deeply analyzed as a comprehensive psychological and behavioral model involving deficiencies in anticipation and imagination, failure of self-control, and difficulty in emotional regulation.

[0003] Procrastinators are often not lacking in time management knowledge, but rather trapped by an avoidance of negative emotions associated with tasks (such as anxiety, fear of failure, and aversion), and an excessive preference for immediate gratification due to a high delay discount rate. Therefore, intervention strategies have gradually shifted their focus from external behavioral correction to the adjustment of internal psychological mechanisms.

[0004] For example, patent application CN119724491A discloses a procrastination identification and intervention system, including a procrastination identification system and a procrastination intervention system. The procrastination identification system includes a multimodal data acquisition module for collecting data; a data analysis and mining module for preprocessing the data and assessing the characteristics and levels of procrastination behavior; and a procrastination identification result module for obtaining procrastination identification results and the levels of related core cognitive abilities. The procrastination intervention system includes a personalized intervention module for generating personalized cognitive and neuromodulation intervention plans; an intervention guidance and feedback module for conducting procrastination intervention training; and an intervention data management module for recording intervention training data.

[0005] For example, patent application CN120048445A discloses a psychological assessment and counseling method and system based on artificial intelligence and large models, belonging to the field of psychological assessment and counseling technology. The method includes the following steps: S1, real-time acquisition and processing of user data; S2, feature selection and extraction; S3, construction of a large psychological assessment model to predict the user's psychological state and its development trend; S4, providing users with psychological counseling plans, timely intervention in the user's psychology, and optimization and adjustment of the psychological counseling plan based on user feedback.

[0006] However, existing procrastination interventions often fail to achieve the desired results in practice and may even cause excessive interference with an individual's normal life and mental state. Summary of the Invention

[0007] The main objective of this application is to provide a method and system for managing procrastination. To solve the aforementioned technical problems, this application specifically adopts the following technical solution: The first aspect of this application is to provide a method for managing procrastination, the method comprising: S101, acquire multimodal user data; the multimodal user data includes: test results of a preset psychological scale, and various behavioral and physiological features updated based on preset time intervals; S102, Identify the user's procrastination type based on the multimodal user data; S103, according to the procrastination type, set the first weight coefficient of the behavioral characteristic item and the second weight coefficient of the physiological characteristic item respectively; S104, calculate the degree of deviation between the measured current value of each feature item and the baseline value of the feature item, and determine the overall degree of deviation based on the degree of deviation of each feature item and the corresponding weight coefficient; S105, Set the response gain coefficient according to the level of intervention acceptance in the current application scenario; Determine the intervention intensity parameter based on the response gain coefficient and the overall deviation. S106, Based on the intervention intensity parameter, invoke the intervention strategy corresponding to the current procrastination type.

[0008] In some embodiments, the intervention strategy includes a cognitive behavioral training strategy and an immediate feedback prompting strategy; the method includes: configuring at least one training parameter of the cognitive behavioral training strategy according to the intervention intensity parameter; the training parameter includes: training format, course content, execution frequency, execution intensity, and single execution duration; the training format includes at least one of online interactive courses and neuromodulation training; configuring at least one prompting parameter of the immediate feedback prompting strategy according to the intervention intensity parameter; the prompting parameter includes: prompting modality, prompting intensity, or presentation method; the prompting modality includes at least one of vibration, audio prompts, or graphic pop-ups; the prompting intensity is adjusted based on at least one of vibration frequency, audio volume and rhythm, or the size and duration of the graphic pop-up.

[0009] In some embodiments, the procrastination type includes a first-level procrastination type, including at least one of starting difficulty, execution difficulty, and hybrid type; the method includes: when the procrastination type is starting difficulty, the first weight coefficient is higher than the second weight coefficient; when the procrastination type is execution difficulty, the second weight coefficient is higher than the first weight coefficient; when the procrastination type is hybrid type, determining the current dominant type as starting difficulty or execution difficulty, and determining the weight coefficient of each feature item according to the current dominant type.

[0010] In some embodiments, in the first-level procrastination type, the maximum difference between the first weight coefficients of each behavioral feature item is less than or equal to a first preset threshold, the maximum difference between the second weight coefficients of each physiological feature item is less than or equal to a second preset threshold, and the maximum difference between the first weight coefficient and the second weight coefficient is less than or equal to a third preset threshold.

[0011] In some embodiments, the procrastination type further includes multiple secondary procrastination types subdivided under each primary procrastination type; in the secondary procrastination type, the maximum difference between the first weight coefficients of each behavioral feature item is greater than a first preset threshold; and / or, the maximum difference between the second weight coefficients of each physiological feature item is greater than a second preset threshold; and / or, the maximum difference between the first weight coefficient and the second weight coefficient is greater than a third preset threshold.

[0012] In some embodiments, the behavioral features include at least one of voice features, facial expression features, limb movement features, and eye movement features; the physiological features include at least one of electrodermal features and pulse wave features.

[0013] In some embodiments, the secondary procrastination type in the initiation difficulty type includes at least one of intention formation difficulty type and intention-action conversion difficulty type; the method includes: when the procrastination type is intention formation difficulty type, the maximum difference between the first weight coefficient and the second weight coefficient is greater than a third preset threshold; when the procrastination type is intention-action conversion difficulty type, the maximum difference between the first weight coefficient of the limb movement feature and the first weight coefficient of the voice feature and facial expression feature is greater than a first preset threshold.

[0014] In some embodiments, the secondary procrastination type in the execution difficulty type includes at least one of self-control deficiency type and emotion regulation deficiency type; when the procrastination type is self-control deficiency type, the maximum difference between the second weight coefficient of the skin conductance feature and the second weight coefficient of the pulse wave feature is greater than a second preset threshold; when the procrastination type is emotion regulation deficiency type, the maximum difference between the first weight coefficient and the second weight coefficient is greater than a third preset threshold.

[0015] In some embodiments, after S103, the method further includes: obtaining scenario information in the current application scenario, wherein the scenario information includes at least one of user schedule status, data collection permission status, and environmental noise level; evaluating the data quality score of each feature item based on the scenario information, and adjusting the weight coefficient of the corresponding feature item according to the data quality score to update the weight coefficient.

[0016] A second aspect of this application is to provide a procrastination management system, the system comprising: The data acquisition module is used to acquire multimodal user data, which includes: test results of a preset psychological scale, and various behavioral and physiological features updated based on preset time intervals. The type identification module is used to identify the user's procrastination type based on the multimodal user data; The weight setting module is used to set the first weight coefficient of the behavioral feature item and the second weight coefficient of the physiological feature item according to the procrastination type. The deviation calculation module is used to calculate the degree of deviation between the measured current value of each feature item and the baseline value of the feature item, and to determine the overall deviation degree based on the degree of deviation of each feature item and the corresponding weight coefficient. The gain setting module is used to set the response gain coefficient according to the intervention acceptance level in the current application scenario; and to determine the intervention intensity parameter based on the response gain coefficient and the overall deviation. The strategy invocation module is used to invoke the intervention strategy corresponding to the current procrastination type based on the intervention intensity parameter.

[0017] Beneficial effects: This application provides a method and system for managing procrastination. Through highly precise quantification of intervention intensity, it achieves accurate control over the timing and intensity of intervention measures, dynamically adapting to the user's individualized causes of procrastination, real-time psychological and physiological state, and current application scenario. This enables personalized configuration, contextualized response, and low-intrusion implementation of intervention strategies. In practical applications, while improving the timeliness and effectiveness of intervention, it fully respects user autonomy and effectively avoids negative effects such as intrusiveness, psychological resistance, or weakened intrinsic motivation caused by inappropriate intervention.

