WO2020168451A1

WO2020168451A1 - Sleep prediction method and apparatus, storage medium, and electronic device

Info

Publication number: WO2020168451A1
Application number: PCT/CN2019/075361
Authority: WO
Inventors: 戴堃; 张寅祥; 陆天洋; 帅朝春; 吴建文
Original assignee: 深圳市欢太科技有限公司; Oppo广东移动通信有限公司
Priority date: 2019-02-18
Filing date: 2019-02-18
Publication date: 2020-08-27
Also published as: CN113170018A

Abstract

A sleep prediction method: when a preset sleep prediction condition is currently met, an electronic device can acquire the data required for performing sleep prediction for a user and acquire a pre-trained sleep prediction model, and thereby perform sleep prediction for the user on the basis of the acquired data required for performing sleep prediction for the user and the sleep prediction model to obtain a prediction result, thus increasing the accuracy of sleep prediction for the user.

Description

Sleep prediction method, device, storage medium and electronic equipment

Technical field

This application belongs to the field of computer technology, and in particular relates to a sleep prediction method, device, storage medium, and electronic equipment.

Background technique

At present, electronic devices such as tablet computers and mobile phones are configured to perform system updates while the user is sleeping, and other operations that affect the user's use or take a long time to avoid affecting the user's use. For this reason, the related technology achieves the foregoing purpose by predicting the user's sleep, such as predicting the user's sleep interval, etc. However, the accuracy of the related technology for predicting the user's sleep is low.

Summary of the invention

The embodiments of the present application provide a sleep prediction method, device, storage medium, and electronic device, which can enable the electronic device to accurately predict the user's sleep.

In the first aspect, an embodiment of the present application provides a sleep prediction method applied to an electronic device, including:

Determine whether the preset sleep prediction conditions are currently met;

If yes, obtain the screen on and off data of the electronic device, and obtain the user's work and rest behavior data and work schedule data;

Obtain a pre-trained sleep prediction model;

Perform sleep prediction on the user according to the screen on and off data, the work and rest behavior data, the work and rest plan data, and the sleep prediction model to obtain a prediction result.

In the second aspect, an embodiment of the present application provides a sleep prediction device applied to an electronic device, including:

The condition judgment module is used to judge whether the preset sleep prediction condition is currently met;

The data acquisition module is used to acquire the screen on and off data of the electronic device when the judgment result of the condition judgment module is yes, and acquire the user's work and rest behavior data and work and rest plan data;

A model acquisition module, configured to select a target sleep prediction model corresponding to the current usage scenario from a set of sleep prediction models;

The sleep prediction module is configured to perform sleep prediction on the user according to the screen on and off data, the work and rest behavior data, the work and rest plan data, and the sleep prediction model to obtain a prediction result.

In a third aspect, an embodiment of the present application provides a storage medium on which a computer program is stored. When the computer program is executed on a computer, the computer is caused to execute the sleep prediction method provided in the embodiment of the present application. step.

In a fourth aspect, an embodiment of the present application provides an electronic device, including a memory and a processor, and the processor is configured to execute the following by calling a computer program stored in the memory:

Determine whether the preset sleep prediction conditions are currently met;

If yes, obtain the screen on and off data of the electronic device, and obtain the user's work and rest behavior data and work and rest plan data;

Obtain a pre-trained sleep prediction model;

In the application embodiment, the electronic device can obtain the screen on and off data of the electronic device when it currently meets the preset sleep prediction conditions, and obtain the user's work and rest behavior data and work and rest plan data. In addition, it can also obtain the pre-trained sleep prediction The model is used to predict the user's sleep based on the acquired screen on and off data, work and rest behavior data, work and rest plan data, and sleep prediction model, and obtain the prediction result, which can improve the accuracy of sleep prediction for the user.

Description of the drawings

The following detailed description of specific implementations of the present application in conjunction with the accompanying drawings will make the technical solutions of the present application and its beneficial effects obvious.

FIG. 1 is a schematic flowchart of a sleep prediction method provided by an embodiment of the present application.

FIG. 2 is a schematic diagram of another flow chart of a sleep prediction method provided by an embodiment of the present application.

Figure 3 is a schematic diagram of an electronic device acquiring a sleep prediction model in an embodiment of the present application.

Fig. 4 is a schematic diagram of an operation configuration interface provided in an embodiment of the present application.

FIG. 5 is a schematic diagram of sleep prediction based on the acquired screen on and off data, work and rest behavior data, work and rest plan data, and sleep prediction model in an embodiment of the present application.

Fig. 6 is a schematic structural diagram of a sleep prediction device provided by an embodiment of the present application.

FIG. 7 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

FIG. 8 is a schematic diagram of another structure of an electronic device provided by an embodiment of the present application.

detailed description

Please refer to the drawings, in which the same component symbols represent the same components, and the principle of the present application is implemented in an appropriate computing environment for illustration. The following description is based on the exemplified specific embodiments of the present application, which should not be regarded as limiting other specific embodiments that are not described in detail herein.

