Work intensity detection method and device, equipment and storage medium

Abstract

The embodiment of the invention provides a working intensity detection method and device, equipment and a storage medium, and belongs to the technical field of artificial intelligence. The method comprises the following steps: acquiring working data of a user, wherein the working data comprises terminal use data and physiological data collected by wearable equipment in a working state; performing time alignment processing on the terminal use data and the physiological data to obtain target working data; performing feature extraction processing on the target working data to obtain a working feature sequence; and performing working intensity detection on the working feature sequence through a working intensity detection model based on a Transform architecture to obtain a working intensity index of the user. According to the embodiment of the invention, the accuracy of work intensity detection can be improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a work intensity detection method, device, equipment and storage medium. Background Art

[0002] Reasonable work intensity not only helps to improve work efficiency, reduce stress, improve work quality and physical and mental health, but also helps to balance work and life and promote sustainable career development.

[0003] Currently, some work management applications can provide some simple work intensity test results, usually through fixed rules or calculation formulas, combined with quantitative indicators (such as task duration, completion) and qualitative indicators (such as task complexity, collaboration difficulty, etc.) to comprehensively calculate the work intensity results. However, different users have different work efficiency, working methods and stress tolerance. This estimation method fails to take into account individual differences, and the accuracy of work intensity detection is not high.

[0004] Therefore, how to improve the accuracy of work intensity detection has become a technical problem that needs to be solved urgently. Summary of the invention

[0005] The main purpose of the embodiments of the present application is to propose a work intensity detection method, device, equipment and storage medium, aiming to improve the accuracy of work intensity detection.

[0006] To achieve the above object, a first aspect of an embodiment of the present application provides a work intensity detection method, the method comprising:

[0007] Acquire the user's work data, wherein the work data includes terminal usage data and physiological data in the working state collected by the wearable device;

[0008] Performing time alignment processing on the terminal usage data and the physiological data to obtain target working data;

[0009] Performing feature extraction processing on the target working data to obtain a working feature sequence;

[0010] The work intensity detection model based on the Transformer architecture is used to perform work intensity detection on the work feature sequence to obtain the work intensity index of the user.

[0011] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a working intensity detection device, the device comprising:

[0012] A working data acquisition module, used to acquire the user's working data, wherein the working data includes terminal usage data and physiological data collected by the wearable device in the working state;

[0013] A time alignment module, used for performing time alignment processing on the terminal usage data and the physiological data to obtain target working data;

[0014] A working feature extraction module, used for performing feature extraction processing on the target working data to obtain a working feature sequence;

[0015] The work intensity detection module is used to perform work intensity detection on the work feature sequence through a work intensity detection model based on the Transformer architecture to obtain the work intensity index of the user.

[0016] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes a computer device, which includes a memory and a processor, the memory stores a computer program, and the processor implements the method described in the first aspect when executing the computer program.

[0017] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect is implemented.

[0018] The work intensity detection method, device, equipment and storage medium proposed in the present application, the work intensity detection method obtains the user's work data, the work data includes terminal usage data and physiological data collected by wearable devices in the working state; performs time alignment processing on the terminal usage data and physiological data to obtain target work data; performs feature extraction processing on the target work data to obtain a work feature sequence; through a work intensity detection model based on the Transformer architecture, the work intensity detection is performed on the work feature sequence to obtain the user's work intensity index. In this way, on the one hand, by performing time alignment processing on multimodal terminal usage data and physiological data collected by wearable devices, the consistency of multimodal data in the time series is ensured, which can provide a more comprehensive and accurate analysis basis for work intensity detection; on the other hand, by performing work intensity detection on the work intensity detection model based on the Transformer architecture, the long-term dependency in the work feature sequence can be effectively captured, so as to better understand the contribution of work behavior and physiological reactions at different time steps to the overall work intensity, and realize more refined and faster automated work intensity detection, thereby improving the accuracy and efficiency of work intensity detection, and providing users with targeted high-precision work intensity index. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flow chart of a working intensity detection method provided in an embodiment of the present application;

