Task-focused intervention method, device, equipment, storage medium and program product

Abstract

This application provides a method, apparatus, device, storage medium, and program product for intervening in focused tasks. The method includes: acquiring the start time, end time, and current time of the current focused task; acquiring a sequence of physiological state characteristics of the user; the physiological state characteristic sequence includes various physiological state characteristics of the user from the start time to the current time; determining the user's flow state recognition result based on the physiological state characteristic sequence; predicting the user's cognitive exhaustion time based on the physiological state characteristic sequence; and adjusting the end time of the current focused task according to the flow state recognition result and the cognitive exhaustion time. This method allows for flexible intervention in the user's learning process.

Description

Technical Field

[0001] This application relates to the field of human-computer interaction technology, and in particular to a focused task intervention method, device, equipment, storage medium, and program product. Background Technology

[0002] With the development of information technology, various digital learning tools, such as the Pomodoro Technique, focus timers, and task management software, have been widely used in individuals' learning and work processes.

[0003] Current digital learning tools often assist users in time planning through preset time blocks and task lists, relying on users' subjective self-control to execute learning and work plans. However, individual states during the learning process are constantly changing, and these tools cannot adaptively adjust to users' learning and work plans based on individual states. Their intervention methods (such as rest reminders) are often relatively fixed. For example, when users are actually fatigued or distracted, they cannot intervene in a timely manner. When users are in a highly focused "flow" state, fixed rest reminders may actually interrupt their efficient learning process.

[0004] Therefore, traditional digital learning tools suffer from the problem of being unable to flexibly intervene in the user's learning process. Summary of the Invention

[0005] The purpose of this application is to at least solve one of the above-mentioned technical defects, in particular the technical defect that existing digital learning tools cannot flexibly intervene in the user's learning process. This application provides a focus task intervention method, device, equipment, storage medium and program product that can flexibly intervene in the user's learning process.

[0006] Firstly, this application provides a focused task intervention method, including:

[0007] Obtain the start time, end time, and current time of the currently focused task, and obtain the user's physiological state feature sequence; the physiological state feature sequence includes the user's physiological state features from the start time to the current time;

[0008] Based on the physiological state feature sequence, the user's flow state recognition result is determined, and based on the physiological state feature sequence, the user's cognitive exhaustion moment is predicted.

[0009] Based on the results of flow state recognition and the time of cognitive exhaustion, the end time of the current focused task is adjusted.

[0010] In one embodiment, the end time of the current focused task is adjusted based on the flow state recognition result and the time of cognitive exhaustion, including:

[0011] If the flow state recognition result indicates that the user is currently in a flow state, extend the end time;

[0012] If the flow state recognition result indicates that the user is not currently in a flow state and the time of cognitive exhaustion is earlier than the time of termination, the time of termination will be adjusted to be before the time of cognitive exhaustion.

[0013] If the flow state recognition result indicates that the user is not currently in a flow state and the end time is earlier than the cognitive exhaustion time, the end time is kept unchanged.

[0014] In one embodiment, physiological state features include attention features, blinking features, and heart rate variability features. Based on the sequence of physiological state features, the user's flow state recognition result is determined, including:

[0015] Based on the heart rate change characteristics, focus characteristics, and blinking characteristics in the physiological state characteristic sequence, determine whether the user currently meets the preset flow state judgment conditions;

[0016] If the user meets the preset criteria for determining the flow state, then the user is determined to be in a flow state.

[0017] In one embodiment, predicting the user's cognitive exhaustion moment based on a sequence of physiological state features includes:

[0018] The physiological state feature sequence is input into each pre-trained cognitive state recognition model to obtain the cognitive state index sequence; the cognitive state index sequence includes the user's cognitive state indicators from the start time to the current time.

[0019] The cognitive state index sequence is input into a pre-trained cognitive exhaustion point prediction model. Based on the user's fatigue curve, the user's cognitive exhaustion point prediction model is used to determine the user's cognitive exhaustion moment.

[0020] In one embodiment, the cognitive state indicators include attention indicators, and the method further includes:

[0021] Determine the user's current distraction duration based on the user's focus index from the start time to the current time;

[0022] Based on the current duration of distraction, determine the intervention strategy for the user's current attention.

[0023] According to the attention intervention strategy, intervene in the user's current attention.

[0024] In one embodiment, cognitive state indicators include fatigue indicators, and the method further includes:

[0025] Based on the user's fatigue index from the start time to the current time, determine whether the user is currently in a state of fatigue;

[0026] If it is determined that the user is in a state of fatigue, the rest duration that matches the fatigue index at the current moment is determined based on the user's fatigue decline curve.

[0027] Based on the rest duration, a rest suggestion message is generated and pushed to the user's device.

[0028] In one embodiment, prior to the steps of obtaining the start time, end time, and current time of the current focused task, the method further includes:

[0029] In response to a task parameter setting operation for the currently focused task, retrieve the task type and start time set in the task parameter setting operation;

[0030] Extract the session data corresponding to each historical focused task belonging to the task type from the user's session database, and determine the predicted focus duration value for the current focused task based on the session data of each historical focused task.

[0031] The end time is determined based on the predicted focus duration and start time.

[0032] In one embodiment, the session data for historical focus tasks includes the user's effective focus time in historical focus tasks. Based on the session data for each historical focus task, a predicted focus time value for the current focus task is determined, including:

[0033] Determine the time decay weight coefficient and focus quality weight coefficient for each historical focus task, and based on the time decay weight coefficient and focus quality weight coefficient for each historical focus task, determine the comprehensive weight coefficient for each historical focus task.

[0034] The effective focus duration of each historical focus task is weighted and summed using the focus quality weight coefficient of each historical focus task to obtain the first predicted value of focus duration. The effective focus duration of each historical focus task is weighted and summed using the comprehensive weight coefficient of each historical focus task to obtain the second predicted value of focus duration.

[0035] The predicted focus duration is determined based on the first and second predicted focus duration values.

[0036] In one embodiment, the method further includes:

[0037] Acquire user session data after completing a focused task; session data should include at least timestamps and physiological metrics.

[0038] Based on the session data of completed focused tasks, extract the user's state feature vector during the task completion process of the completed focused tasks;

[0039] Based on preset mapping rules, virtual object visualization parameters that match the state feature vector are selected.

[0040] Based on the visualization parameters of virtual objects, virtual objects corresponding to the completed focus tasks are generated and displayed in the virtual space; the virtual form of the virtual objects in the virtual space represents the user's physiological state level after completing the focus task.

[0041] In one embodiment, the method further includes:

[0042] In response to a click operation on a virtual object in the virtual space, determine the target virtual object clicked by the virtual object click operation;

[0043] The virtual object displays its morphological change history during the execution of the corresponding target-focused task in the virtual space; the morphological change history is generated based on the session data of the target-focused task; the morphological change history represents the fluctuation of the user's physiological state level during the execution of the target-focused task.