[0018] First, based on multi-source heterogeneous data, support is provided for the accurate quantification of intervention intensity. This includes not only the assessment results of standardized psychological scales but also the simultaneous collection of multi-dimensional behavioral characteristics (such as voice, eye movement, and body movements) and physiological characteristics (such as skin conductance and pulse wave). On this basis, through interpretable rules for calculating the overall deviation of user states, an objective, quantifiable, and traceable intervention intensity is generated.

[0019] Furthermore, in the calculation of the overall deviation, the application of multi-source heterogeneous data is refined and differentiated by adjusting the feature weights, which greatly improves the accuracy and pertinence of the quantitative analysis of intervention intensity.

[0020] First, feature weights are dynamically adjusted according to the type of procrastination. Procrastination is subdivided into primary types and more refined secondary subtypes. Different types of procrastination mechanisms have varying degrees of dependence on behavioral and physiological characteristics, and differentiated weights are assigned accordingly: for example, initiation difficulty emphasizes behavioral characteristics, while execution difficulty emphasizes physiological characteristics. More importantly, the polarization of the weights matches the classification granularity. The finer the classification granularity, the more focused the weights. For example, balanced weights are used in primary types to ensure robustness, while polarized weights are used in secondary subtypes to strengthen the application of key features. This truly achieves targeted treatment, significantly improving the accuracy of procrastination identification and the effectiveness of interventions.

[0021] Secondly, feature weights are dynamically adjusted based on data quality. When the credibility of a feature decreases due to environmental interference (such as a noisy background affecting speech recognition) or permission restrictions (such as the camera being turned off), its weight is automatically reduced to avoid misjudgment and over-intervention caused by noisy data, thereby enhancing decision robustness and preventing user anxiety or resistance.

[0022] Furthermore, by introducing a response gain coefficient based on scenario-aware intervention acceptance, differentiated intervention intensities can be achieved for the same type of procrastination and the same overall degree of deviation. This provides a flexible response strategy that aligns with the user's current situational preferences, effectively avoiding inappropriate interruptions, fully respecting individual autonomy, protecting the user's intrinsic motivation, and preventing the resistance or additional psychological burden that traditional coercive interventions may induce. For example, for the same intervention intensity, voice prompts or interface guidance can be enabled in high-acceptance scenarios (such as self-study periods), while only slight vibrations or delayed reminders are triggered in low-tolerance scenarios (such as meetings or driving). Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. The elements or parts in the drawings are not necessarily drawn to scale. Obviously, the drawings described below are some embodiments of this application; for those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0024] Figure 1 This is a schematic flowchart illustrating a method for managing procrastination provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating a classification of delay types provided in an embodiment of this application; Figure 3 This is a schematic flowchart illustrating yet another method for managing procrastination provided in an embodiment of this application; Figure 4 This is a schematic block diagram of a procrastination management system provided in an embodiment of this application; Figure 5 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0026] The flowcharts shown in the attached diagrams are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily need to be performed in the described order. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation. Furthermore, although functional modules are divided in the system diagram, in some cases, a different module division may be used.

[0027] In this document, suffixes such as “module,” “part,” or “unit” used to denote elements are used only for illustrative purposes and have no specific meaning in themselves. Therefore, “module,” “part,” or “unit” may be used interchangeably.

[0028] In this article, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0029] In this document, unless otherwise explicitly specified and limited, the terms "installed," "equipped with," and "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, a direct connection, or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0030] In this document, the term “and / or” includes any and all combinations of one or more of the listed related items.

[0031] In this article, the term "multiple" means two or more, that is, it includes two, three, four, five, etc.

[0032] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0033] This application provides a method and system for managing procrastination. Through highly precise quantification of intervention intensity, it achieves accurate control over the timing and intensity of intervention measures, dynamically adapting to the user's individualized causes of procrastination, real-time psychological and physiological state, and current application scenario. This enables personalized configuration, contextualized response, and low-intrusion implementation of intervention strategies. In practical applications, while improving the timeliness and effectiveness of intervention, it fully respects user autonomy and effectively avoids negative effects such as intrusiveness, psychological resistance, or weakened intrinsic motivation caused by inappropriate intervention.

[0034] The following detailed description, in conjunction with the accompanying drawings, outlines some embodiments of this application. Unless otherwise specified, the following embodiments and features described herein can be combined with each other. Please refer to... Figure 1 , Figure 1 This is a schematic flowchart illustrating a method for managing procrastination provided in an embodiment of this application, such as... Figure 1 As shown in the figure, this application provides a method for managing procrastination, the method including S101 to S105.

[0035] S101, acquire multimodal user data; the multimodal user data includes: test results of a preset psychological scale, and various behavioral and physiological features updated based on preset time intervals.

[0036] For example, multimodal user data can be collected periodically through user terminal devices such as smartphones, wearable devices, or learning terminals with integrated sensors.

[0037] Among them, the pre-selected psychological scale refers to a standardized questionnaire tool that is pre-selected to assess an individual's cognitive, emotional, or executive function-related psychological traits, such as the Cognitive Scale-Expectation and Imagination Scale, the Cognitive Scale-Adolescent Self-Control Dual System Scale, and the Cognitive Scale-Emotion Regulation Difficulty Scale, etc., which are not limited here. For example, the pre-selected psychological scale can be completed by the user in the form of an interactive questionnaire during the initialization phase or periodic assessment, and the generated structured score or dimension score reflects the user's level in a specific psychological ability.

[0038] Among them, behavioral features refer to the visible indicators captured by sensors or cameras, including at least one of the following: voice features (such as speech rate and pauses), facial features (such as micro-expression types and frequencies), body movement features (such as changes in sitting posture), and eye movement features (such as gaze duration and saccade patterns).

[0039] Physiological characteristics refer to indicators related to the autonomic nervous system collected by wearable or contact sensors, including at least one of skin conductance characteristics and pulse wave characteristics. Skin conductance characteristics can be skin conductance level (SCL) or skin conductance response (SCR); the waveform morphology of the pulse wave reflects cardiovascular status such as vascular elasticity and peripheral resistance, and the temporal variability between its continuous peaks (i.e., fluctuations in the pulse interval) is an important source of heart rate variability (HRV), which can be used to assess autonomic nervous function, stress level, and overall physiological adaptability.

[0040] For example, behavioral and physiological characteristics are automatically collected and updated at preset time intervals (such as every 5 minutes or every 2 minutes) during the user's daily use, based on the built-in sensors or cameras of the terminal device. This improves the continuity and real-time performance of the data and also allows for control of the collection frequency, reducing interference with the user.

[0041] It should be noted that the collection and analysis of behavioral and physiological features can refer to existing technologies. For example, facial expression features can be extracted using computer vision algorithms to extract facial action units (such as AU4 corrugator supercilii activity), micro-expression types, and durations; speech features can be analyzed using signal processing techniques to analyze prosody, speech events, energy, zero-crossing rate, and speech rate changes; eye movement features can be tracked using eye trackers or cameras to track fixation points, saccade trajectories, and pupil diameters; and limb features can be identified based on accelerometers or posture sensors to recognize sitting posture, gestures, and micro-movement frequencies. Regarding physiological features, skin conductance can be used to extract SCR amplitude, latency, and SCL baseline; and pulse waves can be used to calculate the LF, HF, and other frequency domain components in HRV.

[0042] S102, Identify the user's procrastination type based on the multimodal user data.

[0043] Specifically, based on multimodal user data, combined with the test results of a pre-set psychological scale, and / or the user's behavioral and physiological characteristics during task execution, the psychological abilities related to the causes of procrastination that the user lacks are assessed to determine the type of procrastination.

[0044] Psychological abilities include at least one of the following: anticipatory imagination, self-control, and emotion regulation. Anticipatory imagination refers to an individual's ability to simulate and imagine the long-term positive results of completing a task; self-control refers to an individual's ability to resist tempting stimuli (such as distraction) and maintain focus on the task while avoiding interruptions from other short-term temptations; and emotion regulation refers to an individual's ability to regulate negative emotions (such as stress, anxiety, and boredom) and maintain a positive emotional state during task execution, preventing emotional fluctuations from affecting task performance.