[0020] Figure 2 is another flow chart of the working intensity detection method provided in an embodiment of the present application;

[0021] Figure 3 is another flow chart of the working intensity detection method provided in the embodiment of the present application;

[0022] Figure 4 is a schematic block diagram of a working intensity detection device provided in an embodiment of the present application;

[0023] Figure 5 It is a schematic block diagram of the structure of the computer device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0025] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0027] First, some nouns involved in this application are analyzed:

[0028] Artificial intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. AI is a branch of computer science that attempts to understand the essence of intelligence and produce a new type of intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing and expert systems. AI can simulate the information process of human consciousness and thinking. AI is also a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0029] Natural language processing (NLP): NLP uses computers to process, understand and apply human languages ​​(such as Chinese, English, etc.). NLP is a branch of artificial intelligence and an interdisciplinary subject between computer science and linguistics. It is often referred to as computational linguistics. Natural language processing includes grammatical analysis, semantic analysis, and text understanding. Natural language processing is often used in technical fields such as machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, information intent recognition, information extraction and filtering, text classification and clustering, public opinion analysis and opinion mining. It involves data mining related to language processing, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computing.

[0030] Based on this, the embodiments of the present application provide a work intensity detection method, apparatus, device and storage medium, which realize more refined automated work intensity detection through a work intensity detection model based on the Transformer architecture, aiming to improve the accuracy of work intensity detection and provide users with targeted high-precision work intensity index.

[0031] The work intensity detection method, device, equipment and storage medium provided in the embodiments of the present application are specifically illustrated through the following embodiments. First, the work intensity detection method in the embodiments of the present application is described.

[0032] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0033] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0034] The work intensity detection method provided in the embodiment of the present application relates to the field of artificial intelligence technology. The work intensity detection method provided in the embodiment of the present application can be applied to a terminal, can be applied to a server side, or can be software running in a terminal or a server side. In some embodiments, the terminal can be a vehicle-mounted terminal, a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or can be configured as a server cluster or a distributed system composed of multiple physical servers, or can be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the work intensity detection method, etc., but is not limited to the above forms.

[0035] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer computer devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0036] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0037] Figure 1 is an optional flow chart of the work intensity detection method provided in the embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S104.

[0038] Step S101, obtaining the user's work data, where the work data includes terminal usage data and physiological data collected by the wearable device in the working state.

[0039] Terminal usage data refers to monitoring data generated when a user uses a terminal to carry out work. Terminal usage data includes, but is not limited to, work behavior data such as office software usage monitoring values, mouse movement monitoring values, and / or keyboard input monitoring values.

[0040] Physiological data includes, but is not limited to, physiological response data such as heart rate, body temperature, blood pressure, blood oxygen and / or number of exercise steps.

[0041] Wearable devices include smart bracelets, smart watches, etc. Wearable devices can be equipped with different types of sensors such as electrocardiogram (ECG) sensors, temperature sensors, blood pressure sensors, photoplethysmography (PPG) sensors, accelerometers, etc.

[0042] In this way, when the user wears the wearable device while working, the ECG sensor of the wearable device can be used to collect the user's heart rate during work, the temperature sensor can be used to collect the user's body temperature during work, the blood pressure sensor can be used to collect the user's blood pressure during work, the PPG sensor can be used to collect the user's blood oxygen during work, and / or the accelerometer can be used to collect the user's exercise steps during work and other physiological data during the user's working state.

[0043] In this way, by obtaining detailed and accurate work data, a more comprehensive analysis basis can be provided for subsequent work intensity testing.

[0044] Step S102: Time alignment processing is performed on the terminal usage data and the physiological data to obtain target working data.