[0044] Secondly, this application provides a focus task intervention device, comprising:

[0045] The acquisition module is used to acquire the start time, end time, and current time of the currently focused task, as well as to acquire the user's physiological state feature sequence; the physiological state feature sequence includes the user's physiological state features from the start time to the current time.

[0046] The identification module is used to determine the user's flow state identification result based on the physiological state feature sequence, and to predict the user's cognitive exhaustion moment based on the physiological state feature sequence.

[0047] The intervention module is used to adjust the end time of the current focused task based on the results of flow state identification and the time of cognitive exhaustion.

[0048] Thirdly, this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0049] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0050] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0051] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0052] The focus task intervention method, device, equipment, storage medium, and program product provided in this application acquire the start time, end time, and current time of the current focus task, as well as the user's physiological state feature sequence. The physiological state feature sequence includes various physiological state characteristics of the user from the start time to the current time. Based on the physiological state feature sequence, the user's flow state recognition result is determined, and the user's cognitive exhaustion time is predicted based on the physiological state feature sequence. The end time of the current focus task is adjusted according to the flow state recognition result and the cognitive exhaustion time. Thus, by analyzing the physiological state feature sequence, the user's flow state recognition result is accurately determined, and the cognitive exhaustion time is accurately predicted, thereby achieving automatic adjustment of the end time of the current focus task. This allows for flexible management of personalized learning time based on the user's real-time state, making intervention more precise and timely, reducing the user's decision-making burden, and further improving the user's learning efficiency. Attached Figure Description

[0053] 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. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A flowchart illustrating a focus task intervention method provided in an embodiment of this application;

[0055] Figure 2 This is a flowchart illustrating a focused task intervention method in another embodiment;

[0056] Figure 3 This is a structural block diagram of a focus task intervention device in one embodiment;

[0057] Figure 4 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0058] 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 embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0059] Existing learning applications (such as Pomodoro timers, focus timers, and task management tools) primarily rely on user self-control and cannot perceive learning states (focus, fatigue, emotions, etc.) in real time. While existing EEG devices can measure certain attention / emotion indicators, their applications are mainly concentrated in experimental research and basic psychology training, lacking application models deeply integrated with learning tasks. Therefore, existing technologies have the following problems: First, they rely on subjective reports; the judgment of a user's focus state depends entirely on the user starting and pausing the timer themselves, failing to objectively and quantitatively reflect the user's true cognitive state. Second, they lack real-time intervention; applications can only passively record interruptions after the user becomes distracted, unable to proactively perceive and intervene in real-time when the user is distracted, fatigued, or experiencing emotional fluctuations. Third, they lack personalized learning models and dynamic planning capabilities; user focus tasks are typically timed according to a fixed 25-minute or 45-minute Pomodoro timer, not changing with the user's real-time cognitive state.

[0060] Therefore, there is an urgent need for a comprehensive learning system that combines personalized learning plans based on the user's physiological state with immersive incentives. This application provides an intelligent focus management method based on real-time EEG perception and long-term learning profiles. It can detect the user's focus, fatigue, and emotional state in real time; assess future cognitive decline trends using time-series prediction models; provide dynamic, closed-loop real-time intervention strategies; introduce visual incentives centered on the growth of virtual flowers; build personal efficiency profiles based on the user's historical conversation data; and automatically identify optimal learning periods to optimize learning plans and task sequences. Please refer to the detailed descriptions of the following embodiments for specifics.

[0061] In one exemplary embodiment, such as Figure 1 As shown, a focused task intervention method is provided. Taking the application of this method to a server as an example, the method includes the following steps S102 to S106. Wherein:

[0062] Step S102: Obtain the start time, end time, and current time of the current focused task, and obtain the user's physiological state feature sequence; the physiological state feature sequence includes the user's physiological state features from the start time to the current time.

[0063] The current focus task refers to the learning task that the user is currently undertaking. In a learning app, this can refer to the timer that the user is using during the current learning process.

[0064] The start time of the current focused task refers to the start time of the timer.

[0065] The end time of the current focused task refers to the end time of the timer. The end time can be the end time set by the user when setting the learning timer, or it can be the end time after the system adaptively adjusts as the user's cognitive state changes.

[0066] In this context, the current moment of the currently focused task refers to the current timing moment of the timer.

[0067] Physiological state characteristics include at least electroencephalogram (EEG) state characteristics and heart rate state characteristics. EEG state characteristics include power spectral density and blink frequency, while heart rate state characteristics include heart rate variability (HRV).

[0068] In practical applications, brainwave state vectors can be constructed based on brainwave state features. ,in, , These represent the power in the alpha, beta, and theta frequency bands, respectively. The ratio of Beta waves to Alpha waves is commonly used to assess attention levels, while BlinkRate represents the blink rate. Physiological state feature vectors can also be constructed based on EEG and heart rate characteristics.

[0069] Among them, the physiological state feature sequence is a sequence of features that can quantitatively represent the user's instantaneous physiological state, obtained by calculating and analyzing the original physiological signal sequence, and arranged in chronological order. The physiological state feature sequence can be composed of physiological state feature vectors corresponding to each time from the start time to the current time.

[0070] In practical applications, the acquired EEG signal sequences can be processed by bandpass filtering and artifact recognition (EMG, EOG) to obtain EEG state feature sequences, which can then be used as physiological state feature sequences.

[0071] Optionally, the server receives physiological signal data uploaded in real time from user terminals or sensing devices, aligns it by timestamp to form a continuous physiological signal sequence, and then uses feature extraction algorithms (e.g., calculating the power spectral density of EEG, heart rate variability HRV, blink frequency, etc.) to calculate physiological state features from the physiological signal sequence in time intervals or in real time, and arranges them in chronological order to form a physiological state feature sequence.

[0072] The aforementioned physiological signal sequence refers to the raw time-series data stream that reflects the user's physiological state, continuously collected by wearable devices or biosensors (such as electroencephalography (EEG), electrooculography (EOG), and electrocardiography (ECG / PPG), including signal sequences such as brain waves, eye movements, and heart rate.