Please refer to FIG. 1, which is a schematic flowchart of a sleep prediction method provided by an embodiment of the present application. The sleep prediction method can be applied to electronic devices. The process of the sleep prediction method may include:

In 101, it is determined whether a preset sleep prediction condition is currently met.

It should be noted that there are no specific restrictions on the setting of sleep prediction conditions in the embodiments of the present application, and can be set by those of ordinary skill in the art according to actual needs. For example, you can set the sleep prediction condition that the environmental brightness of the current environment of the electronic device is lower than the preset brightness. In this way, the electronic device can detect the environmental brightness of its environment in real time, for example, through the set ambient light sensor. The environment brightness of the environment is detected, and when the environment brightness of the environment is lower than the preset brightness, it is determined that the sleep prediction condition is currently met.

In 102, if yes, obtain the screen on and off data of the electronic device, and obtain the user's work and rest behavior data and work and rest plan data.

In the embodiment of the present application, the electronic device triggers the sleep prediction of the user when it determines that the preset sleep prediction condition is currently met. First, the electronic device acquires the data required for the sleep prediction of the user. The data required for the sleep prediction of the user includes at least the screen on and off data of the electronic device and the user's work and rest behavior data and work schedule data.

In 103, a pre-trained sleep prediction model is obtained.

In the embodiment of the present application, a sleep prediction model for predicting sleep of the user is also pre-trained, where the sleep prediction model can be stored locally in the electronic device or in a remote server. In this way, after the electronic device obtains the data required for the sleep prediction of the user, it further obtains the sleep prediction model used for the sleep prediction of the user locally, or obtains the sleep prediction model for the user from the remote server. Sleep prediction model.

It should be noted that the sleep prediction model is obtained through machine learning algorithm training in advance, and the machine learning algorithm can realize various functions through continuous feature learning. For example, it can be based on the user's work and rest behavior data, work schedule, and the on-off screen of electronic equipment. The data makes sleep predictions for users. Among them, machine learning algorithms may include: decision tree models, logistic regression models, Bayes models, neural network models, clustering models, and so on.

The algorithm types of machine learning algorithms can be divided according to various situations. For example, machine learning algorithms can be divided into supervised learning algorithms, non-supervised learning algorithms, semi-supervised learning algorithms, reinforcement learning algorithms, etc. based on learning methods.

Under supervised learning, the input data is called "training data", and each set of training data has a clear identification or result, such as "spam" and "non-spam" in the anti-spam system, and recognition of handwritten digits. "1, 2, 3, 4" etc. Common application scenarios for supervised learning are classification problems and regression problems. Common algorithms include Logistic Regression and Back Propagation Neural Network.

In unsupervised learning, the data is not specifically identified, the model is to infer some internal structure of the data. Common application scenarios include the learning of association rules and clustering. Common algorithms include Apriori algorithm and k-Means algorithm.

Semi-supervised learning algorithm. In this learning mode, the input data is partially identified. This learning model can be used for type recognition, but the model first needs to learn the internal structure of the data in order to organize the data reasonably for prediction. Application scenarios include classification and regression. Algorithms include some extensions to commonly used supervised learning algorithms. These algorithms first try to model unidentified data, and then predict the identified data on this basis. Graph inference algorithm (Graph Inference) or Laplacian SVM (Laplacian SVM), etc.

Reinforcement learning algorithm. In this learning mode, the input data is used as feedback to the model. Unlike the supervised model, the input data is only used as a way to check whether the model is right or wrong. Under reinforcement learning, the input data is directly fed back to the model. The model must be adjusted for this immediately. Common application scenarios include dynamic systems and robot control. Common algorithms include Q-Learning and Temporal difference learning.

In addition, machine learning algorithms can also be divided into:

Regression algorithm, common regression algorithms include: Ordinary Least Square, Logistic Regression, Stepwise Regression, Multivariate Adaptive Regression Splines and Local Scatter Smoothing Estimate (Locally Estimated Scatterplot Smoothing).

Example-based algorithms include k-Nearest Neighbor (KNN), Learning Vector Quantization (LVQ), and Self-Organizing Map (SOM).

Regularization methods, common algorithms include: Ridge Regression, Least Absolute Shrinkage and Selection Operator (LASSO), and Elastic Net (Elastic Net).

Decision tree algorithm, common algorithms include: Classification and Regression Tree (CART), ID3 (Iterative Dichotomiser 3), C4.5, Chi-squared Automatic Interaction Detection (CHAID), Decision Stump, Random Forest (Random Forest), Multiple Adaptive Regression Spline (MARS) and Gradient Boosting Machine (GBM).