[0045] Taking into account that the sampling time of terminal usage data and physiological data is not the same, for example, physiological data may be collected once per second, while terminal usage data may be monitored once per minute, therefore, the terminal usage data and physiological data can be time-aligned to synchronize the terminal usage data and physiological data to a unified time granularity (such as every minute) and timeline.

[0046] Exemplarily, for example, the terminal usage data and the physiological data may be resampled to ensure the consistency of the terminal usage data and the physiological data in the same time series.

[0047] The terminal usage data and physiological data after time alignment processing constitute the target work data.

[0048] It can be understood that the terminal usage data after time alignment is presented in the form of a time series, and each time step in the time series corresponds to an office software usage monitoring value, a mouse movement monitoring value and / or a keyboard input monitoring value. Similarly, the data after time alignment is also presented in the form of a time series, and each time step in the time series corresponds to a heart rate, a body temperature, a blood pressure, a blood oxygen and / or a number of exercise steps. Among them, the time series is determined by the total time range of the collected work data. For example, if the work data from 8:00 to 19:00 is continuously collected, then the time series is from 8:00 to 19:00; the time step refers to the resampling time window (such as a resampling time window of every 10 minutes is a time step).

[0049] Step S103, performing feature extraction processing on the target working data to obtain a working feature sequence.

[0050] The target work data can be processed for feature extraction to obtain a work feature sequence, which facilitates the understanding and processing of the work intensity detection model.

[0051] It can be understood that the working characteristic sequence includes a time series and a working characteristic vector, and the terminal usage characteristic vector and the physiological characteristic vector of each time step in the time series constitute a working characteristic vector.

[0052] The terminal usage feature vector includes, but is not limited to, an office software usage feature vector (such as the duration of office software usage), a mouse movement feature vector (such as the number of mouse clicks), and / or a keyboard input feature vector (such as the keyboard input frequency).

[0053] Physiological feature vectors include, but are not limited to, heart rate feature vectors (such as average heart rate), body temperature feature vectors (such as body temperature fluctuation amplitude), blood pressure feature vectors (such as blood pressure fluctuation amplitude), blood oxygen feature vectors (blood oxygen fluctuation amplitude) and / or exercise step feature vectors (such as average exercise step number).

[0054] Step S104, performing work intensity detection on the work feature sequence through a work intensity detection model based on the Transformer architecture to obtain the user's work intensity index.

[0055] Among them, the work intensity detection model based on the Transformer architecture is obtained by pre-training a preset large language model.

[0056] Exemplarily, the large language model can be a natural language processing (NLP) model that is good at processing complex time series data, such as a BERT model based on the Transformer architecture or a GPT series model (such as the GPT-2 model). The work intensity detection model has strong robustness, adaptability, and generalization capabilities, which helps to improve its ability to cope with individual differences. Facing different users, it can effectively capture the implicit potential correlation between the terminal usage feature vector and the physiological feature vector at each time step in the work feature sequence, and output a clear work intensity index.

[0057] Based on this, the work intensity detection model can be used to detect the work intensity of the work feature sequence. The work intensity detection model can effectively understand the correlation between the work feature vector of each time step in the work feature sequence and the work feature vectors of other time steps through the multi-head attention mechanism, thereby capturing the global dependency between the work feature vectors of different time steps. It can also take into account the work behavior and physiological reactions at different time steps, determine the fluctuation range of work intensity, and obtain an accurate work intensity index.

[0058] Exemplarily, a work intensity index such as a work intensity score (eg, 1-10 points) or a work intensity label (eg, low, medium, high) is an indicator used to quantify work intensity.

[0059] In this way, through the work intensity detection model based on the Transformer architecture, high-precision work intensity detection is achieved, which can improve the efficiency, comprehensiveness and convenience of work intensity detection.