[0073] In practical applications, wearable EEG devices can be used to collect users' EEG signal sequences. During collection, user behavior data (e.g., time period, subject, task type, operational events, etc.) and subjective evaluation data are collected simultaneously. Task types can be specific learning tasks. In practice, task types can be reading / input (e.g., browsing literature, watching online course videos), memory / recitation (e.g., memorizing vocabulary, reciting texts), logic / problem-solving (e.g., solving math problems, writing code), and creation / output (e.g., writing essays, drawing sketches). Operational events refer to the interactive behavior data generated by the user during application use, used to help determine the user's learning intentions and subjective state. These can include: starting / ending a task, pausing / resuming (e.g., going to the restroom) or taking a break as suggested by the system; setting goals (the preset learning duration entered by the user before starting), switching the current learning subject; feedback and evaluation events, such as scoring task completion after completion; and abnormal / interference events, such as app screen switching interactions, turning on the screen, or viewing collected data. Before starting a task, users set a learning time goal and select the corresponding subject. After completion, they subjectively score the task completion rate (1-100%), thus obtaining subjective evaluation data from users.

[0074] The physiological state feature sequence obtained through step S102 can provide an objective and real-time data foundation for subsequent analysis.

[0075] Step S104: Based on the physiological state feature sequence, determine the user's flow state recognition result, and based on the physiological state feature sequence, predict the user's cognitive exhaustion time.

[0076] The result of flow state recognition can indicate whether the user is currently in a flow state or not.

[0077] Cognitive exhaustion refers to the critical point in time when a user's cognitive performance will significantly decline or become unsustainable.

[0078] Optionally, the server identifies whether the user is currently in a flow state based on the user's physiological state feature sequence, and predicts the user's cognitive exhaustion moment based on the user's physiological state feature sequence.

[0079] Step S106: Adjust the end time of the current focused task based on the flow state recognition results and the time of cognitive exhaustion.

[0080] Optionally, the server determines whether the end time of the current focused task should be extended, brought forward, or kept unchanged, based on whether the user is currently in a flow state and the predicted time of cognitive exhaustion.

[0081] In step S106, based on the flow state recognition results and the moment of cognitive exhaustion, the task duration is dynamically adjusted, breaking the rigid and fixed task time pattern in traditional learning tools. This allows the arrangement of learning tasks to flexibly adapt to the user's real-time cognitive capacity, appropriately extending the efficient learning period when the user is in a good state and stopping it in time when the state declines, thereby maximizing the effective learning time and improving the overall efficiency and experience of the learning process.

[0082] The aforementioned focused task intervention method acquires the start time, end time, and current time of the current focused task, as well as the user's physiological state feature sequence. This sequence includes various physiological state characteristics of the user from the start time to the current time. Based on this sequence, the method determines the user's flow state identification result and predicts the user's cognitive exhaustion time. The end time of the current focused task is adjusted according to the flow state identification result and the cognitive exhaustion time. Thus, by analyzing the physiological state feature sequence, the method accurately determines the user's flow state identification result and accurately predicts the cognitive exhaustion time, thereby achieving automatic adjustment of the end time of the current focused task. This allows for flexible management of personalized learning time based on the user's real-time state, making intervention more precise and timely. It also helps reduce the user's decision-making burden and further improves the user's learning efficiency.

[0083] Flow is a special physiological and psychological state. When a user enters a flow state, although the time may have reached the "cognitive exhaustion point" predicted by the cognitive exhaustion point prediction model, the user's actual brain performance is at its peak. The following embodiment provides a specific flow protection method that can extend the end time in a short period of time on a trial basis. It should be noted that the flow state and the cognitive exhaustion point do not conflict. The flow state can resolve the conflict between prediction and reality.

[0084] In one exemplary embodiment, the end time of the current focused task is adjusted based on the flow state recognition result and the cognitive exhaustion time, including: extending the end time when the flow state recognition result indicates that the user is currently in a flow state; adjusting the end time to before the cognitive exhaustion time when the flow state recognition result indicates that the user is not currently in a flow state and the cognitive exhaustion time is earlier than the end time; and maintaining the end time unchanged when the flow state recognition result indicates that the user is not currently in a flow state and the end time is earlier than the cognitive exhaustion time.

[0085] Among them, the flow state recognition result is a judgment on whether the user is in a highly immersive, efficient and pleasant "flow" psychological state after analyzing the user's physiological state feature sequence.

[0086] Optionally, the server adjusts the end time of the current focused task according to the following dynamic decision-making logic: First, it checks the flow state recognition result. If the result is yes, then regardless of other conditions, it directly performs the extension operation, pushing the end time back by a preset or dynamically calculated time period. If the result is no, the server further compares the order of the cognitive exhaustion time and the current end time. If the exhaustion time is earlier than the end time, it means that the user will have exhausted their energy before the end of the current focused task, and the server will advance the end time to before the cognitive exhaustion time. If the cognitive exhaustion time is later than or equal to the end time, it means that the planned duration of the current focused task is within the user's cognitive capacity, and the original end time remains unchanged. The adjustment information will be synchronized to the user interface in real time.

[0087] In this embodiment, when the user is in a state of flow, all preset rest reminders or end suggestions are suppressed, and the task duration is dynamically extended (e.g., by 5 minutes each time) until the flow characteristics subside.

[0088] In this embodiment, by combining the flow state recognition result with the two core judgment conditions of cognitive exhaustion, a refined, adaptive and goal-oriented dynamic intervention rule is formed. This rule can prioritize the protection and prolongation of the flow state of efficient learning, proactively prevent cognitive exhaustion to stop ineffective efforts, and enable the management of the learning process to flexibly respond to the user's real psychological and physiological state, achieving the optimal balance between improving learning efficiency and protecting cognitive health.

[0089] In an exemplary embodiment, the physiological state features include focus features, blinking features, and heart rate change features. Based on the physiological state feature sequence, the user's flow state recognition result is determined, including: determining whether the user currently meets the preset flow state judgment conditions based on the heart rate change features, focus features, and blinking features in the physiological state feature sequence; and determining that the user is currently in a flow state if the user currently meets the preset flow state judgment conditions.

[0090] Flow is a psychological state characterized by complete immersion in an activity, accompanied by a high degree of engagement, pleasure, and high efficiency. Physiologically, flow is often manifested by a specific pattern of heart rate variability (HRV), a moderate level of focus, and a low blink rate.

[0091] Optionally, the server analyzes the heart rate variability (HRV), attention feature value (beta / alpha), and blink frequency in the physiological state feature sequence in real time. When "high beta / low alpha" and extremely low blink frequency (deep immersion) are present, and combined with heart rate variability, it is determined that the user currently meets the criteria for determining the flow state. Therefore, it is determined that the user is currently in a flow state.

[0092] In this embodiment, the user's flow state is accurately identified through physiological state feature sequences, which is beneficial for intelligently extending task time in the flow state. This achieves accurate capture and active protection of the efficient learning window, and solves the problem that the user's optimal learning state may be interrupted due to the inability to identify the flow state.