[0045] Please see Figure 2 , Figure 2 This is a classification diagram of delay types provided in an embodiment of this application, such as... Figure 2 As shown, in some embodiments, the procrastination type includes a primary procrastination type; the primary procrastination type includes at least one of initiation difficulty, execution difficulty, and hybrid type. Initiation difficulty manifests as difficulty in starting a plan or task; execution difficulty (also known as execution maintenance difficulty) manifests as difficulty in maintaining focus and persisting until completion after the task has begun; hybrid type manifests as the simultaneous presence of task initiation difficulty and execution maintenance difficulty.

[0046] For example, priming difficulties primarily stem from deficiencies in anticipatory imagination. This ability can be assessed using dimensions such as the Cognitive Scale-Anticipatory Imagination Scale, and the user's vocal and facial features during the task, to identify priming difficulties. For instance, in terms of vocal features, users exhibit flat rhythm, low energy (e.g., a negative RMS slope in the vocal signal), and hesitant pauses (e.g., abnormal zero-crossing rate in the vocal signal). In terms of facial features, they exhibit a lack of activation of positive muscle groups (e.g., the zygomaticus major muscle) and eye avoidance.

[0047] It should be understood that the intention to perform a task stems from a trade-off between the positive value of the task's outcome (such as rewards and a sense of accomplishment upon completion) and the negative utility of the task's process (such as negative emotions like aversion and boredom). If the positive value of the outcome outweighs the negative utility, the individual will have the intention to do it immediately; otherwise, they will not. When there is a deficiency in the ability to anticipate and imagine, the value of the positive outcome is weakened in cognitive evaluation, making it unable to surpass the negative utility in value comparison, thus increasing the risk of difficulty in initiating the task.

[0048] For example, executive dysfunction primarily stems from deficiencies in self-control and emotion regulation. This can be assessed using the Cognitive Scale-Adolescent Self-Control Dual System Scale and the Emotion Regulation Difficulty Scale, as well as skin conductance characteristics during task performance (such as increased SCR, decreased SCR latency, increased power in the low-frequency band and decreased power in the high-frequency band) and pulse wave characteristics (such as increased heart rate, decreased pulse wave amplitude, and decreased power in the high-frequency band). It should be understood that during task performance, individuals experience negative emotions, leading to overactivation of the sympathetic nervous system and withdrawal of vagal tone, which in turn affects skin conductance and pulse wave characteristics.

[0049] For example, if a user is found to have deficiencies in their ability to anticipate and imagine, their self-control, and their emotional regulation, they can be classified as a mixed type.

[0050] In some embodiments, each primary procrastination type is further subdivided into multiple secondary procrastination types.

[0051] For example, such as Figure 2 As shown, the secondary types of procrastination within the "difficulty in initiating" category include at least one of "difficulty in forming intention" and "difficulty in converting intention into action." "Difficulty in forming intention" corresponds to a deficiency in the content of the anticipated imagery, meaning it's difficult to vividly construct or extract the positive value of the completed task, resulting in a severe lack of "meaning in doing this." "Difficulty in converting intention into action" corresponds to a deficiency in the connection between anticipated imagery and action, meaning one can imagine a positive outcome but cannot transform "future benefits" into an "immediate driving force" to initiate current action.

[0052] For example, such as Figure 2 As shown, the secondary procrastination type within the executive difficulty category includes at least one of the following: self-control deficit-dominated type and emotion regulation deficit-dominated type. Specifically, the dominant ability type can be determined based on the severity of the ability deficiency.

[0053] In some embodiments, step S102 can be implemented manually or automatically. For example, the integrated multimodal user data (including psychological scale test results, temporal changes in behavioral and physiological characteristics) can be provided to a professional physician, who can then determine the user's missing skill type through online or offline consultations or by combining clinical experience, and accordingly identify the corresponding primary and / or secondary procrastination types. For example, a machine learning model can be pre-trained based on pre-collected training data (including labeled skill deficiency tags, corresponding procrastination types, and corresponding psychological scale test results, behavioral and physiological characteristics); when multimodal data from a new user is input, the model can automatically output the skill deficiency assessment results and the corresponding procrastination type, thereby achieving efficient and scalable automated classification.

[0054] It should be noted that the procrastination types listed in the embodiments of this application (such as initiation difficulty, execution difficulty, mixed type, and their secondary subtypes) are merely illustrative examples and do not constitute a limitation on the scope of protection of this invention. In practical applications, procrastination types can be further subdivided based on broader psychological ability dimensions (such as working memory capacity, time perception bias, etc.) to form a more individualized and mechanism-oriented classification system. Correspondingly, for each refined procrastination type, prior experiments or clinical studies can be combined to evaluate the discriminative contribution of each behavioral and physiological characteristic to that type; and the applicable weight configuration under each procrastination type can be verified through controlled experiments and stored in the weight configuration rule base.

[0055] S103, according to the procrastination type, set the first weight coefficient of the behavioral characteristic item and the second weight coefficient of the physiological characteristic item respectively.

[0056] Specifically, based on the type of delay, the weight coefficients of different features are retrieved from the preset weight configuration rule library to make different features contribute differently to the overall degree of deviation.

[0057] Specifically, based on the type of procrastination, the weight coefficients of different feature items are retrieved from a pre-defined weight configuration rule base, so that different feature items have differentiated contributions when calculating the overall degree of deviation. The weight configuration rule base refers to a pre-built set of structured rules that associates procrastination types with the weight coefficients of each feature item. The weight allocation matches the causes of user procrastination, thereby improving the targeting and accuracy of subsequent intervention intensity calculations.

[0058] In some embodiments, when the procrastination type is a difficult-to-start type, the first weighting coefficient is higher than the second weighting coefficient. For example, the weighting coefficients of each feature can be configured as follows: voice feature: 0.17; facial expression feature: 0.20; eye movement feature: 0.18; limb movement feature: 0.20; skin conductance feature: 0.13; pulse wave feature: 0.14.

[0059] In some embodiments, when the procrastination type is execution difficulty, the second weighting coefficient is higher than the first weighting coefficient. For example, the weighting coefficients of each feature can be configured as follows: voice feature: 0.13; facial expression feature: 0.15; eye movement feature: 0.14; limb movement feature: 0.16; skin conductance: 0.22; pulse wave: 0.20.

[0060] It should be understood that when a task is identified as having difficulty initiating, the first weight coefficient of behavioral features such as voice features and facial features should be increased, as these can effectively reflect the vividness of the user's expected imagination and the degree of emotional investment before the task is initiated; when a task is identified as having difficulty in execution, the weight of physiological features such as skin conductance and pulse wave should be increased, as these can more sensitively reflect the emotional fluctuations and cognitive load during the task execution process.

[0061] In some embodiments, when the delay type is a hybrid type, the current dominant type is determined to be either start-up difficulty type or execution difficulty type, and the weight coefficient of each feature is determined according to the current dominant type.

[0062] It should be understood that for users with mixed procrastination patterns, the current dominant type (difficult to start or difficult to execute) is determined based on the stage of the task, and the corresponding weight allocation for the dominant type is applied. For example, if the task is in the planning or initial stage, the dominant type tends to be "difficult to start"; if the task has entered the execution process, the dominant type tends to be "difficult to execute". Furthermore, the dominant type is dynamically adjusted based on the latest stage of the task, or the secondary procrastination type corresponding to the dominant type is identified and updated based on behavioral and physiological characteristics, and the weight configuration of the secondary procrastination type is further applied. For example, during task execution, if abnormal physiological characteristics indicating the accumulation of negative emotions, such as a rapid and continuous increase in SCL and a significant decrease in high-frequency power based on heart rate variability of pulse waves, are detected, the current dominant type will be re-determined as "emotion regulation deficit dominant type" (difficult to execute), and the weight configuration of the secondary procrastination type will be immediately switched.