[0060] The work intensity detection method provided in the above embodiment obtains the work data of the user, and the work data includes the terminal usage data and the physiological data collected by the wearable device in the working state; performs time alignment processing on the terminal usage data and the physiological data to obtain the target work data; performs feature extraction processing on the target work data to obtain the work feature sequence; and performs work intensity detection on the work feature sequence through the work intensity detection model based on the Transformer architecture to obtain the work intensity index of the user. In this way, on the one hand, by performing time alignment processing on the multimodal terminal usage data and the physiological data collected by the wearable device, the consistency of the multimodal data in the time series is ensured, which can provide a more comprehensive and accurate analysis basis for work intensity detection; on the other hand, by performing work intensity detection on the work intensity detection model based on the Transformer architecture, the long-term dependency in the work feature sequence can be effectively captured, so as to better understand the contribution of work behaviors and physiological reactions at different time steps to the overall work intensity, and realize more refined and faster automated work intensity detection, thereby improving the accuracy and efficiency of work intensity detection, so as to provide users with targeted high-precision work intensity index.

[0061] In step S102 of some embodiments, the target work data can be cleaned to obtain clean work data, which includes clean terminal usage data and clean physiological data; the clean terminal usage data can be processed by time series feature extraction to obtain a terminal usage feature sequence; the clean physiological data can be processed by time-frequency domain feature extraction to obtain a physiological feature sequence; the terminal usage feature sequence and the physiological feature sequence can be fused to obtain a work feature sequence, which includes a time series and a work feature vector.

[0062] In order to ensure the reliability of the input of the work intensity detection model, the target work data may be cleaned to obtain clean work data, which includes clean terminal usage data and clean physiological data.

[0063] In order to facilitate the understanding of the work intensity detection model, the time series feature extraction processing can be performed on the cleaning terminal usage data to obtain the terminal usage feature sequence, which includes the time series and the terminal usage feature vector. Then, the time-frequency domain feature extraction processing can be performed on the cleaning physiological data to obtain the physiological feature sequence, which includes the time series and the physiological feature vector.

[0064] The terminal usage feature sequence and the physiological feature sequence are then fused to obtain a working feature sequence, which includes a time series and a working feature vector. For example, by concatenating the terminal usage feature vector and the physiological feature vector at each time step in the time series, a comprehensive working feature vector at each time step in the time series can be obtained.

[0065] Through the above method, important work feature sequences that are more helpful for work intensity detection can be accurately extracted, so that the detection efficiency and detection accuracy of the work intensity detection model are improved.

[0066] In some embodiments, the target working data is cleaned to obtain clean working data, and the target working data may be denoised to obtain denoised target working data; the denoised target working data may be filled with missing values ​​to obtain filled target working data; the filled target working data may be corrected for outliers to obtain corrected target working data; and the corrected target working data may be standardized to obtain clean working data.

[0067] Exemplarily, the cleaning process mainly includes denoising, missing value filling, outlier correction and standardization.

[0068] Considering that the target working data is usually affected by the noise of the user's working environment, the target working data may be subjected to denoising processing, for example, filtering processing is performed on the target working data to obtain denoised target working data.

[0069] The denoised target working data may also be processed to fill missing values, for example, by using mean, median filling, or interpolation to fill missing values ​​to obtain filled target working data.

[0070] The filled target working data may also be corrected for outliers, for example, by detecting outliers through box plots, standard deviations, etc., and adjusting the outliers to a normal range, or deleting the outliers to obtain corrected target working data.

[0071] The corrected target working data may also be subjected to standardization processing, for example, the corrected target working data may be subjected to normalization processing to obtain target cleaning working data.

[0072] In this way, by performing a detailed cleaning process on the target work data, higher quality clean work data can be provided.

[0073] See also Figure 2 In some embodiments, step S104 may include but is not limited to steps S201 to S204.

[0074] Step S201, inputting the work feature sequence into the work intensity detection model;

[0075] Step S202, encoding the working feature sequence to obtain a coding vector sequence;

[0076] Step S203, using a multi-head attention mechanism, the encoding vector sequence is processed to capture global feature dependencies to obtain global context features;

[0077] Step S204: perform work intensity prediction processing on the global context features to obtain a work intensity index.