Bayesian method algorithms include: Naive Bayes algorithm, Averaged One-Dependence Estimators (AODE), and Bayesian Belief Network (BBN).

In 104, a sleep prediction is performed on the user according to the obtained screen on and off data, work and rest behavior data, work and rest plan data, and sleep prediction model, and the prediction result is obtained.

In the embodiments of the present application, after the electronic device obtains the screen on and off data, the user's work and rest behavior data, and the work schedule data of the electronic device, and obtains the sleep prediction model, it can be based on the on and off screen data, work and rest behavior data, The work and rest plan data and the sleep prediction model are used to predict the user's sleep and obtain the prediction results. In the sleep prediction, the weight of the screen on and off data is greater than the weight of the work and rest behavior data and the work plan data.

It should be noted that the sleep prediction for the user includes but is not limited to the time when the user enters sleep, the time when the user ends sleep, and the sleep interval composed of the time when the user enters sleep and the time when sleep ends. For example, according to the sleep prediction model to predict the user’s sleep, the time when the user enters sleep is 23:30 of the day, the time of receiving sleep is 06:60 the next day, and the corresponding user's sleep interval is 23:30-06 the next day :60.

It can be seen from the above that the electronic device in the embodiment of the present application can obtain the screen on and off data of the electronic device when it currently meets the preset sleep prediction conditions, and obtain the user's work and rest behavior data and work and rest plan data. In addition, it also obtains The pre-trained sleep prediction model can predict the user's sleep based on the acquired screen on and off data, work and rest behavior data, work schedule data, and sleep prediction model, and obtain the prediction result, which can improve the accuracy of sleep prediction for the user.

Please refer to FIG. 2, which is a schematic diagram of another flow chart of the sleep prediction method provided by an embodiment of the application. The sleep prediction method can be applied to electronic devices. The process of the sleep prediction method may include:

In 201, the electronic device determines whether a preset sleep prediction condition is currently met.

It should be noted that there is no specific limitation on the setting of the sleep recognition condition in the embodiment of the present application, and can be set by a person of ordinary skill in the art according to actual needs.

For example, as an optional implementation manner, the sleep recognition condition can be configured as:

The duration of the static state reaches the preset duration.

Among them, the electronic device can start a timer for timing while entering a static state (for example, the electronic device can detect whether there is acceleration in any direction according to the built-in three-axis acceleration sensor, and if it does not exist, it is determined to be in a static state). The timing duration of the timer represents the continuous duration of the electronic device in the static state, where the electronic device stops counting the timer and resets the timer when the timing duration of the timer reaches the preset duration or exits the static state. In this way, when the timing duration of the timer reaches the preset duration, that is, when the duration of its static state reaches the preset duration, the electronic device determines that the sleep recognition condition is currently met.

As another optional implementation manner, the sleep recognition condition may be configured as:

The preset sleep prediction moment is reached.

Among them, the embodiment of the present application does not specifically limit the value of the sleep prediction time, which can be configured by a person of ordinary skill in the art according to actual needs. For example, it can be fixedly set to 21:00 every natural day, so that the electronic device After reaching 21:00 of each natural day, it is determined that the sleep recognition condition is satisfied; for another example, the sleep prediction time can also be set to a time 30 minutes before the time when the user enters sleep that was predicted last time.

As yet another optional implementation manner, the sleep recognition condition may be configured as:

The ambient brightness of the environment is lower than or equal to the preset brightness.

Among them, there are no specific restrictions on the value of the preset brightness in the embodiments of the present application, and can be configured by those of ordinary skill in the art according to actual needs. For example, the preset brightness can be configured to 300 nits, so that the electronic device can use it The configured ambient light sensor detects the ambient brightness of the environment in real time. When the ambient brightness of the environment is detected to be lower than or equal to 300 nits, it is determined that the sleep recognition conditions are met.

In 202, if yes, the electronic device obtains the on-off screen data of the electronic device, and obtains the user's work and rest behavior data and work and rest plan data.

Among them, the screen on and off data includes, but is not limited to, the switching time used to describe the switching time of the electronic device from the off-screen to the on-screen state, the switching time from the on-screen to the off-screen, and the switching from the off-screen to the on-screen state. The screen-on duration and screen-off duration obtained by "switching time" and "switching time for switching from on to off".

The work and rest behavior data includes but is not limited to data describing when and how the user rests (such as sleep, nap, etc.), data describing when the user exercises and how to exercise data, etc.

Work and rest plan data includes but is not limited to data describing when the user plans to do something (such as schedule), data describing when the user plans to take a break, and data describing when the user ends the break, etc.

In 203, the electronic device determines the current usage scenario of the electronic device.

In 204, the electronic device obtains a sleep prediction model corresponding to the current use scenario from the multiple pre-trained sleep prediction models, where different sleep prediction models in the multiple sleep prediction models correspond to different use scenarios.