[0063] S104 Calculate the degree of deviation between the measured current value of each feature item and the baseline value of the feature item, and determine the overall degree of deviation based on the degree of deviation of each feature item and the corresponding weight coefficient.

[0064] Among them, the baseline value can refer to a reference value that is established individually for each user or each user group, representing a certain characteristic under its typical or stable state. It can be determined based on historical data of historical multimodal data under good functional states (such as low stress, high focus state).

[0065] Specifically, the measured values ​​of each feature item are obtained at the current time point, and the degree of deviation between them and the baseline value is calculated by means of relative change rate or standardized difference, so as to quantify the current level of abnormality or fluctuation under each feature item dimension. Then, the degree of deviation of each feature item is multiplied by its corresponding weight coefficient to obtain a weighted deviation value. These deviation values ​​are added together to obtain the overall deviation degree reflecting the current overall abnormality level of the user, which serves as the basis for subsequent intervention intensity calculation.

[0066] S105, Set the response gain coefficient according to the level of intervention acceptance in the current application scenario; Determine the intervention intensity parameter based on the response gain coefficient and the overall deviation.

[0067] Specifically, the system acquires contextual information about the current application scenario, such as the user's task status (e.g., whether they are in a meeting, studying, or resting period) and personal physical condition. Based on this, it assesses the user's acceptance of different intervention methods and sets corresponding response gain coefficients for different levels of acceptance. This coefficient is used to amplify or suppress the response to the overall deviation. It typically ranges from 0 to 1, with higher values ​​indicating a more suitable scenario for stronger intervention. Furthermore, multiplying the response gain coefficient by the overall deviation yields the intervention intensity parameter at the current moment.

[0068] For example, when a user's task status is "focused on studying and preparing for exams," task completion is the priority in this type of application scenario. Users are more receptive to moderate external encouragement and can act as active motivators. It can be set to a medium-high value, such as 0.7.

[0069] For example, when a user's task status is "self-directed work / creation," the social norms of the professional environment need to be maintained in such application scenarios. Users have a low acceptance of intervention methods and become silent assistants. It can be set to a low value, such as 0.3. This can avoid high-intensity interventions from excessively encroaching on an individual's autonomy, undermining their intrinsic motivation, or even triggering rebellious psychology or anxiety, thus increasing their psychological burden.

[0070] For example, when an individual's physical state is "healthy or emotionally managed," the core objective in such application scenarios is psychological protection and stress relief, and users have a high acceptance of intervention methods (or a high demand for them). Set as an adaptive value, and it exhibits a non-linear positive correlation with the deviation of physiological characteristics (e.g., exponential growth). For example, when data fluctuations corresponding to severe physiological stress (e.g., abnormal HRV) are detected in physiological characteristics, Adjust to a higher value, such as 0.9.

[0071] For example, when a user's task status is "Procrastination Ability Training Course," the goal is to cultivate a specific ability within a controlled training environment. It can be set to a fixed median value, such as 0.5.

[0072] S106, Based on the intervention intensity parameter, invoke the intervention strategy corresponding to the current procrastination type.

[0073] Specifically, based on the threshold range of the intervention intensity parameter, a specific intervention strategy matching the current procrastination type and intensity is retrieved from a pre-defined intervention strategy library and executed. This intervention strategy library can be organized according to different procrastination types (such as difficulty initiating, difficulty in execution, and their subtypes), and multiple intensity levels of intervention plans are predefined for each type. For example, low intensity triggers a brief notification in the corner of the screen (e.g., "30 minutes remaining until the planned task completion time"); medium intensity triggers a micro-vibration notification; and high intensity triggers a high-frequency continuous vibration.

[0074] In some embodiments, after determining the intervention intensity parameter, it is normalized to ensure that its value is stable and comparable across users and scenarios. The normalized value directly corresponds to the preset intervention intensity level, as shown in Table 1.

[0075] Table 1. Intervention strategies under different intervention intensity parameters In some embodiments, the intervention strategy includes cognitive behavioral training strategies applied to the ability training section in Table 1. The goal of this section is to repair and improve long-term psychological abilities, in which the user is in a controlled environment with low interference and high focus (such as scheduled training sessions).

[0076] The cognitive behavioral training strategy is configured with at least one training parameter based on the intervention intensity parameter. The training parameters include: training format, course content, execution frequency, execution intensity, and single execution duration. The training format includes at least one of online interactive courses and neuromodulation training. Neuromodulation training can be transcranial direct current stimulation (tDCS) or other non-invasive neuromodulation training. For example, for severely impaired users, tDCS can be used, targeting the orbitofrontal cortex (OFC), with an execution frequency of once daily, a single execution duration of 20-30 minutes, and an execution intensity of 1.5 mA. The treatment duration and frequency will be personalized based on the deviation in physiological characteristics (such as the degree of abnormality in skin conductance response relative to baseline data).

[0077] In some embodiments, the intervention strategy includes an immediate feedback prompting strategy, applied to the daily training section in Table 1, which aims at immediate behavioral adjustment in real-life scenarios. Based on highly real-time behavioral and physiological characteristics, the timing of intervention can be more accurately determined, effectively preventing procrastination at its onset.

[0078] Specifically, at least one prompting parameter in the instant feedback prompting strategy is configured according to the intervention intensity parameter; the prompting parameter includes: prompting modality, prompting intensity, or presentation mode; the prompting modality includes at least one of vibration, audio prompt, or graphic pop-up; the prompting intensity is adjusted based on at least one of vibration frequency, audio volume and rhythm, or graphic pop-up size and duration.

[0079] In some embodiments, in the first-level procrastination type, the maximum difference between the first weight coefficients of each behavioral feature item is less than or equal to a first preset threshold, the maximum difference between the second weight coefficients of each physiological feature item is less than or equal to a second preset threshold, and the maximum difference between the first weight coefficient and the second weight coefficient is less than or equal to a third preset threshold.

[0080] It should be understood that when there is a lack of sufficiently accurate or personalized user data, or when the system is in a cold start phase, the granularity of procrastination classification is limited to the first-level procrastination type. At this stage, weight allocation follows a balance constraint, meaning that the weight differences between behavioral features, between physiological features, and between behavioral and physiological features are all controlled within preset thresholds. This ensures that the contribution of each feature to the overall assessment is relatively balanced, reflecting the principle of prudence in intervention strategies. It avoids misjudgments due to excessive weighting of individual noise signals, and also prevents overly strong or inappropriate interventions triggered based on incomplete information.

[0081] In some embodiments, different values ​​may be applied to the normalized intervention intensity parameter depending on the type of delay. Hierarchical mapping standard This approach aims to match the sensitivity and needs of different user types to interventions. For example, under Level 1 procrastination (such as difficulty in initiating or difficulty in execution), a balanced interval grading strategy is adopted, evenly dividing the intervention intensity parameter into three intervals (e.g., [0, 0.33], (0.33, 0.66], (0.66, 1]) to ensure a stable and conservative intervention response, suitable for Level 1 procrastination. After identifying more refined Level 2 procrastination types, a grading strategy referencing a normal distribution is adopted, with the mild range set to [0, 0.3], the moderate range expanded to (0.3, 0.7), and the severe range to [0.7, 1]. This approach is more sensitive to moderate deviations when key competency deficiencies are clear, enabling timely triggering of appropriate interventions to prevent the problem from worsening, while reserving high-intensity interventions for truly urgent situations, thereby improving the timeliness and accuracy of interventions.

[0082] In some embodiments, in the secondary procrastination type, the maximum difference between the first weight coefficients of each behavioral feature item is greater than a first preset threshold; and / or, the maximum difference between the second weight coefficients of each physiological feature item is greater than a second preset threshold; and / or, the maximum difference between the first weight coefficient and the second weight coefficient is greater than a third preset threshold.