[0093] In one exemplary embodiment, predicting a user's cognitive exhaustion moment based on a physiological state feature sequence includes: inputting the physiological state feature sequence into each pre-trained cognitive state recognition model to obtain a cognitive state index sequence; the cognitive state index sequence includes each cognitive state index of the user from the start time to the current time; inputting the cognitive state index sequence into a pre-trained cognitive exhaustion point prediction model, and determining the user's cognitive exhaustion moment based on the user's fatigue curve through the pre-trained cognitive exhaustion point prediction model.

[0094] Among them, the cognitive state index is a quantitative indicator output by the model that describes the user's cognitive ability level changing over time. Cognitive state indicators can be such as focus index, fatigue index, relaxation index, and mood index.

[0095] Among them, the pre-trained cognitive state recognition model refers to a machine learning or deep learning model (such as support vector machine, random forest, neural network) trained using historical labeled data (such as associating physiological characteristics with the cognitive state exhibited by the user's subjective evaluation or behavior). Its function is to map the input physiological state features into an estimate of the user's cognitive state. The pre-trained cognitive state recognition model can refer to a pre-trained attention recognition model, a pre-trained fatigue recognition model, a pre-trained relaxation recognition model, and a pre-trained emotion recognition model.

[0096] In practical applications, during the pre-training process, the model first collects EEG signal data through designed experiments, and then uses multi-dimensional EEG features, such as time-domain features (calculating the statistical features of the signal: mean, variance, skewness, kurtosis, Hjorth parameter) and frequency-domain features (calculating the absolute power of each frequency band using power spectral density PSD). The task-state data is standardized using baseline data, and then EEG state feature vectors are constructed using feature dimensionality reduction. Machine learning algorithms are then used to train attention recognition models, fatigue recognition models, relaxation recognition models, and emotion recognition models.

[0097] Among them, the pre-trained cognitive exhaustion prediction model is a time series prediction model (such as LSTM model, Transformer model). It learns the pattern of cognitive resource consumption by analyzing the changing trends of users' historical and current cognitive state indicators, thereby predicting the critical time point when future cognitive performance will decline significantly or become unsustainable, i.e., the moment of cognitive exhaustion.

[0098] For example, a pre-trained cognitive exhaustion prediction model predicts the "cognitive exhaustion point" in the next 5-20 minutes based on the data stream that just occurred from the start of the current focused task (0 minutes) to the current moment (T minutes). It calculates that fatigue should be felt at a certain time (e.g., at the 45th minute) based on the fatigue curve.

[0099] The fatigue curve is a function curve that describes how user fatigue changes over time.

[0100] Optionally, the server inputs the physiological state feature sequence into multiple pre-trained cognitive state recognition models. These models output the user's focus index, fatigue index, relaxation index, and mood index at different time points. The server integrates these indicators, sorts them by time, and generates a cognitive state indicator sequence. Then, the sequence containing the user's cognitive state indicators from the start of the current focused task to the current time is input into a cognitive exhaustion point prediction model. This model analyzes the pattern of the sequence and, combined with the user's personalized fatigue curve, infers at what point in the future the user's cognitive resources are expected to be exhausted or reduced to an ineffective level if the task is not interrupted, and outputs this predicted time as the cognitive exhaustion time.

[0101] In this embodiment, a pre-trained cognitive state recognition model is used to "translate" physiological state characteristics into directly understandable cognitive state indicators, which can indirectly measure the user's cognitive process and model the user's cognitive state. Furthermore, a pre-trained cognitive exhaustion point prediction model can proactively identify "cognitive exhaustion moments," which can prevent users from experiencing a sharp drop in learning effectiveness or developing feelings of boredom due to excessive consumption. This achieves a more intelligent and humanized learning process management than fixed durations or simple reminders.

[0102] In one exemplary embodiment, the cognitive state indicators include a focus index, and the method further includes: determining the user's current distraction duration based on the user's focus index from the start time to the current time; determining an attention intervention strategy for the user's current attention based on the current distraction duration; and intervening in the user's current attention according to the attention intervention strategy.

[0103] The current distraction duration refers to the cumulative time during which the user's attention has been diverted from the learning task, calculated based on the duration when the focus index is below a preset threshold.

[0104] Among them, attention intervention strategies refer to a series of pre-set methods to help users refocus.

[0105] Optionally, the server continuously monitors the real-time output focus index. When the focus index is consistently below the preset value of 30, it can be judged as distraction, start a timer, accumulate the "current distraction duration", and match different intensity intervention strategies (such as mild reminders, moderate intervention, and strong guidance) from the strategy library according to different intervals of distraction duration (such as short-term inattentiveness, moderate distraction, and severe detachment) and execute them.

[0106] In this application, attention intervention strategies can include three types: subconscious repair, peripheral perception cues, and interactive blocking. When distraction lasts 1-2 minutes, a subconscious repair strategy is used, which involves not interrupting the user but superimposing a 40Hz brainwave using the neural entrainment effect onto the background white noise to subtly bring attention back. When distraction lasts 2-5 minutes, a peripheral perception cues strategy is used, which involves triggering micro-vibrations and providing gentle voice reminders. When distraction lasts longer than 5 minutes, an interactive blocking strategy is used, which involves indicating poor current concentration and suggesting stopping learning or switching tasks.

[0107] In this embodiment, by quantifying the duration of distraction and implementing graded attention recall based on the degree of distraction, a refined and personalized approach to the problem of attention deficit is achieved, effectively enhancing the coherence and controllability of the learning process.

[0108] In one exemplary embodiment, the cognitive state indicators include fatigue indicators, and the method further includes: determining whether the user is currently in a fatigued state based on the user's fatigue indicators from the start time to the current time; if the user is determined to be in a fatigued state, determining a rest duration that matches the fatigue indicators at the current time based on the user's fatigue decay curve; and generating rest suggestion prompts based on the rest duration and pushing them to the user's terminal.

[0109] The fatigue decline curve is a personalized function or model describing the time required for a user to recover from different fatigue levels to a baseline state, and can be learned from historical data. In this application, the fatigue decline curve is fitted based on "effective rest records" from historical data.

[0110] The matching rest duration refers to the recommended rest time that can theoretically effectively relieve fatigue, calculated based on the user's current fatigue level and their individual recovery characteristics.

[0111] Optionally, the server analyzes real-time fatigue indicators. If the fatigue index continues to exceed the preset fatigue threshold of 90, it determines that the user is not suitable for learning and is in a state of fatigue. Based on the fatigue decline curve, it determines the recommended rest duration and suggested activities (such as looking into the distance, drinking water, and light activity) and pushes them to the user's device.

[0112] In this embodiment, rest suggestions are triggered based on fatigue indicators, and personalized rest durations are recommended according to individual recovery characteristics. This achieves scientific energy management and solves the problem that rest arrangements in traditional learning tools are either rigid or rely entirely on the user's self-discipline. It ensures the timeliness and effectiveness of rest intervention, helps users recover in time before fatigue accumulates, and maintains long-term learning sustainability.