It should be noted that in the embodiments of the present application, multiple sleep prediction models are pre-trained, and different sleep prediction models are suitable for predicting sleep of users in different usage scenarios, where the usage scenarios describe where the user uses the electronic device. Scenes, including but not limited to home vacation scenes, travel scenes, work trip scenes and daily work scenes.

In this way, in order to be able to predict the user’s sleep more accurately, after the electronic device obtains the data required for the user’s sleep prediction, it further determines the current usage scenario of the electronic device, thereby obtaining it from multiple pre-trained sleep prediction models The sleep prediction model corresponding to the current usage scenario is used for subsequent sleep prediction of the user.

It should be noted that the aforementioned pre-trained multiple sleep prediction models can all be stored locally in the electronic device, or all can be stored in a remote server, or partly stored locally in the electronic device and partly stored in a remote server. .

For example, referring to Figure 3, there are four pre-trained sleep prediction models stored in the local memory of the electronic device, namely A sleep prediction model suitable for sleep prediction in the home vacation scene, and sleep prediction model suitable for the sleep prediction in the travel scene B sleep prediction model, C sleep prediction model suitable for sleep prediction in work travel scenarios, and D sleep prediction model suitable for sleep prediction in daily work scenarios. If the electronic device determines that its current use scene is a home vacation scene, it will obtain a sleep prediction model for sleep prediction of the user; if the electronic device determines that its current use scene is a travel scene, it will obtain a sleep prediction model for the user Perform sleep prediction; if the electronic device determines that its current use scene is a work trip scenario, obtain the C sleep prediction model for sleep prediction of the user; if the electronic device determines that its current use scene is a daily work scenario, obtain a D sleep prediction model Used to predict user sleep.

In 205, the electronic device predicts the user's sleep based on the acquired screen on and off data, work and rest behavior data, work and rest plan data, and sleep prediction model, and obtains the prediction result.

In the embodiments of the present application, after the electronic device obtains the screen on and off data, the user's work and rest behavior data, and the work schedule data of the electronic device, and obtains the sleep prediction model, it can be based on the on and off screen data, work and rest behavior data, The work and rest plan data and the sleep prediction model are used to predict the user's sleep and obtain the prediction results. In the sleep prediction, the weight of the screen on and off data is greater than the weight of the work and rest behavior data and the work plan data. It should be noted that the sleep prediction of the user includes but is not limited to the time when the user enters sleep, the time when the user ends sleep, and the sleep interval composed of the time when the user enters sleep and the time when sleep ends. For example, according to the sleep prediction model to predict the user’s sleep, the time when the user enters sleep is 23:30 of the day, the time of receiving sleep is 06:60 the next day, and the corresponding user's sleep interval is 23:30-06 the next day :60.

In one embodiment, the prediction result is the user's sleep interval. After the obtained screen on/off data, work and rest behavior data, work schedule data, and sleep prediction model are obtained, sleep prediction is performed on the user, and after the prediction result is obtained, the following can be executed:

If it reaches the sleep interval, the electronic device performs a preset operation.

In the embodiment of the present application, after predicting the sleeping interval of the user, if the electronic device reaches the predicted sleeping interval, it executes a preset operation configured in advance and executed when the user is in a sleeping state. Wherein, the preset operation includes but is not limited to at least one of a system update operation, an application update operation, and a power consumption control operation, which may be manually configured by the user or may be configured by the electronic device by default.

For example, the electronic device can configure the system update operation as a preset operation, so as to perform the system update operation when the predicted sleep interval is reached, and update the system to the latest version; the electronic device can also configure the application update operation as a preset operation, thereby The application update operation is performed when the predicted sleep interval is reached, and the installed application is updated to the latest version; the electronic device can also configure the "application preset power consumption control strategy for reducing power consumption" as a preset operation, Thus, when the predicted sleep interval is reached, the preset power consumption control strategy for reducing power consumption is applied, and the power consumption of the electronic device is reduced, and so on.

For another example, please refer to Figure 4. The electronic device provides a preset operation configuration interface. As shown in Figure 4, the preset operation configuration interface includes the prompt message "Please select the operation performed during sleep", operation selection box, drop-down button, drop-down Menu, OK button, and Cancel button. Among them, the drop-down menu is called out according to the user's click operation on the drop-down button. The drop-down menu provides various operations that the electronic device can perform during the user's sleep interval, as shown in the system update in Figure 4 Operation, application update operations, etc., the user can select the operation performed by the electronic device in the user's sleep interval according to actual needs, and after selecting the operation that needs to be performed by the electronic device in the user's sleep interval, click the OK button to instruct the electronic device to The operation selected by the user is the aforementioned preset operation. Or, if the user finds that there is no need for an operation performed by the electronic device during the user's sleep interval, he can click the cancel button to instruct the electronic device to perform a preset operation of the default configuration.