[0083] It should be understood that as the granularity of procrastination type classification becomes more refined, the understanding of the causes of individual procrastination becomes more accurate. Correspondingly, the weights of certain key features can be strengthened to improve identification accuracy. Under the second-level procrastination type, significant differences in weights are allowed or even required: this can be achieved by widening the weight differences within the same category of features, as well as between different categories of features. This enhances the sensitivity of identifying specific procrastination subtypes and the accuracy of intervention matching, making it particularly suitable for specialized, goal-oriented skills training modules, and intervention scenarios with stable and reliable user data.

[0084] The first preset threshold, the second preset threshold, and the third preset threshold are the upper limits for constraining the weight differences within behavioral features, physiological features, and between behavioral and physiological features, respectively. The specific values ​​can be set according to the number of features and application scenario experience. For example, in a typical configuration, the first preset threshold can be set to 0.03, the second preset threshold can be set to 0.03, and the third preset threshold can be set to 0.1.

[0085] In some embodiments, when the procrastination type is the intention-forming difficulty type, the maximum difference between the first weight coefficient and the second weight coefficient is greater than a third preset threshold, and / or, the maximum difference between the first weight coefficients of each behavioral feature item is greater than a first preset threshold.

[0086] It should be understood that the core discriminative features in the intention formation difficulty category are flat speech and lack of positive facial expressions, which are assigned high weights. Furthermore, in the pure intention formation stage, physiological features may not have undergone significant changes, and are assigned low weights. Therefore, the weight coefficients of each feature can be configured as follows: speech features: 0.25; facial expressions: 0.25; eye movement features: 0.22; limb movement features: 0.18; skin conductance features: 0.05; pulse wave features: 0.05.

[0087] In some embodiments, when the procrastination type is the difficulty in converting intention into action, the maximum difference between the first weight coefficient of the limb movement feature and the first weight coefficient of the voice feature and facial expression feature is greater than a first preset threshold.

[0088] It should be understood that the core discriminant features in the difficulty of converting intention into action are micro-limb movement features such as initial action hesitation, which have high weight. Voice and facial expressions are used to verify the accuracy of the discriminant and have medium weight. Physiological features may be accompanied by mild decision-making conflict anxiety, but are not the primary discriminant features. Therefore, the weight coefficients of each feature can be configured as follows: voice feature: 0.17; facial expression feature: 0.17; eye movement feature: 0.17; limb movement feature: 0.22; skin conductance feature: 0.13; pulse wave feature: 0.14.

[0089] In some embodiments, when the procrastination type is dominated by self-control deficiency, the maximum difference between the second weighting coefficient of the skin conductance feature and the second weighting coefficient of the pulse wave feature is greater than a second preset threshold.

[0090] It should be understood that the core discriminant feature in the dominant type of self-control deficit is the low-frequency component of heart rate variability reflecting cognitive effort and conflict, or subtle fluctuations in pulse wave morphology during the task period, which may manifest as sluggish regulation when faced with temptation. Electrodermal activity (EDA) is used to differentiate whether interruption is due to impulsivity (possibly accompanied by a brief SCR) or due to accumulated emotions (a sustained increase in SCL). Furthermore, limb movement and eye movement features in the task context can directly monitor the frequency, duration, and speed of attentional rebound of distracting events, and have higher discriminant value than other physiological features. Therefore, the weighting coefficients for each feature can be configured as follows: Verbal features: 0.14; Facial expression features: 0.14; Eye movement features: 0.16; Limb movement features: 0.16; EDA features: 0.22; Pulse wave features: 0.18.

[0091] In some embodiments, when the procrastination type is dominated by emotion regulation deficiency, the maximum difference between the first weighting coefficient and the second weighting coefficient is greater than a third preset threshold.

[0092] It should be understood that the core discriminative features in the emotion regulation deficit-dominated type are skin conductance activity and persistent pulse wave, which are assigned high weights accordingly. For example, a sustained, drifting increase in SCL is the gold standard indicator of accumulated emotional load. Simultaneously, a progressive decrease in high-frequency power of heart rate variability (i.e., the HF index of HRV) reflects the decline in emotional recovery function. Micro-expressions and phonological prosody can reflect the capture of aversion and distress-related micro-expressions (such as AU4 activity of the corrugator supercilii muscle) and negative vocal events such as sighing and complaining during the task, and have higher discriminative value than other physiological features. Therefore, the weight coefficients of each feature can be configured as follows: phonological features: 0.14; facial expression features: 0.14; eye movement features: 0.11; limb movement features: 0.11; skin conductance features: 0.25; pulse wave features: 0.25.

[0093] In some embodiments, after S103, the method further includes: obtaining scene information in the current application scenario; and evaluating the data quality score of each feature item based on the scene information.

[0094] The scenario information includes at least one of the following: user schedule status (such as meeting, study, or free time), data collection permission status (such as the activation status of camera, microphone, or physiological sensor), and environmental noise level (such as background sound intensity detected by microphone).

[0095] Based on scenario information within the current application context, the data reliability of each feature item is evaluated, generating a corresponding data quality score. For example, in a high-noise environment, the data quality score for voice features decreases; when the schedule status is "in a meeting" or microphone access is disabled, voice and facial expression feature collection may involve user privacy or cannot be collected, resulting in a corresponding decrease in the data quality score. Subsequently, the weight coefficient of the corresponding feature item is reduced based on this data quality score, such as by 0.1, to obtain the updated weight coefficient. This effectively prevents misjudgments and over-intervention caused by unreliable data dominating the evaluation, improving reliability in complex real-world environments.

[0096] In some embodiments, each feature item is collected and analyzed at a corresponding first preset period to update parameter values, thereby updating the intervention intensity parameter. When the weight coefficient of any feature item is greater than or equal to a first coefficient threshold, the collection and analysis period of the corresponding feature item is adjusted to a second preset period, which is shorter than the first preset period. When the updated weight coefficient of any feature item is less than the second coefficient threshold, the collection and analysis period of the corresponding feature item is adjusted to a third preset period, which is longer than the first preset period. Thus, while enhancing the accuracy of key feature perception, the system resource overhead and user device burden are effectively reduced.

[0097] In some embodiments, if the data quality score of a feature item remains below a preset score threshold for a first preset time period, the collection period is maintained without shortening, and the weight coefficient of the corresponding feature item is adjusted according to the data quality score to update the weight coefficient. This hierarchical linkage adjustment mechanism between the collection period and the weight coefficient ensures the reliability of the evaluation while avoiding high-frequency collection or misjudgment interventions triggered by invalid data, thereby reducing disruption to users.

[0098] In some embodiments, since the intervention intensity parameters are updated frequently in real time, directly triggering prompts based on each update can easily lead to frequent and fragmented interventions, causing user discomfort, psychological resistance, or interruption of attention. To avoid such excessive disruption, a response interval for the intervention strategy is preset. Within this time window, multiple intervention intensity parameters are temporarily stored and analyzed. Only at the end of the interval is an optimized intervention strategy output after comprehensive evaluation. This significantly reduces the frequency of prompts while ensuring the timeliness of intervention, thereby improving user experience and intervention effectiveness.

[0099] In some embodiments, to balance the timeliness of intervention with user experience, differentiated response intervals are configured for different types of intervention strategies. Specifically, after each intervention strategy is executed, the response interval for the next intervention strategy is dynamically determined based on the intervention strategy executed this time. If no intervention strategy is executed within a second preset time period (e.g., one hour), a general value (e.g., 3 minutes) is switched to apply. For example, low-intensity prompts correspond to shorter response intervals (e.g., 30 seconds) to quickly respond to changes in status; while high-intensity interventions trigger longer response intervals (e.g., 5 minutes) to avoid continuous strong interventions that could cause user aversion or cognitive overload. The specific value of the response interval can be flexibly configured according to the actual application scenario, user preferences, or historical interaction feedback.