[0078] First, the work feature sequence is used as the input of the work intensity detection model; then, the work feature sequence is encoded in the work intensity detection model to obtain a coding vector sequence; then, the coding vector sequence is processed for global feature dependency capture through the multi-head attention mechanism, so that the work intensity detection model can capture the potential dependency between the terminal usage feature vector and the physiological feature vector (for example, the keyboard input frequency may also change while the heart rate fluctuates greatly), thereby obtaining the global context feature; finally, the global context feature is processed for work intensity prediction to obtain a more accurate work intensity index.

[0079] In step S202 of some embodiments, the working feature sequence may be embedded to obtain an embedded vector sequence; and the embedded vector sequence may be positionally encoded to obtain an encoded vector sequence.

[0080] The work feature sequence is encoded. Specifically, the work feature sequence can be first embedded through the embedding layer of the work intensity detection model, so as to map the work feature sequence to a high-dimensional space and obtain an embedded vector sequence; then the embedded vector sequence is positionally encoded to obtain an encoded vector sequence, thereby ensuring that the work intensity detection model can understand the temporal relationship of the work feature vectors in the work feature sequence.

[0081] In step S204 of some embodiments, the global context features may be transformed to obtain deep-level global context features; the deep-level global context features may be pooled to obtain high-level global context features; and the high-level global context features may be fully connected to obtain a work intensity index.

[0082] The global context features are processed for work intensity prediction. Specifically, the global context features can be first transformed nonlinearly through the feed-forward neural network (FFN) of the work intensity detection model to obtain more complex deep global context features; the deep global context features are then pooled through the pooling layer of the work intensity detection model to obtain high-level global context features that can represent the overall work process; finally, the high-level global context features are fully connected through the fully connected layer of the work intensity detection model to obtain the work intensity index for output, thereby ensuring the validity and accuracy of the work intensity index.

[0083] Before step S104 of some embodiments, the large language model may be pre-trained. Taking BERT as an example, the training process of the large language model includes: obtaining the work feature sequence samples of the work data samples and the actual work intensity index of the work data samples; inputting the work feature sequence samples into the BERT model for iterative training, effectively learning the predicted work intensity index of the work data samples; calculating the loss between the actual work intensity index and the predicted work intensity index, thereby calculating the loss function of the BERT model according to the loss, and then updating the parameters of the BERT model according to the loss function, until the number of iterations is greater than the preset number threshold, stopping the training of the BERT model, and obtaining a trained work intensity detection model for more refined and faster automated work intensity detection.

[0084] See also Figure 3 In some embodiments, after step S104, it may include but is not limited to steps S301 to S302.

[0085] Step S301: Generate a work intensity report for a user, the work intensity report including a work intensity index, work intensity evaluation information, and work arrangement suggestion information;

[0086] Step S302: Visually display the work intensity report.

[0087] After the work intensity index of the user is detected by the work intensity detection model, a personalized work intensity report can be generated for the user, wherein the work intensity report includes the work intensity index, and work intensity evaluation information and work arrangement suggestion information generated based on the work intensity index and combined with work data.

[0088] For example, if the user's work intensity index is 8 points, the work intensity evaluation information may be "the work intensity is high, please take proper rest and adjust", and the work arrangement suggestion information may be "try the Pomodoro Technique", "please optimize the distribution of work tasks" or "please try to arrange high-priority work tasks in the morning".

[0089] The work intensity report can be displayed visually, allowing users to intuitively view their work intensity index so that they can understand their work intensity in a timely manner and make necessary adjustments to avoid excessive fatigue and improve work efficiency.

[0090] See also Figure 4 , Figure 4 It is a schematic block diagram of a work intensity detection device provided in an embodiment of the present application. The work intensity detection device can be configured in a computer device to execute the aforementioned work intensity detection method.