In one embodiment, when obtaining the user's work and rest behavior data and work rest plan data, the following can be performed:

(1) The electronic device obtains the corresponding user's motion sensor data recorded by the sports application, and generates the user's work and rest behavior data according to the obtained motion sensor data;

(2) The electronic device obtains the event schedule data configured by the user in the memo application and the alarm clock data configured by the user in the alarm clock application, and generates the user's schedule data based on the obtained event schedule data and alarm clock data.

In the embodiment of the present application, when the electronic device obtains the user's work and rest behavior data, it can obtain the corresponding user's motion sensor data recorded by the sports application (motion sensors include but are not limited to three-axis acceleration sensors, gyroscopes, and geomagnetic sensors, etc.), thereby Generate the user's work and behavior data based on the acquired motion sensor data. For example, the electronic device generates the user's work and behavior data based on the acquired motion sensor data as "The user walks 10,000 steps between 20:30-21:00, and at 21: 00-21:00 rested for 30 minutes".

When the electronic device obtains the user's schedule data, it can obtain the user's schedule data configured in the memo application and the alarm clock data configured by the user in the alarm clock application, so as to generate the user's schedule according to the obtained schedule data and alarm clock data Data, for example, the electronic device obtains the user configuration data in the memo application as "departure to visit XX customer at 08:00 the next day", and obtains the alarm clock data configured by the user in the alarm clock application as "wake up alarm clock at 06:30 the next day" , The electronic device can generate the user’s schedule data based on the acquired event scheduling data and alarm clock data as "the user plans to wake up at 06:30 the next day and set off to visit the XX customer at 08:00"

In one embodiment, according to the obtained screen on and off data, work and rest behavior data, work and rest plan data, and sleep prediction model, perform sleep prediction on the user, and when the prediction result is obtained, you can execute:

(1) The electronic device preprocesses the acquired screen on and off data, work and rest behavior data, and work and rest plan data;

(2) The electronic device inputs the preprocessed on-screen data, work and rest behavior data, and work and rest plan data into the sleep prediction model to obtain the prediction results output by the sleep prediction model for the user's sleep prediction.

Referring to Figure 5, the electronic device predicts the user's sleep based on the screen on and off data, work and rest behavior data, schedule data, and sleep prediction model, and when the prediction result is obtained, it can first compare the acquired screen on and off data, work and rest behavior The data and schedule data are preprocessed, and then the preprocessed screen on and off data, work and rest behavior data, and schedule data are input into the sleep prediction model, and the sleep prediction model is based on the input on and off screen data, work and rest behavior data And the schedule data for sleep prediction of the user, and output the prediction result.

It should be noted that when the electronic device preprocesses the acquired screen on and off data, work and rest behavior data, and schedule data, it can perform data cleaning processing on the acquired screen on and off data, work and rest behavior data, and schedule data. , Data integration processing, data transformation processing and data reduction processing.

Among them, data cleaning processing is the process of re-examining and verifying data, with the purpose of deleting duplicate information, correcting existing errors, and providing data consistency.

Data integration processing is to integrate the data of a single dimension into a higher and more abstract dimension. After the integration, more accurate, richer, and more targeted data can be obtained.

Data transformation processing requires the data to meet certain conditions when performing statistical analysis on the data. For example, in the analysis of variance, the test error is required to be independent, unbiased, uniform and normal in variance, but in actual analysis , Independence and unbiasedness are easier to satisfy, homogeneity of variance can be satisfied in most cases, and normality sometimes cannot be satisfied. At this time, if the data is properly transformed, such as square root transformation, logarithmic transformation, square root arcsine transformation, etc., the data can meet the requirements of variance analysis. This kind of data conversion is called data conversion.

Data reduction means to minimize the amount of data while maintaining the original appearance of the data as much as possible (the necessary prerequisite for completing this task is to understand the mining task and be familiar with the content of the data itself). There are two main ways of data reduction: attribute selection and data sampling, respectively, for attributes and records in the original data set.

In an embodiment, when determining the current usage scenario of the electronic device, the following may be executed:

(1) The electronic device acquires current state information of the electronic device, where the current state information includes information describing the current use state, location state, and environmental state of the electronic device;

(2) The electronic device determines, from the multiple usage scenarios, the usage scenario whose status information matches the current status information according to the prestored status information of the multiple usage scenarios, as the current usage scenario of the electronic device.

It should be noted that the current status information includes, but is not limited to, relevant information used to describe the current use status, location status, and environmental status of the electronic device. For example, an electronic device generates state information describing its use state based on gravity sensor data and acceleration sensor data, generates state information describing its position based on positioning sensor data, and generates state information describing its environment based on sound sensors and light sensors. State information, etc.