[0100] In some embodiments, multiple intervention intensity parameters are acquired within the response interval; based on a preset hierarchical mapping standard, each intervention intensity parameter is mapped to a corresponding candidate intervention strategy. The hierarchical mapping standard can map intervention intensity parameters in different value ranges to corresponding intervention strategies. For example, the hierarchical mapping standard for intervention intensity parameters is [0, 0.3], (0.3, 0.7], and (0.7, 1], corresponding to mild, moderate, and severe deficiency levels, respectively. The daily training module will correspond to low, medium, and high intensity intervention strategies.

[0101] In some embodiments, if the number of occurrences of any candidate intervention strategy within the response interval reaches a preset minimum, the candidate intervention strategy with the highest occurrence frequency is selected as the target intervention strategy; if the number of occurrences of all candidate intervention strategies does not reach the preset minimum, the target intervention strategy is selected from the candidate intervention strategies according to a preset joint mapping relationship between the response gain system and the hierarchical mapping standard. The preset minimum number is determined based on the response interval duration and the user's procrastination type, and is not limited here.

[0102] In some embodiments, two types of joint mapping relationships are used to coordinate decision-making intervention strategies, dynamically adapting the appropriateness of intervention in the current scenario, and achieving personalized configuration, contextualized response, and low-intrusion implementation of intervention strategies.

[0103] For example, one type of joint mapping relationship includes: comparing the current response gain coefficient with a preset response threshold to determine the user's current level of acceptance of the intervention; establishing a joint mapping relationship between different comparison results and strategy selection tendencies, including: when the response gain coefficient is greater than the preset response threshold, the user's current level of acceptance of the intervention is high, and the strategy selection tendency is "enabling a prompt with a higher level of intervention"; when the response gain coefficient is less than or equal to the preset response threshold, the user's current level of acceptance of the intervention is low, and the strategy selection tendency is "enabling a prompt with a lower level of intervention"; wherein, the preset response threshold can be dynamically determined based on the user's historical interaction feedback, task type, and physiological state.

[0104] For example, the intervention intensity parameter can be divided into multiple intervals according to a preset grading mapping standard, each corresponding to a different degree of delay defect (e.g., mild, moderate, severe), and associated with corresponding intervention strategies. Correspondingly, a two-level joint mapping relationship is established between the response gain coefficient and the grading mapping standard. Subsequently, recommended intervention strategies can be directly mapped and output based on the response gain coefficient. For instance, a mapping relationship is established between the [0, 0.3] interval of the grading mapping standard (corresponding to mild defects, suitable for low-intensity prompts) and the [0.1, 0.4] interval of the response gain coefficient; between the (0.3, 0.7] interval of the grading mapping standard (corresponding to moderate defects, suitable for moderate-intensity prompts) and the (0.4, 0.8] interval of the response gain coefficient; and between the (0.7, 1] interval of the grading mapping standard (corresponding to severe defects, suitable for high-intensity prompts) and the (0.8, 1] interval of the response gain coefficient.

[0105] For example, within a preset response interval, two low-intensity cues, three medium-intensity cues, and one high-intensity cue are generated. If the preset minimum number of cues is three, the medium-intensity cue is applied; if the preset minimum number of cues is four, no strategy satisfies the stable occurrence condition. In this case, if the current response gain coefficient is 0.5, and according to the preset response threshold of 0.6 in a first-class joint mapping relationship, the low-intensity cue can be selected as the final intervention strategy. According to a second-class joint mapping relationship, the medium-intensity cue is preferentially selected as the final intervention strategy.

[0106] In some embodiments, a two-class joint mapping relationship can be preferentially applied to determine the applicable recommended intervention strategy; if the recommended intervention strategy exists in the candidate strategy set generated within the response interval, the recommended intervention strategy is directly selected as the target intervention strategy; if it does not exist, a one-class joint mapping relationship is applied to determine the strategy selection tendency; the candidate intervention strategy that is closest to the recommended intervention strategy in terms of cue intensity and conforms to the strategy selection tendency is selected from the candidate strategy set as the final target intervention strategy.

[0107] For example, when the response gain coefficient is 0.5, based on the two-class joint mapping relationship, the recommended intervention strategy is a medium-intensity prompt. If a medium-intensity prompt option is generated within the preset response interval, it is adopted directly; if only no prompt, low-intensity prompt, and high-intensity enhancement options are generated within the preset response interval, then based on the preset response threshold of 0.6 in the first-class joint mapping relationship, it tends to "enable a prompt with a lower level of intervention". The low-intensity prompt, which is closest to the medium-intensity prompt and meets the requirement of "enabling a prompt with a lower level of intervention", can be selected as the final intervention strategy.

[0108] In some embodiments, the intervention intensity parameter The calculation formula can be: ; in, For a specific moment; The first time at time t One feature term; For the first Measured values ​​of each feature term; For the first The weight coefficients of each feature term; For response gain coefficient; This serves as the user's normative baseline value.

[0109] In some embodiments, the normalization process includes: normalizing the intervention intensity parameter using a sigmoid function (such as the sigmoid function) to map it to the (0,1) interval. This method is simple to calculate, requires no storage of historical data, and produces stable output. The parameters of this function can be set based on prior data, ensuring that the normalized values ​​have stable theoretical boundaries and comparability. This is particularly suitable for scenarios in the early stages of data accumulation or in situations with high privacy requirements where long-term storage of personal historical data is not desired.

[0110] For example, intervention intensity parameter The normalization formula can be: ; in, The normalized intervention intensity parameter.

[0111] In other embodiments, the normalization process includes: obtaining the user's historical intervention intensity parameters within a sliding time window (e.g., the last 7 days), the historical intervention intensity parameters including the historical maximum and minimum values ​​of the user's intervention intensity parameters within the sliding time window; using the historical maximum and minimum values ​​as dynamic boundaries to linearly map the current intervention intensity parameters to the [0, 1] interval; if historical data is insufficient, normalization is performed using preset theoretical boundaries (e.g., upper and lower limits set by expert experience).

[0112] Therefore, it can sensitively capture relative changes in a user's own state that deviate from the norm, better adapt to the baseline differences among different users, and achieve highly personalized strategy responses, which is especially suitable for long-term personalized health management and monitoring of gradual changes. For example, a slight increase in the physiological indicators of a user who is usually emotionally stable may be mapped to a higher ΔI(t), thus receiving timely attention.

[0113] For example, intervention intensity parameter The normalization formula can be: ; in, and These represent the historical maximum and minimum values ​​of the dynamic changes in intervention intensity, respectively.

[0114] Please see Figure 3 , Figure 3 This is a schematic flowchart illustrating yet another method for managing procrastination provided in an embodiment of this application. Figure 3 As shown, the procrastination management method provided in this application is constructed based on multi-source heterogeneous multimodal user data, providing support for the accurate quantification of intervention intensity. It not only incorporates the assessment results of standardized psychological scales but also simultaneously collects multi-dimensional behavioral characteristics (such as voice, eye movement, and body movements) and physiological characteristics (such as skin conductance and pulse wave) to determine the user's deficiencies in predictive imagination, self-control, and emotional regulation abilities from multiple dimensions.

[0115] Based on this, objective, quantitative, and traceable intervention intensity is generated by using interpretable rules for calculating the overall deviation of user status (such as the calculation and normalization formula of intervention intensity parameters).

[0116] In the calculation of the overall deviation, the feature weights are adjusted. This enables refined and differentiated applications of multi-source heterogeneous data, significantly improving the accuracy and relevance of intervention intensity quantification.