[0113] In an exemplary embodiment, prior to the steps of obtaining the start time, end time, and current time of the current focused task, the method further includes: in response to a task parameter setting operation for the current focused task, obtaining the task type and start time set by the task parameter setting operation; extracting session data corresponding to each historical focused task belonging to the task type from the user's session database, and determining a focus duration prediction value for the current focused task based on the session data of each historical focused task; and determining the end time based on the focus duration prediction value and the start time.

[0114] The task parameter setting operation refers to the operation of setting task parameters in advance before entering a focused task. At least the user needs to set the type of task to be completed and the start time.

[0115] The session database stores data on the entire lifecycle of a user's focused tasks. Each session record includes a timestamp of a task, physiological indicators, intervention records, and information on the learning subjects marked by the user.

[0116] The focus duration prediction value is based on the user's historical performance data for similar tasks, predicting the effective focus duration that the current focus task may achieve.

[0117] Optionally, when a user creates a new focus task in the user interface and sets the task type (such as reading, programming, or writing) and the planned start time, the server responds to the operation by filtering all historical task records of the same task type from the user's session database, using this historical data (such as past effective focus time and completion status) to make predictions, generating a focus time prediction value for the current focus task, and adding the focus time prediction value to the planned start time to calculate the end time as the end point of the task timing.

[0118] It should be noted that if the user is a novice user and the system does not store the historical data of novice users, the general default value (such as 45 minutes) will be used as the predicted focus time value, that is, the default focus time for focus tasks is 45 minutes.

[0119] In this embodiment, before the task begins, the duration of the current focused task is predicted in a personalized way using the user's historical behavior data. Based on the user's learning patterns and past experience, personalized durations can be flexibly recommended to the user.

[0120] In an exemplary embodiment, the session data of historical focus tasks includes the effective focus time of the user in the historical focus tasks. Based on the session data of each historical focus task, a predicted focus time value for the current focus task is determined, including: determining a time decay weight coefficient and a focus quality weight coefficient for each historical focus task, and determining a comprehensive weight coefficient for each historical focus task based on the time decay weight coefficient and the focus quality weight coefficient; using the focus quality weight coefficient of each historical focus task, weighting and summing the effective focus time of each historical focus task to obtain a first predicted focus time value; and using the comprehensive weight coefficient of each historical focus task, weighting and summing the effective focus time of each historical focus task to obtain a second predicted focus time value; and determining the predicted focus time value based on the first predicted focus time value and the second predicted focus time value.

[0121] The time decay weight coefficient is assigned a weight based on the time elapsed since the historical task occurred; generally, the more recent the task, the higher the weight.

[0122] The focus quality weighting coefficient is a weighting based on the quality of historical task completion (such as the stability of focus and whether the task was completed proactively in advance).

[0123] The comprehensive weighting coefficient is the final weight that combines time decay and quality assessment, and is used to more comprehensively evaluate the value of each historical record to the current prediction.

[0124] Optionally, the server calculates two weights for each historical task record: a time decay weight calculated based on a decay function of the time elapsed since the task occurred, and a focus quality weight evaluated based on session data of that task (such as average focus level and number of interruptions). Then, the two weights are multiplied to obtain a comprehensive weight coefficient. The historical effective focus duration is then weighted using the focus quality weight to obtain a first predicted value reflecting the absolute true ability level. The historical effective focus duration is then weighted using the comprehensive weight coefficient to obtain a second predicted value reflecting the recent trend. Finally, the two predicted values ​​are weighted and summed to determine the final predicted focus duration value.

[0125] To facilitate understanding by those skilled in the art, the following method for determining the predicted focus duration is provided. Specifically, firstly, historical data from the user session database is extracted, including all learning session records for the same subject or task type, and data with excessively short durations (e.g., <5 minutes) or marked as "abnormal interruptions" is removed. Secondly, since learning habits are often cyclical (e.g., final exam review weeks, quarterly assessments), a "long-short-term memory fusion" model is used to calculate the prediction in two parts: a recent trend prediction calculation and a long-term baseline calculation. The results of the two parts are then weighted and summed using the following formula: ,in, This is the result of recent trend calculations. This is the result of long-term baseline calculation.

[0126] In the recent trend calculation, a time decay-focused quality-weighted moving average algorithm is used to determine the personalized recommendation duration. If the average focus level for a particular task is extremely high (entering a flow state), then the duration of that task is more valuable. The calculation formula is: ,in, , This represents the overall weighting coefficient for the i-th task; This represents the effective focus time of the i-th historical focus task (referring to the actual time the user persisted before the system judged fatigue / distraction or the efficient time when the user actively ended the task). Represents the overall weight coefficient of the i-th historical focus task; The focus quality factor is calculated based on the average focus index for the current task, which is the average focus divided by 100. This represents the time decay factor, which is more important the closer it is to the current moment. It can be used to reflect the timeliness of user habits, meaning that the closer the record is to the present moment, the greater its reference value. , For the current date, For the date of the i-th record, The decay constant is set to 0.95. Data from 1 day ago has a weight of 0.95, data from 1 week ago has a weight of 0.7 (still has high reference value), data from 2 weeks ago has a weight of 0.49 (weight halved), and data from 1 month ago has a weight of 0.21.

[0127] The long-term baseline calculation process uses historical data from the past 90 days, without using a time decay factor, and only uses a weighted average based on the focus quality factor, which can represent the user's true ability level in a specific subject.

[0128] In this embodiment, a dual weighting mechanism of time decay and focus quality assessment is introduced to predict focus duration. This allows the predicted focus duration to not only reflect the user's long-term ability baseline but also to keenly capture recent state change trends. This refined prediction method makes the automatically set end time for the current focus task more closely match the user's latest and most realistic ability performance, making it more personalized.

[0129] In an exemplary embodiment, the method further includes: acquiring user session data after completing a focused task; the session data includes at least timestamps and physiological indicator data; extracting the user's state feature vector during the task completion process based on the session data of the completed focused task; filtering out virtual object visualization parameters that match the state feature vector based on preset mapping rules; generating virtual objects corresponding to the completed focused task based on the virtual object visualization parameters and displaying them in a virtual space; the virtual form of the virtual object in the virtual space represents the user's physiological state level after completing the focused task.

[0130] The session data includes at least timestamps and physiological indicators, and may also include intervention records and user-annotated learning subject information.

[0131] This application stores the session logs of each user’s focused task.

[0132] Among them, the state feature vector is a set of features extracted from a complete task session data that can summarize the user's overall physiological state in that task, such as average focus index, flow duration, fatigue recovery efficiency, and emotional stability.

[0133] Among them, the virtual object visualization parameters are a set of values ​​used to control the appearance attributes of virtual objects.