In the embodiment of the present application, the electronic device locally prestores the state information of a plurality of different usage scenarios (in other words, a plurality of different state information is used to describe a plurality of different usage scenarios), such as the state information of a home vacation scenario, and going out. Status information of travel scenes, status information of work trip scenes, and status information of daily work scenes. In this way, when the electronic device determines its current usage scenario, it determines from the multiple usage scenarios a usage scenario whose status information matches the current status information based on the prestored status information of the multiple usage scenarios, as the current usage scenario of the electronic device.

Among them, the electronic device can determine whether the two status information matches according to the similarity between the two status information. In this way, the electronic device can obtain the status of each usage scene separately when determining the usage scenarios in which the status information matches its current status information. The similarity between the information and its current state information, and the use scenario where the similarity reaches the preset similarity is determined as the use scenario that matches the state information and its current state information.

It should be noted that there is no specific limitation on the value of the preset similarity in the embodiments of the present application, and a person of ordinary skill in the art can select an appropriate value according to actual needs.

For example, suppose that the electronic device prestores state information of a home vacation scene, state information of a travel scene, state information of a work trip scene, and state information of a daily work scene, and the preset similarity is configured to be 85%. If the electronic device obtains that the status information of the home vacation scene is similar to its current status information at 40%, the status information of the outing travel scene is similar to its current status information at 45%, and the status information of the work trip scene is similar to its current status information. The similarity is 70%, and the similarity between the status information of the daily work scene and its current status information is 86%. It can be seen that the similarity between the status information of the daily work scene and the current status information of the electronic device reaches the preset similarity (85 %), the electronic device determines the daily work scene as the use scene whose status information matches its current status information.

Wherein, when the electronic device obtains the similarity between the status information of each usage scene and its current status information, the electronic device performs feature extraction on any one of the pre-stored status information of multiple usage scenes to obtain The word vector set of the state information of each usage scene is recorded, and the word vector set of the state information of each usage scene is recorded as the first word vector set. In addition, the electronic device also performs feature extraction on its current state information, and obtains the word vector set of its current state information, which is recorded as the second word vector set.

After the electronic device obtains the first word vector set of the state information of each usage scene and the second word vector set of its current state information, it calculates the difference between each first word vector set and the second word vector set. The distance is calculated as the similarity between the status information of each usage scene and its current status information.

Among them, a person of ordinary skill in the art can select any characteristic distance (such as Euclidean distance, Manhattan distance, Chebyshev distance, cosine distance, etc.) according to actual needs to measure the distance between two word vector sets.

For example, the cosine distance between the first word vector set and the second word vector set can be obtained by referring to the following formula:

Among them, e represents the cosine distance between the first word vector set and the second word vector set, f represents the first word vector set, and N represents the dimensions of the first word vector set and the second word vector set (the dimensions of the two word vector sets Same), f _i represents the word vector of the i-th dimension in the first word vector set (the state information of a usage scenario includes state information of multiple dimensions, such as use state information, location state information, environmental state information, etc., The dimension word vector is the word vector of the state information of the i-th dimension), and g _i represents the word vector of the i-th dimension in the second word vector set.

Among them, when the electronic device obtains the word vector set of its current state information and obtains the second word vector set, it can perform word segmentation operation on its current state information and then input it into the encoder neural network, which is processed by the encoder neural network and the output corresponds to the aforementioned The word vector of the current state information, correspondingly, the electronic device uses the word vector set of the current state information output by the encoder neural network as the second word vector set.

It should be noted that the embodiments of this application do not limit the specific model and topology of the encoder neural network. For example, a single-layer recurrent neural network can be used for training to obtain the encoder neural network, or a multi-layer recurrent neural network can be used. The encoder neural network obtained by training may also be trained using a convolutional neural network, or its variants, or a neural network of other network structures to obtain an encoder neural network.

Correspondingly, when the electronic device separately obtains the word vector sets of the state information of each use scene, and obtains multiple first word vector sets, it can input the state information of each use scene into the encoder neural network, and send the encoder neural network to the encoder neural network. The output word vector set of the status information of each usage scene is used as the first word vector set.

In addition, it is also possible to pre-train a usage scene recognition model for usage scene recognition, and configure the usage scene recognition model locally in the electronic device. In this way, when the electronic device determines its current usage scenario based on its current status information, it can input its current status information into the configured usage scenario recognition model, and the usage scenario recognition model can identify the usage scenario corresponding to the aforementioned current status information, and Output. Correspondingly, the electronic device uses the use scene corresponding to the aforementioned current state information output by the use scene recognition model as its current use scene.

Please refer to FIG. 6, which is a schematic structural diagram of a sleep prediction device provided by an embodiment of the application. The sleep prediction device can be applied to electronic equipment. The sleep prediction device may include: a condition determination module 401, a data acquisition module 402, a model acquisition module 403, and a sleep prediction module 404.