[0117] First, feature weights are dynamically adjusted according to the type of procrastination. Procrastination is subdivided into primary types and more refined secondary subtypes. Different types of procrastination mechanisms have varying degrees of dependence on behavioral and physiological characteristics, and differentiated weights are assigned accordingly: for example, initiation difficulty emphasizes behavioral characteristics, while execution difficulty emphasizes physiological characteristics. More importantly, the polarization of the weights matches the classification granularity. The finer the classification granularity, the more focused the weights. For example, balanced weights are used in primary types to ensure robustness, while polarized weights are used in secondary subtypes to strengthen the application of key features. This truly achieves targeted treatment, significantly improving the accuracy of procrastination identification and the effectiveness of interventions.

[0118] Secondly, feature weights are dynamically adjusted based on data quality. When the credibility of a feature decreases due to environmental interference (such as a noisy background affecting speech recognition) or permission restrictions (such as the camera being turned off), its weight is automatically reduced to avoid misjudgment and over-intervention caused by noisy data, thereby enhancing decision robustness and preventing user anxiety or resistance.

[0119] In addition, a response gain coefficient is introduced based on scene-aware intervention acceptance. This approach allows for differentiated intervention intensities for the same type of procrastination and the same overall degree of deviation. It provides a flexible response strategy that aligns with the user's current situation and preferences, effectively avoiding inappropriate interruptions, fully respecting individual autonomy, protecting the user's intrinsic motivation, and preventing the resistance or additional psychological burden that traditional coercive interventions might induce. For example, for the same intervention intensity, voice prompts or interface guidance can be enabled in high-acceptance scenarios (such as self-study periods), while only slight vibrations or delayed reminders are triggered in low-tolerance scenarios (such as meetings or driving).

[0120] Finally, based on the real-time generated intervention intensity parameters, the intervention strategy corresponding to the current procrastination type is invoked. This strategy is highly adapted to the user's individualized procrastination causes, real-time psychological and physiological state, and current application scenario.

[0121] This application embodiment also provides a method for managing procrastination, the method comprising: acquiring multimodal user data; the multimodal user data including: test results of a preset psychological scale, multiple behavioral and physiological features updated based on preset time intervals; identifying the user's procrastination type based on the multimodal user data, including: determining the classification granularity of the procrastination type according to the user data and / or the system activation stage; identifying the user as a first-level procrastination type, or a second-level procrastination type further subdivided under the first-level procrastination type, according to the classification granularity; setting a first weight coefficient for the behavioral features and a second weight coefficient for the physiological features according to the procrastination type, wherein the finer the classification granularity, the greater the weight difference within the same type of feature and / or the greater the weight difference between different types of features. The greater the difference in weights between the features; calculate the deviation between the measured current value of each feature and the baseline value of the feature, and determine the overall deviation based on the deviation of each feature and its corresponding weight coefficient; set the response gain coefficient according to the level of intervention acceptance in the current application scenario; determine the intervention intensity parameter based on the response gain coefficient and the overall deviation; based on the intervention intensity parameter, call the intervention strategy corresponding to the current procrastination type, including: normalizing the intervention intensity parameter, and applying different hierarchical mapping standards to the normalized intervention intensity parameter according to the classification granularity; wherein, a balanced hierarchical strategy is used under the first-level procrastination type; and a hierarchical strategy referencing a normal distribution is switched under the second-level procrastination type.

[0122] It should be understood that, based on the current depth of understanding of the causes of user procrastination, the classification granularity is dynamically set, and an adaptive mechanism is implemented accordingly: for the same multimodal user data, different feature weight configurations are selectively invoked according to the classification granularity; for the same normalized intervention intensity parameter, it is selectively mapped to different intervention strategy grading standards according to the classification granularity. It should be noted that the classification granularity in the embodiments of this application is not limited to first-level procrastination type and second-level procrastination type. Second-level procrastination type can be further subdivided into third-level procrastination type, thereby achieving refined procrastination management.

[0123] Please see Figure 4 , Figure 4 This is a schematic block diagram of a procrastination management system provided in an embodiment of this application. The procrastination management system can be configured on a server to execute the aforementioned procrastination management method and its specific implementation steps.

[0124] like Figure 4 As shown in the illustration, this application also provides a procrastination management system 200, the system comprising: The data acquisition module 201 is used to acquire multimodal user data; the multimodal user data includes: test results of a preset psychological scale, and various behavioral and physiological features updated based on preset time intervals; Type assessment module 202 is used to identify the user's procrastination type based on the multimodal user data; The weight setting module 203 is used to set the first weight coefficient of the behavioral feature item and the second weight coefficient of the physiological feature item according to the procrastination type. The deviation calculation module 204 is used to calculate the degree of deviation between the measured current value of each feature item and the baseline value of the feature item, and to determine the overall deviation degree based on the degree of deviation of each feature item and the corresponding weight coefficient. The gain setting module 205 is used to set the response gain coefficient according to the intervention acceptance level in the current application scenario; and to determine the intervention intensity parameter based on the response gain coefficient and the overall deviation. The strategy invocation module 206 is used to invoke the intervention strategy corresponding to the current procrastination type based on the intervention intensity parameter.

[0125] For example, the procrastination management system 200 also includes: The scene information module is used to obtain scene information under the current application scenario. The scene information includes at least one of the following: user schedule status, collection permission status, and environmental noise level. The data quality module is used to evaluate the data quality score of each feature item based on the scenario information, and adjust the weight coefficient of the corresponding feature item according to the data quality score to update the weight coefficient.

[0126] For example, the procrastination type includes a first-level procrastination type, including at least one of initiation difficulty, execution difficulty, and mixed type; the procrastination management system 200 further includes: A type of weighting module is used to adjust the first weighting coefficient to be higher than the second weighting coefficient when the delay type is a difficult start type; The second type of weight module is used to adjust the second weight coefficient to be higher than the first weight coefficient when the delay type is execution difficulty type. Three weighting modules are used to determine whether the current dominant type is difficult to start or difficult to execute when the delay type is mixed, and to determine the weight coefficient of each feature item according to the current dominant type.

[0127] For example, in the first-level procrastination type, the maximum difference between the first weight coefficients of each behavioral feature item is less than or equal to a first preset threshold, the maximum difference between the second weight coefficients of each physiological feature item is less than or equal to a second preset threshold, and the maximum difference between the first weight coefficient and the second weight coefficient is less than or equal to a third preset threshold.

[0128] For example, the procrastination type further includes multiple secondary procrastination types subdivided under each primary procrastination type; in the secondary procrastination type, the maximum difference between the first weight coefficients of each behavioral feature item is greater than a first preset threshold; and / or, the maximum difference between the second weight coefficients of each physiological feature item is greater than a second preset threshold; and / or, the maximum difference between the first weight coefficient and the second weight coefficient is greater than a third preset threshold.

[0129] For example, the behavioral features include at least one of voice features, facial expression features, body movement features, and eye movement features; the physiological features include at least one of electrodermal features and pulse wave features.

[0130] For example, the secondary procrastination type in the initiation difficulty category includes at least one of intention formation difficulty and intention-action conversion difficulty. The procrastination management system 200 also includes: Four weighting modules are used to adjust the maximum difference between the first weighting coefficient and the second weighting coefficient to be greater than a third preset threshold when the procrastination type is the difficulty in forming the intention. Five weighting modules are used to adjust the maximum difference between the first weighting coefficient of the limb movement feature and the first weighting coefficient of the voice feature and facial expression feature when the procrastination type is the difficulty in converting intention into action.

[0131] For example, the secondary procrastination type in the executive difficulty category includes at least one of self-control deficit-dominated type and emotion regulation deficit-dominated type; the procrastination management system 200 further includes: Six weighting modules are used to adjust the maximum difference between the second weighting coefficient of the skin conductance feature and the second weighting coefficient of the pulse wave feature to be greater than a second preset threshold when the procrastination type is dominated by self-control deficiency. The seven weighting modules are used to adjust the maximum difference between the first weighting coefficient and the second weighting coefficient to be greater than a third preset threshold when the procrastination type is dominated by emotion regulation deficiency.