[0134] In practical applications, virtual objects can refer to flowers in a user's personal virtual garden. Each time a user completes a focus task, their personal virtual garden generates a flower corresponding to that task. Visualization parameters for virtual objects can include the flower's shape and abundance, color and saturation, and degree of opening and closing, as well as its posture. Specifically, focus duration and flow rate are mapped to shape and abundance; the longer the focus time and the deeper the flow, the thicker the flower's stems and leaves, and the more layers of petals it has. Emotional state and relaxation index are mapped to color and saturation; a calm and focused state corresponds to clear, bright colors, while a high anxiety level during the task may result in darker flowers or flowers with mottled textures, directly reflecting psychological strain. Average focus score is mapped to degree of opening and closing, with a high average focus score corresponding to a fully bloomed flower, and a low average focus score corresponding to a budding or slightly curled shape.

[0135] The preset mapping rules define the correspondence between different physiological state characteristics and the values ​​of visualization parameters.

[0136] Optionally, when a focus task is completed, the server retrieves the complete session data of the focus task and calculates the average focus index, flow duration, fatigue recovery efficiency and emotional stability of the focus task from the data, and generates a state feature vector. Then, according to the preset mapping rules, a unique virtual plant is generated and displayed in the user's personal virtual garden.

[0137] In this embodiment, abstract and invisible physiological characteristics are transformed into concrete virtual objects, providing positive feedback for the user's learning process. This solves the problem of the single form of feedback in traditional learning tools, allowing users to more intuitively perceive their efforts and growth, thus forming a virtuous cycle.

[0138] In one exemplary embodiment, the method further includes: in response to a virtual object click operation in the virtual space, determining the target virtual object clicked by the virtual object click operation; displaying the morphological change history of the target virtual object in the virtual space during the execution of the corresponding target focus task; the morphological change history is generated based on the session data of the target focus task; the morphological change history characterizes the fluctuation of the user's physiological state level during the execution of the target focus task.

[0139] Among them, the morphological change history is a dynamic replay or visual summary that shows how the appearance of a virtual object changes synchronously with the user's real-time physiological state throughout the entire process from the start to the end of a focused task.

[0140] Among them, the target focus task refers to the focus task that has been completed, represented by the virtual object clicked by the user.

[0141] Optionally, when a user clicks on a virtual object in the virtual space, the server identifies the clicked target virtual object and its associated target focus task, then retrieves the detailed session data of the target focus task, and re-enacts the change process of the virtual object in the form of animation, thereby reproducing the history of the virtual object's morphological changes during task execution as the user's physiological state fluctuates.

[0142] In practical applications, each time a user completes a focused task (such as 2 hours of study), the server generates a final-form flower based on the average focus level of that task and stores it in the user's "personal achievement." Users can intuitively perceive the results of their long-term efforts through the lushness of their personal virtual garden. When a user clicks on a flower, the system can replay the generation process, showing in fast-forward the dynamic changes of the flower as it grows and opens / closes with the task progress (corresponding to physiological fluctuations in the original task timeline), allowing users to review their focused journey and intuitively experience the fruits of their labor.

[0143] In this embodiment, by providing an interactive virtual object form change playback function, the fluctuations in the user's physiological state during a learning task are externalized and visualized, allowing the user to intuitively review the fluctuations in their attention, fatigue points, and flow moments during their learning process. This visual feedback can help users understand their own learning patterns and improve their self-regulation ability.

[0144] In practical applications, besides displaying a personal virtual garden on the user interface, the app also provides data dashboards that can chart key indicator trends for daily, weekly, and monthly periods: total focus time, average focus score, optimal focus time, mood fluctuation curves, and the probability of flow. The app also offers focus training and meditation functions, allowing users to train their focus and meditation according to their needs, such as performing 10 minutes of meditation before starting to study. The training effect is quantified in real-time through EEG signal feedback. The app can also create a "personal efficiency profile" by analyzing users' historical conversation data. It statistically analyzes users' focus records over several weeks or months to obtain average focus distribution over different time periods, focus performance in different subjects, fatigue recovery curves, the correlation between mood fluctuations and learning quality, and learning pattern clustering results. This allows the app to uncover users' individual biorhythm patterns (such as daily optimal performance times) and cognitive characteristics (such as efficiency differences across subjects), thereby providing forward-looking and predictive planning suggestions.

[0145] The aforementioned learning pattern clustering results refer to several typical and significantly characteristic learning state types automatically summarized by the system using machine learning algorithms (such as K-Means, DBSCAN, etc.) to conduct multi-dimensional analysis of users' long-term EEG data, behavioral data, and subjective feedback. For example, clustering based on subject-time matching, through correlation analysis of subject labels, timestamps, and EEG states, gives corresponding types, divided into morning logical type and nighttime creative type. The morning logical type has the highest focus score when performing logical tasks such as mathematics and programming in the morning (e.g., 10:00-11:30), while the nighttime creative type has active alpha waves, high relaxation, and stable emotional state in the evening. Clustering based on "recovery ability," by analyzing the brainwave recovery of users after rest, is divided into rapid recovery type and deep recovery type. The rapid recovery type only requires 5 minutes of light rest, and the brainwaves recover quickly. It can recover to baseline levels, but deep recovery requires longer (more than 15 minutes) or specific forms of rest (such as meditation or sleep) to eliminate brain fatigue; based on clustering of "focus duration", users are classified into sprint type and long-distance endurance type according to the data characteristics of maintaining high focus duration. Sprint type has extremely high focus, but the duration is short, and then the focus drops sharply. Long-distance endurance type is slower to enter the state (slow start), but once it enters the "flow" state (high beta / low alpha, low blink rate), it can maintain efficient learning for more than 45-90 minutes.

[0146] In the process of creating a personal efficiency profile, the system analyzes the user's data across all dates to accurately identify the "golden time" of the day when the user's focus is most concentrated and their emotions are most stable. For example, "historical data shows that your focus is on average 25% higher between 10:00 and 11:30 in the morning than in the afternoon."

[0147] In the process of biorhythm modeling, the user's biorhythm patterns and cognitive characteristics are mined to automatically identify prime learning times. For example, concentration levels are on average 25% higher between 10:00 and 11:30 AM than in the afternoon. Thus, when users set task parameters, the system can combine task type and historical session data to provide recommendations on task order and learning time, and guide rest intervals and task duration.

[0148] In another embodiment, such as Figure 2 As shown, a focused task intervention method is provided, and its application to a server is illustrated, including the following steps:

[0149] Step S202: Obtain the start time, end time, and current time of the current focused task, and obtain the user's physiological state feature sequence; the physiological state feature sequence includes the user's physiological state features from the start time to the current time.