The condition judgment module 401 is used to judge whether the preset sleep prediction condition is currently met;

The data acquisition module 402 is configured to acquire the screen on and off data of the electronic device when the judgment result of the condition judgment module 401 is yes, and acquire the user's work and rest behavior data and work and rest plan data;

The model acquisition module 403 is used to acquire a pre-trained sleep prediction model;

The sleep prediction module 404 is configured to perform sleep prediction on the user according to the obtained screen on and off data, work and rest behavior data, work and rest plan data, and sleep prediction model to obtain a prediction result.

In an embodiment, when acquiring the user's work and rest behavior data and work rest plan data, the data acquisition module 402 may be used to:

Obtain the corresponding user's motion sensor data recorded by the sports application, and generate the user's work and rest behavior data according to the obtained motion sensor data;

Obtain the event schedule data configured by the user in the memo application, and obtain the alarm clock data configured by the user in the alarm clock application, and generate the user's schedule data based on the obtained event schedule data and alarm clock data.

In one embodiment, when performing sleep prediction on the user according to the obtained screen on and off data, work and rest behavior data, work and rest plan data, and sleep prediction model, and obtain the prediction result, the sleep prediction module 404 can be used to:

Preprocess the acquired screen on and off data, work and rest behavior data, and work and rest plan data;

Input the preprocessed on-off screen data, work and rest behavior data, and work and rest plan data into the sleep prediction model to obtain the prediction results output by the sleep prediction model for the user's sleep prediction.

In one embodiment, when preprocessing the acquired screen on and off data, work and rest behavior data, and work and rest plan data, the sleep prediction module 404 may be used to:

Perform data cleaning processing, data integration processing, data transformation processing, and data reduction processing on the acquired screen on and off data, work and rest behavior data, and work and rest plan data.

In one embodiment, the sleep prediction conditions include:

The duration of the static state reaches the preset duration;

Or, reach the preset sleep prediction time.

In an embodiment, when acquiring a pre-trained sleep prediction model, the model acquiring module 403 may be used to:

Determine the current usage scenarios of electronic equipment;

From the multiple pre-trained sleep prediction models, a sleep prediction model corresponding to the current use scenario is obtained, where different sleep prediction models in the multiple sleep prediction models correspond to different use scenarios.

In an embodiment, when determining the current usage scenario of the electronic device, the model acquisition module 403 may be used to:

Acquiring current state information of the electronic device, where the current state information includes information describing the current use state, location state, and environmental state of the electronic device;

According to the pre-stored state information of multiple use scenarios, a use scenario whose state information matches the current state information is determined from the multiple use scenarios, as the current use scenario of the electronic device.

In an embodiment, when the use scenario whose state information matches the current state information is determined from the multiple use scenarios based on the state information of multiple prestored use scenarios, the model acquisition module 403 may be used to:

The similarity between the status information of each usage scene and its current status information is respectively acquired, and the usage scene with the similarity reaching the preset similarity is determined as the usage scene matching the status information and its current status information.

In an embodiment, when acquiring the similarity between the status information of each usage scenario and its current status information, the model acquisition module 403 may be used to:

Obtain the word vector sets of the status information of each usage scene respectively to obtain multiple first word vector sets;

Obtain the word vector set of the aforementioned current state information, and obtain the second word vector set;

Calculate the distance between each first word vector set and the second word vector set;

The calculated distances are taken as the similarity between the state information of each usage scene and the aforementioned current state information.

In an embodiment, when acquiring the word vector sets of the status information of each usage scenario, and obtaining multiple first word vector sets, the model acquiring module 403 may be used to:

The state information of each use scene is input into the encoder neural network, and the word vector set of the state information of each use scene output by the encoder neural network is used as the first word vector set.

In an embodiment, when obtaining the word vector set of the aforementioned current state information to obtain the second word vector set, the model obtaining module 403 may be used to:

Input the aforementioned current state information into the encoder neural network;

The word vector set of the foregoing current state information output by the encoder neural network is used as the second word vector set.

In an embodiment, after obtaining the current state information of the electronic device, the model obtaining module 403 may also be used to:

According to the aforementioned current status information and the pre-trained usage scene recognition model, the usage scene corresponding to the aforementioned current status information is identified as the current usage scene of the electronic device.

In an embodiment, the prediction result includes the user's sleep interval, and the sleep prediction device further includes an operation execution module for:

After the sleep prediction module 404 predicts the user's sleep based on the acquired screen on and off data, work and rest behavior data, work and rest plan data, and sleep prediction model, and obtains the user's sleep interval, if the aforementioned sleep interval is reached, the preset operation is performed .

In an embodiment, the aforementioned preset operation includes at least one of a system update operation, an application update operation, and a power consumption control operation.

The embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the stored computer program is executed on a computer, the computer is caused to execute the steps in the sleep prediction method provided in the embodiment of the present application.