[0132] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the system and its modules and units described above can be referred to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0133] The methods and systems described in this application can be used in a wide variety of general-purpose or special-purpose computing system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer terminal devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices.

[0134] Please see Figure 5 , Figure 5 This is a schematic block diagram illustrating the structure of a computer device according to an embodiment of this application. The computer device may be a terminal device or a server.

[0135] For example, the above-described method or system can be implemented as a computer program, which can be used in, for example... Figure 5 The computer device shown in this application also provides a computer device, which includes: a memory for storing a computer program; and a processor for executing the computer program and implementing the steps and specific implementation steps of the procrastination management method provided in any embodiment of this application when executing the computer program.

[0136] like Figure 5 As shown, the computer device includes a processor, memory, and network interface connected via a system bus, wherein the memory may include non-volatile storage media and internal memory.

[0137] Non-volatile storage media can store operating systems and computer programs. These computer programs include program instructions that, when executed, cause the processor to perform any procrastination management method and its specific implementation steps.

[0138] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0139] Internal memory provides an environment for the execution of computer programs in non-volatile storage media. When the computer program is executed by the processor, it enables the processor to execute any procrastination management method and its specific implementation steps.

[0140] This network interface is used for network communication, such as sending assigned tasks.

[0141] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be 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. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0142] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps: S101, acquire multimodal user data; the multimodal user data includes: test results of a preset psychological scale, and various behavioral and physiological features updated based on preset time intervals; S102, Identify the user's procrastination type based on the multimodal user data; S103, according to the procrastination type, set the first weight coefficient of the behavioral characteristic item and the second weight coefficient of the physiological characteristic item respectively; S104, calculate the degree of deviation between the measured current value of each feature item and the baseline value of the feature item, and determine the overall degree of deviation based on the degree of deviation of each feature item and the corresponding weight coefficient; S105, Set the response gain coefficient according to the level of intervention acceptance in the current application scenario; Determine the intervention intensity parameter based on the response gain coefficient and the overall deviation. S106, Based on the intervention intensity parameter, invoke the intervention strategy corresponding to the current procrastination type.

[0143] For example, the processor is used to run a computer program stored in the memory, and is also used to implement the steps and specific implementation steps of the procrastination management method provided in any embodiment of this application, which will not be repeated here.

[0144] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program causes the processor to implement the steps and specific implementation steps of the procrastination management method provided in any embodiment of this application. For example, the computer program includes program instructions, and the processor executes the program instructions to implement the steps and specific implementation steps of the procrastination management method provided in any embodiment of this application.

[0145] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0146] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered 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.

Claims

1. A method for managing procrastination, characterized in that, The method includes: S101, acquire multimodal user data; the multimodal user data includes: test results of a preset psychological scale, and various behavioral and physiological features updated based on preset time intervals; S102, Identify the user's procrastination type based on the multimodal user data; S103, according to the procrastination type, set the first weight coefficient of the behavioral characteristic item and the second weight coefficient of the physiological characteristic item respectively; S104, calculate the degree of deviation between the measured current value of each feature item and the baseline value of the feature item, and determine the overall degree of deviation based on the degree of deviation of each feature item and the corresponding weight coefficient; S105, Set the response gain coefficient according to the level of intervention acceptance in the current application scenario; Determine the intervention intensity parameter based on the response gain coefficient and the overall deviation. S106, Based on the intervention intensity parameter, invoke the intervention strategy corresponding to the current procrastination type.

2. The method according to claim 1, characterized in that, The intervention strategies include cognitive behavioral training strategies and immediate feedback prompting strategies; The method includes: Configure at least one training parameter in the cognitive behavior training strategy according to the intervention intensity parameter; The training parameters include: training format, course content, execution frequency, execution intensity, and duration of a single execution; the training format includes at least one of online interactive courses and neuromodulation training. Configure at least one prompting parameter in the instant feedback prompting strategy according to the intervention intensity parameter; The prompt parameters include: prompt modality, prompt intensity, or presentation mode; the prompt modality includes at least one of vibration, audio prompt, or graphic pop-up; the prompt intensity is adjusted based on at least one of vibration frequency, audio volume and rhythm, or graphic pop-up size and duration.

3. The method according to claim 1, characterized in that, The procrastination types include first-level procrastination types, including at least one of the following: difficulty in initiation, difficulty in execution, and mixed types. The method includes: When the delay type is a difficult start type, the first weighting coefficient is higher than the second weighting coefficient; When the delay type is execution difficulty type, the second weighting coefficient is higher than the first weighting coefficient; When the delay type is a mixed type, the current dominant type is determined to be either start-up difficulty type or execution difficulty type, and the weight coefficient of each feature item is determined according to the current dominant type.

4. The method according to claim 3, characterized in that, In the first-level procrastination type, the maximum difference between the first weight coefficients of each behavioral feature item is less than or equal to a first preset threshold, the maximum difference between the second weight coefficients of each physiological feature item is less than or equal to a second preset threshold, and the maximum difference between the first weight coefficient and the second weight coefficient is less than or equal to a third preset threshold.

5. The method according to claim 1, characterized in that, The procrastination type also includes multiple secondary procrastination types subdivided under each primary procrastination type; in the secondary procrastination type, the maximum difference between the first weight coefficients of each behavioral feature item is greater than a first preset threshold. And / or, the maximum difference between the second weighting coefficients of each physiological characteristic item is greater than the second preset threshold; And / or, the maximum difference between the first weighting coefficient and the second weighting coefficient is greater than a third preset threshold.

6. The method according to any one of claims 1 to 5, characterized in that, The behavioral features include at least one of the following: voice features, facial expression features, limb movement features, and eye movement features; the physiological features include at least one of the following: skin conductance features and pulse wave features.

7. The method according to claim 6, characterized in that, The secondary procrastination types within the initiation difficulty category include at least one of intention formation difficulty and intention-action conversion difficulty. The method includes: When the procrastination type is "difficult to form intention", the maximum difference between the first weight coefficient and the second weight coefficient is greater than the third preset threshold. When the procrastination type is "difficulty in converting intention into action", the maximum difference between the first weight coefficient of the limb movement feature and the first weight coefficient of the voice feature and facial expression feature is greater than the first preset threshold.

8. The method according to claim 6, characterized in that, The second-level procrastination type in the executive difficulty category includes at least one of the self-control deficit-dominated type and the emotion regulation deficit-dominated type; When the procrastination type is dominated by self-control deficiency, the maximum difference between the second weighting coefficient of the skin conductance feature and the second weighting coefficient of the pulse wave feature is greater than the second preset threshold. When the procrastination type is dominated by emotion regulation deficiency, the maximum difference between the first weighting coefficient and the second weighting coefficient is greater than the third preset threshold.

9. The method according to claim 1, characterized in that, Following S103, the following is also included: Obtain scenario information for the current application scenario, including at least one of user schedule status, data collection permission status, and environmental noise level; The data quality score of each feature item is evaluated based on the scenario information, and the weight coefficient of the corresponding feature item is adjusted according to the data quality score to update the weight coefficient.

10. A procrastination management system, characterized in that, The system includes: The data acquisition module is used to acquire multimodal user data, which includes: test results of a preset psychological scale, and various behavioral and physiological features updated based on preset time intervals. The type identification module is used to identify the user's procrastination type based on the multimodal user data; The weight setting module is used to set the first weight coefficient of the behavioral feature item and the second weight coefficient of the physiological feature item according to the procrastination type. The deviation calculation module is used to calculate the degree of deviation between the measured current value of each feature item and the baseline value of the feature item, and to determine the overall deviation degree based on the degree of deviation of each feature item and the corresponding weight coefficient. The gain setting module is used to set the response gain coefficient according to the intervention acceptance level in the current application scenario; and to determine the intervention intensity parameter based on the response gain coefficient and the overall deviation. The strategy invocation module is used to invoke the intervention strategy corresponding to the current procrastination type based on the intervention intensity parameter.

Images