[0150] Step S204: Based on the physiological state feature sequence, determine the user's flow state identification result, and based on the physiological state feature sequence, predict the user's cognitive exhaustion time.

[0151] Step S206: If the flow state recognition result indicates that the user is currently in a flow state, extend the end time.

[0152] Step S208: If the flow state recognition result indicates that the user is not currently in a flow state and the time of cognitive exhaustion is earlier than the end time, adjust the end time to before the time of cognitive exhaustion.

[0153] Step S210: If the flow state recognition result indicates that the user is not currently in a flow state and the end time is earlier than the cognitive exhaustion time, the end time is kept unchanged.

[0154] It should be noted that the specific limitations of the above steps can be found in the specific limitations of a focus task intervention method described above.

[0155] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0156] Based on the same inventive concept, this application also provides a focus task intervention device for implementing the focus task intervention method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more focus task intervention device embodiments provided below can be found in the limitations of the focus task intervention method described above, and will not be repeated here.

[0157] In one embodiment, this application also provides a focus task intervention device, such as... Figure 3 As shown, it includes: an acquisition module 302, a recognition module 304, and an intervention module 306, wherein:

[0158] The acquisition module 302 is used to acquire the start time, end time, and current time of the currently focused task, and to acquire the user's physiological state feature sequence; the physiological state feature sequence includes the user's physiological state features from the start time to the current time;

[0159] The identification module 304 is used to determine the user's flow state identification result based on the physiological state feature sequence, and to predict the user's cognitive exhaustion time based on the physiological state feature sequence.

[0160] Intervention module 306 is used to adjust the end time of the current focused task based on the flow state recognition result and the cognitive exhaustion time.

[0161] In one embodiment, the intervention module 306 is specifically configured to: extend the end time when the flow state recognition result indicates that the user is currently in a flow state; adjust the end time to before the cognitive exhaustion time when the flow state recognition result indicates that the user is not currently in a flow state and the cognitive exhaustion time is earlier than the end time; and maintain the end time unchanged when the flow state recognition result indicates that the user is not currently in a flow state and the end time is earlier than the cognitive exhaustion time.

[0162] In one embodiment, the physiological state features include focus features, blinking features, and heart rate change features. The recognition module 304 is specifically used to determine whether the user currently meets the preset flow state judgment conditions based on the heart rate change features, focus features, and blinking features in the physiological state feature sequence; if the user currently meets the preset flow state judgment conditions, the user is determined to be in a flow state.

[0163] In one embodiment, the identification module 304 is specifically used to input the physiological state feature sequence into each pre-trained cognitive state identification model to obtain a cognitive state index sequence; the cognitive state index sequence includes each cognitive state index of the user from the start time to the current time; the cognitive state index sequence is input into a pre-trained cognitive exhaustion point prediction model, and the user's cognitive exhaustion time is determined based on the user's fatigue curve by the pre-trained cognitive exhaustion point prediction model.

[0164] In one embodiment, the cognitive state indicators include a focus index, and the device further includes: an attention recall module, configured to determine the user's current distraction duration based on the user's focus index from the start time to the current time; determine an attention intervention strategy for the user's current attention based on the current distraction duration; and intervene in the user's current attention according to the attention intervention strategy.

[0165] In one embodiment, the cognitive state indicators include fatigue indicators, and the device further includes: an inefficient state truncation module, used to determine whether the user is currently in a fatigue state based on the user's fatigue indicators from the start time to the current time; if the user is determined to be in a fatigue state, to determine a rest duration that matches the fatigue indicators at the current time based on the user's fatigue decay curve; and to generate rest suggestion prompts based on the rest duration and push them to the user's terminal.

[0166] In one embodiment, the device further includes: a personalized duration recommendation module, configured to, in response to a task parameter setting operation for the current focused task, obtain the task type and start time set in the task parameter setting operation; extract session data corresponding to each historical focused task belonging to the task type from the user's session database, and determine a predicted focus duration value for the current focused task based on the session data of each historical focused task; and determine an end time based on the predicted focus duration value and the start time.

[0167] In one embodiment, the session data of historical focus tasks includes the user's effective focus time in historical focus tasks. The personalized focus time recommendation module is specifically used to determine the time decay weight coefficient and focus quality weight coefficient for each historical focus task, and to determine the comprehensive weight coefficient for each historical focus task based on the time decay weight coefficient and focus quality weight coefficient; to use the focus quality weight coefficient of each historical focus task to perform a weighted sum of the effective focus time of each historical focus task to obtain a first predicted value of focus time; and to use the comprehensive weight coefficient of each historical focus task to perform a weighted sum of the effective focus time of each historical focus task to obtain a second predicted value of focus time; and to determine the predicted value of focus time based on the first predicted value of focus time and the second predicted value of focus time.

[0168] In one embodiment, the device further includes: a visualization module for acquiring user session data after completing a focused task; the session data includes at least timestamps and physiological indicator data; based on the session data of the completed focused task, extracting the user's state feature vector during the task completion process; based on preset mapping rules, filtering out virtual object visualization parameters that match the state feature vectors; based on the virtual object visualization parameters, generating virtual objects corresponding to the completed focused task and displaying them in a virtual space; the virtual form of the virtual object in the virtual space represents the user's physiological state level after completing the focused task.

[0169] In one embodiment, the visualization module is specifically configured to, in response to a virtual object click operation in the virtual space, determine the target virtual object clicked by the virtual object click operation; display the morphological change history of the target virtual object in the virtual space during the execution of the corresponding target-focused task; the morphological change history is generated based on the session data of the target-focused task; the morphological change history characterizes the fluctuation of the user's physiological state level during the execution of the target-focused task.

[0170] The modules in the aforementioned task-focused intervention device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0171] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the aforementioned focus task intervention method. The steps of the focus task intervention method described above can be steps from one of the focus task intervention methods in the various embodiments described above.

[0172] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, causes the processor to perform the steps of the aforementioned focus task intervention method. The steps of the focus task intervention method described above may be steps from one of the focus task intervention methods in the various embodiments described above.

[0173] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, causes the processor to perform the steps of the aforementioned focus task intervention method. The steps of the focus task intervention method described above may be steps from one of the focus task intervention methods in the various embodiments described above.

[0174] Indicatively, such as Figure 4 As shown, Figure 4 This is a schematic diagram of the internal structure of a computer device 400 provided in an embodiment of this application. The computer device 400 can be provided as a server. (Refer to...) Figure 4 The computer device 400 includes a processing component 402, which further includes one or more processors, and memory resources represented by memory 401 for storing instructions, such as application programs, that can be executed by the processing component 402. The application programs stored in memory 401 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 402 is configured to execute instructions to perform the focus task intervention method of any of the above embodiments.