[0091] like Figure 4 As shown, the work intensity detection device 400 includes: a work data acquisition module 401, a time alignment module 402, a work feature extraction module 403 and a work intensity detection module 404.

[0092] A work data acquisition module 401 is used to acquire the user's work data, wherein the work data includes terminal usage data and physiological data in a working state collected by a wearable device;

[0093] A time alignment module 402 is used to perform time alignment processing on the terminal usage data and the physiological data to obtain target working data;

[0094] A working feature extraction module 403 is used to perform feature extraction processing on the target working data to obtain a working feature sequence;

[0095] The work intensity detection module 404 is used to perform work intensity detection on the work feature sequence through a work intensity detection model based on the Transformer architecture to obtain the work intensity index of the user.

[0096] In one embodiment, the work intensity detection module 404 is further used to:

[0097] inputting the work feature sequence into the work intensity detection model;

[0098] Performing encoding processing on the working feature sequence to obtain an encoding vector sequence;

[0099] Through a multi-head attention mechanism, the encoding vector sequence is processed by capturing global feature dependencies to obtain global context features;

[0100] The global context feature is processed for work intensity prediction to obtain the work intensity index.

[0101] In one embodiment, the work intensity detection module 404 is further used to:

[0102] Embedding the working feature sequence to obtain an embedding vector sequence;

[0103] Perform position encoding on the embedded vector sequence to obtain an encoded vector sequence.

[0104] In one embodiment, the work intensity detection module 404 is further used to:

[0105] Transforming the global context features to obtain deep-level global context features;

[0106] Performing pooling processing on the deep-level global context features to obtain high-level global context features;

[0107] The high-level global context features are subjected to full connection operation processing to obtain the work intensity index.

[0108] In one embodiment, the working feature extraction module 403 is further used to:

[0109] Performing cleaning processing on the target work data to obtain cleaning work data, wherein the cleaning work data includes cleaning terminal usage data and cleaning physiological data;

[0110] Performing time series feature extraction processing on the cleaning terminal usage data to obtain a terminal usage feature sequence;

[0111] Performing time-frequency domain feature extraction processing on the clean physiological data to obtain a physiological feature sequence;

[0112] The terminal usage feature sequence and the physiological feature sequence are fused to obtain the working feature sequence, which includes a time series and a working feature vector.

[0113] In one embodiment, the working feature extraction module 403 is further used to:

[0114] Performing denoising processing on the target working data to obtain denoised target working data;

[0115] Performing missing value filling processing on the denoised target working data to obtain filled target working data;

[0116] Performing outlier correction processing on the filled target working data to obtain corrected target working data;

[0117] The corrected target working data is standardized to obtain the cleaning working data.

[0118] In one embodiment, the work intensity detection device 400 further includes a work intensity report generating module and a work intensity report displaying module.

[0119] The work intensity report generating module is used to generate a work intensity report of the user, wherein the work intensity report includes a work intensity index, work intensity evaluation information and work arrangement suggestion information;

[0120] The work intensity report display module is used to visually display the work intensity report.

[0121] Among them, each module in the above-mentioned work intensity detection device 400 corresponds to each step in the above-mentioned work intensity detection method embodiment, and its functions and implementation processes are no longer repeated here one by one.

[0122] The work intensity detection device 400 can execute the work intensity detection method provided in the embodiment of the present application, and therefore can achieve the beneficial effects that can be achieved by the work intensity detection method provided in the embodiment of the present application. Please refer to the previous embodiment for details and will not be repeated here.

[0123] The method and apparatus of the present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0124] Exemplarily, the above method and apparatus may be implemented in the form of a computer program. The computer program may be implemented in the form of a computer program. Figure 5 Runs on the computer device shown.

[0125] See also Figure 5 , Figure 5 It is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application.

[0126] See also Figure 5 The computer device includes a processor and a memory connected via a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0127] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0128] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any work intensity detection method.

[0129] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0130] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0131] Acquire the user's work data, wherein the work data includes terminal usage data and physiological data in the working state collected by the wearable device;

[0132] Performing time alignment processing on the terminal usage data and the physiological data to obtain target working data;

[0133] Performing feature extraction processing on the target working data to obtain a working feature sequence;

[0134] The work intensity detection model based on the Transformer architecture is used to perform work intensity detection on the work feature sequence to obtain the work intensity index of the user.

[0135] In one embodiment, when the processor implements the work intensity detection model based on the Transformer architecture to perform work intensity detection on the work feature sequence to obtain the work intensity index of the user, it is used to implement:

[0136] inputting the work feature sequence into the work intensity detection model;

[0137] Performing encoding processing on the working feature sequence to obtain an encoding vector sequence;

[0138] Through a multi-head attention mechanism, the encoding vector sequence is processed by capturing global feature dependencies to obtain global context features;

[0139] The global context feature is subjected to work intensity prediction processing to obtain the work intensity index. In one embodiment, when the processor implements the encoding processing of the work feature sequence to obtain the encoding vector sequence, it is used to implement:

[0140] Embedding the working feature sequence to obtain an embedding vector sequence;

[0141] The embedded vector sequence is positionally encoded to obtain an encoded vector sequence.

[0142] In one embodiment, when the processor performs work intensity prediction processing on the global context feature to obtain the work intensity index, it is used to implement:

[0143] Transforming the global context features to obtain deep-level global context features;

[0144] Performing pooling processing on the deep-level global context features to obtain high-level global context features;

[0145] The high-level global context features are subjected to full connection operation processing to obtain the work intensity index.

[0146] In one embodiment, when the processor performs feature extraction processing on the target working data to obtain a working feature sequence, it is used to implement:

[0147] Performing cleaning processing on the target work data to obtain cleaning work data, wherein the cleaning work data includes cleaning terminal usage data and cleaning physiological data;

[0148] Performing time series feature extraction processing on the cleaning terminal usage data to obtain a terminal usage feature sequence;

[0149] Performing time-frequency domain feature extraction processing on the clean physiological data to obtain a physiological feature sequence;

[0150] The terminal usage feature sequence and the physiological feature sequence are fused to obtain the working feature sequence, which includes a time series and a working feature vector.

[0151] In one embodiment, when the processor performs cleaning processing on the target working data to obtain clean working data, it is used to implement:

[0152] Performing denoising processing on the target working data to obtain denoised target working data;

[0153] Performing missing value filling processing on the denoised target working data to obtain filled target working data;

[0154] Performing outlier correction processing on the filled target working data to obtain corrected target working data;

[0155] The corrected target working data is standardized to obtain the cleaning working data.

[0156] In one embodiment, after implementing the work intensity detection model based on the Transformer architecture to perform work intensity detection on the sleep feature subsequence to obtain the work intensity index of the user, the processor is further configured to implement:

[0157] Generating a work intensity report of the user, wherein the work intensity report includes a work intensity index, work intensity evaluation information, and work arrangement suggestion information;

[0158] The work intensity report is visually displayed.

[0159] The computer device can execute the work intensity detection method provided in the embodiment of the present application, and therefore can achieve the beneficial effects that can be achieved by the work intensity detection method provided in the embodiment of the present application. Please refer to the previous embodiment for details and will not be repeated here.

[0160] An embodiment of the present application also provides a computer-readable storage medium.

[0161] The computer-readable storage medium of the present application stores a computer program, and when the computer program is executed by a processor, the steps of the work intensity detection method as described above are implemented.

[0162] The computer-readable storage medium may be an internal storage unit of the work intensity detection device or computer device described in the above embodiments, such as a hard disk or memory of the work intensity detection device or computer device. The computer-readable storage medium may also be an external storage device of the work intensity detection device or computer device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital card (Secure Digital Card, SD Card), a flash card (Flash Card), etc., equipped on the work intensity detection device or computer device.