[0175] The computer device 400 may also include a power supply component 403 configured to perform power management of the computer device 400, a wired or wireless network interface 404 configured to connect the computer device 400 to a network, and an input / output (I / O) interface 405. The computer device 400 may operate on an operating system stored in memory 401, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0176] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0177] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0178] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0179] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0180] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A focus task intervention method, characterized in that, The method includes: In response to a task parameter setting operation for the currently focused task, obtain the task type and start time set by the task parameter setting operation; Extract session data corresponding to each historical focus task belonging to the task type from the user's session database; the session data of the historical focus task includes the user's effective focus time in the historical focus task; Based on the session data of each historical focus task, a predicted focus duration for the current focus task is determined, specifically including: determining a time decay weight coefficient and a focus quality weight coefficient for each historical focus task; and determining a comprehensive weight coefficient for each historical focus task based on the time decay weight coefficient and the focus quality weight coefficient; using the focus quality weight coefficient of each historical focus task, a weighted sum of the effective focus durations of each historical focus task is obtained to get a first predicted focus duration; and using the comprehensive weight coefficient of each historical focus task, a weighted sum of the effective focus durations of each historical focus task is obtained to get a second predicted focus duration; and determining the predicted focus duration based on the first predicted focus duration and the second predicted focus duration. Based on the predicted focus duration and the start time, the end time is determined; The system acquires the start time, the end time, and the current time, and acquires the user's physiological state feature sequence; the physiological state feature sequence includes each physiological state feature of the user from the start time to the current time; the physiological state features include focus features, blinking features, and heart rate change features; Based on the physiological state feature sequence, the user's flow state recognition result is determined, and based on the physiological state feature sequence, the user's cognitive exhaustion time is predicted. Specifically, this includes: determining whether the user currently meets the preset flow state judgment conditions based on the heart rate change characteristics, focus characteristics, and blinking characteristics in the physiological state feature sequence; if the user currently meets the preset flow state judgment conditions, determining that the user is currently in a flow state; inputting the physiological state feature sequence into each pre-trained cognitive state recognition model to obtain a cognitive state index sequence; the cognitive state index sequence includes each cognitive state index of the user from the start time to the current time; inputting the cognitive state index sequence into a pre-trained cognitive exhaustion point prediction model, and determining the user's cognitive exhaustion time based on the user's fatigue curve through the pre-trained cognitive exhaustion point prediction model; Based on the flow state recognition results and the cognitive exhaustion time, the end time of the current focused task is adjusted.

2. The method according to claim 1, characterized in that, The step of adjusting the end time of the current focused task based on the flow state recognition result and the cognitive exhaustion time includes: If the flow state recognition result indicates that the user is currently in a flow state, the end time is extended; If the flow state recognition result indicates that the user is not currently in the flow state and the cognitive exhaustion time is earlier than the end time, the end time is adjusted to be before the cognitive exhaustion time; If the flow state recognition result indicates that the user is not currently in the flow state and the end time is earlier than the cognitive exhaustion time, the end time is kept unchanged.

3. The method according to claim 1, characterized in that, The cognitive state indicators include attention indicators, and the method further includes: Based on the user's focus index from the start time to the current time, determine the user's current distraction duration; Based on the current distraction duration, determine the attention intervention strategy for the user's current state; According to the attention intervention strategy, the user's current attention is intervened.

4. The method according to claim 1, characterized in that, The cognitive state indicators include fatigue indicators, and the method further includes: Based on the user's fatigue index from the start time to the current time, determine whether the user is currently in a state of fatigue; If it is determined that the user is in the state of fatigue, a rest duration matching the fatigue index at the current moment is determined based on the user's fatigue decline curve. Based on the rest duration, a rest suggestion message is generated and pushed to the user's terminal.

5. The method according to claim 1, characterized in that, The method further includes: Acquire the user's session data after completing a focused task; the session data includes at least timestamps and physiological indicator data. Based on the session data of the completed focus task, extract the user's state feature vector during the task completion process of the completed focus task; Based on preset mapping rules, virtual object visualization parameters that match the state feature vector are selected. Based on the visualization parameters of the virtual object, a virtual object corresponding to the completed focus task is generated and displayed in the virtual space; the virtual form of the virtual object in the virtual space represents the user's physiological state level under the completed focus task.

6. The method according to claim 5, characterized in that, The method further includes: In response to a click operation on a virtual object in the virtual space, the target virtual object clicked by the virtual object click operation is determined; The virtual space displays the morphological change history of the target virtual object during the execution of the corresponding target focus task; the morphological change history is generated based on the session data of the target focus task; the morphological change history represents the fluctuation of the user's physiological state level during the execution of the target focus task.

7. A focus task intervention device, characterized in that, The device includes: A personalized focus duration recommendation module is used to respond to a task parameter setting operation for the current focus task, obtain the task type and start time set in the task parameter setting operation; extract session data corresponding to each historical focus task belonging to the task type from the user's session database; the session data of the historical focus tasks includes the user's effective focus duration in the historical focus tasks; and determine the focus duration prediction value for the current focus task based on the session data of each historical focus task, specifically including: determining the time decay weight coefficient and focus quality weight coefficient for each historical focus task, and based on the session data of each historical focus task... The time decay weight coefficient and focus quality weight coefficient are used to determine a comprehensive weight coefficient for each historical focus task; the effective focus duration of each historical focus task is weighted and summed using the focus quality weight coefficient to obtain a first predicted value of focus duration; and the effective focus duration of each historical focus task is weighted and summed using the comprehensive weight coefficient to obtain a second predicted value of focus duration; the predicted value of focus duration is determined based on the first and second predicted values ​​of focus duration; and the end time is determined based on the predicted value of focus duration and the start time. The acquisition module is used to acquire the start time, the end time, and the current time, and to acquire the user's physiological state feature sequence; the physiological state feature sequence includes the user's physiological state features from the start time to the current time; the physiological state features include focus features, blinking features, and heart rate change features; The identification module is used to determine the user's flow state identification result based on the physiological state feature sequence, and to predict the user's cognitive exhaustion time based on the physiological state feature sequence. Specifically, it includes: determining whether the user currently meets the preset flow state judgment conditions based on heart rate change features, attention features, and blinking features in the physiological state feature sequence; if the user currently meets the preset flow state judgment conditions, determining that the user is currently in a flow state; inputting the physiological state feature sequence into each pre-trained cognitive state identification model to obtain a cognitive state index sequence; the cognitive state index sequence includes each cognitive state index of the user from the start time to the current time; inputting the cognitive state index sequence into a pre-trained cognitive exhaustion point prediction model, and determining the user's cognitive exhaustion time based on the user's fatigue curve through the pre-trained cognitive exhaustion point prediction model; An intervention module is used to adjust the end time of the current focused task based on the flow state recognition result and the cognitive exhaustion time.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images