[0163] Since the computer program stored in the computer-readable storage medium can execute any one of the work intensity detection methods provided in the embodiments of the present application, the beneficial effects that can be achieved by any one of the work intensity detection methods provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0164] Furthermore, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, etc.; the data storage area may store data created according to the use of the blockchain node, etc.

[0165] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0166] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0167] The above description is only a specific implementation mode of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and these modifications or substitutions should be included in the protection scope of the present application.

Claims

1. A working intensity detection method, characterized in that: The method comprises: Acquire the user's work data, wherein the work data includes terminal usage data and physiological data in the working state collected by the wearable device; Performing time alignment processing on the terminal usage data and the physiological data to obtain target working data; Performing feature extraction processing on the target working data to obtain a working feature sequence; The work intensity detection model based on the Transformer architecture is used to perform work intensity detection on the work feature sequence to obtain the work intensity index of the user.

2. The method according to claim 1, characterized in that The work intensity detection model based on the Transformer architecture is used to perform work intensity detection on the work feature sequence to obtain the work intensity index of the user, including: inputting the work feature sequence into the work intensity detection model; Performing encoding processing on the working feature sequence to obtain an encoding vector sequence; Through a multi-head attention mechanism, the encoding vector sequence is processed by capturing global feature dependencies to obtain global context features; The global context feature is processed for work intensity prediction to obtain the work intensity index.

3. The method according to claim 2, characterized in that The encoding process of the working feature sequence to obtain the encoding vector sequence includes: Embedding the working feature sequence to obtain an embedding vector sequence; The embedded vector sequence is positionally encoded to obtain an encoded vector sequence.

4. The method according to claim 2, characterized in that: The performing work intensity prediction processing on the global context feature to obtain the work intensity index includes: Transforming the global context features to obtain deep-level global context features; Performing pooling processing on the deep-level global context features to obtain high-level global context features; The high-level global context features are subjected to full connection operation processing to obtain the work intensity index.

5. The method according to claim 1, characterized in that The step of performing feature extraction processing on the target working data to obtain a working feature sequence includes: Performing cleaning processing on the target work data to obtain cleaning work data, wherein the cleaning work data includes cleaning terminal usage data and cleaning physiological data; Performing time series feature extraction processing on the cleaning terminal usage data to obtain a terminal usage feature sequence; Performing time-frequency domain feature extraction processing on the clean physiological data to obtain a physiological feature sequence; The terminal usage feature sequence and the physiological feature sequence are fused to obtain the working feature sequence, which includes a time series and a working feature vector.

6. The method according to claim 5, characterized in that The cleaning process of the target working data to obtain clean working data includes: Performing denoising processing on the target working data to obtain denoised target working data; Performing missing value filling processing on the denoised target working data to obtain filled target working data; Performing outlier correction processing on the filled target working data to obtain corrected target working data; The corrected target working data is standardized to obtain the cleaning working data.

7. The method according to any one of claims 1 to 6, characterized in that: After performing work intensity detection on the sleep feature subsequence by using a work intensity detection model based on a Transformer architecture to obtain the work intensity index of the user, the method further includes: Generating a work intensity report of the user, wherein the work intensity report includes a work intensity index, work intensity evaluation information, and work arrangement suggestion information; The work intensity report is visually displayed.

8. A working intensity detection device, characterized in that: The device comprises: A working data acquisition module, used to acquire the user's working data, wherein the working data includes terminal usage data and physiological data collected by the wearable device in the working state; A time alignment module, used for performing time alignment processing on the terminal usage data and the physiological data to obtain target working data; A working feature extraction module, used for performing feature extraction processing on the target working data to obtain a working feature sequence; The work intensity detection module is used to perform work intensity detection on the work feature sequence through a work intensity detection model based on the Transformer architecture to obtain the work intensity index of the user.

9. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the work intensity detection method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the work intensity detection method according to any one of claims 1 to 7 is implemented.

Images