An embodiment of the present application further provides an electronic device including a memory and a processor, and the processor executes the steps in the sleep prediction method provided in the embodiment of the present application by calling a computer program stored in the memory.

Please refer to FIG. 7, which is a schematic structural diagram of an electronic device provided by an embodiment of the application. The electronic device may include a memory 601 and a processor 602. A person of ordinary skill in the art can understand that the structure of the electronic device shown in FIG. 7 does not constitute a limitation on the electronic device, and may include more or less components than those shown in the figure, or a combination of certain components, or different component arrangements. .

The memory 601 can be used to store application programs and data. The application program stored in the memory 601 contains executable code. Application programs can be composed of various functional modules. The processor 602 executes various functional applications and data processing by running application programs stored in the memory 601.

The processor 602 is the control center of the electronic device. It uses various interfaces and lines to connect the various parts of the entire electronic device, and executes the electronic device by running or executing the application program stored in the memory 601 and calling the data stored in the memory 601 The various functions and processing data of the electronic device can be used to monitor the electronic equipment as a whole.

In the embodiment of the present application, the processor 602 in the electronic device will load the executable code corresponding to the process of one or more audio processing programs into the memory 601 according to the following instructions, and the processor 602 will run and store the executable code The application program in the memory 601 thus executes:

Determine whether the preset sleep prediction conditions are currently met;

Obtain a pre-trained sleep prediction model;

According to the obtained screen on and off data, work and rest behavior data, work and rest plan data, and sleep prediction model, a sleep prediction is performed on the user to obtain a prediction result.

Please refer to FIG. 8. FIG. 8 is another schematic structural diagram of the electronic device provided by an embodiment of the application. The difference from the electronic device shown in FIG. 7 is that the electronic device further includes components such as an input unit 603 and an output unit 604.

The input unit 603 can be used to receive input numbers, character information or user characteristic information (such as fingerprints), and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

The output unit 604 may be used to output information input by the user or information provided to the user, such as a speaker.

Determine whether the preset sleep prediction conditions are currently met;

Obtain a pre-trained sleep prediction model;

In an embodiment, when acquiring the user's work and rest behavior data and work and rest plan data, the processor 602 may execute:

In an embodiment, when performing sleep prediction on the user according to the acquired screen on and off data, work and rest behavior data, work and rest plan data, and sleep prediction model, and when the prediction result is obtained, the processor 602 may execute:

In an embodiment, when preprocessing the acquired screen on and off data, work and rest behavior data, and work and rest plan data, the processor 602 may execute:

In one embodiment, the sleep prediction conditions include:

The duration of the static state reaches the preset duration;

Or, reach the preset sleep prediction time.

In an embodiment, when acquiring the pre-trained sleep prediction model, the processor 602 may execute:

Determine the current usage scenarios of electronic equipment;

In an implementation manner, when determining the current usage scenario of the electronic device, the processor 602 may execute:

In an embodiment, when determining a usage scenario whose status information matches the current status information from the multiple usage scenarios according to the prestored status information of multiple usage scenarios, the processor 602 may execute:

In an implementation manner, when acquiring the similarity between the status information of each usage scenario and its current status information, the processor 602 may execute:

In an embodiment, when acquiring the word vector sets of the status information of each usage scenario, and obtaining multiple first word vector sets, the processor 602 may execute:

In one embodiment, when obtaining the word vector set of the aforementioned current state information to obtain the second word vector set, the processor 602 may execute:

In an embodiment, after acquiring the current state information of the electronic device, the processor 602 may execute:

In one embodiment, the prediction result includes the user's sleep interval. After the obtained screen on and off data, work and rest behavior data, work schedule data, and sleep prediction model are obtained, sleep prediction is performed on the user, and the user's sleep interval is obtained. The device 602 can execute:

If the aforementioned sleep interval is reached, the preset operation is performed.

In the foregoing embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in an embodiment, please refer to the detailed description of the sleep prediction method above, which will not be repeated here.

The sleep prediction device/electronic device provided by the embodiment of the application belongs to the same concept as the sleep prediction method in the above embodiment. Any method provided in the sleep prediction method embodiment can be run on the sleep prediction device/electronic device. For the implementation process, refer to the embodiment of the sleep prediction method, which will not be repeated here.

It should be noted that for the sleep prediction method of the embodiment of the present application, those of ordinary skill in the art can understand that all or part of the process of implementing the sleep prediction method of the embodiment of the present application can be completed by controlling the relevant hardware through a computer program. The program may be stored in a computer readable storage medium, such as stored in a memory, and executed by at least one processor, and the execution process may include a process such as an embodiment of the sleep prediction method. Among them, the storage medium may be a magnetic disk, an optical disc, a read only memory (ROM, Read Only Memory), a random access memory (RAM, Random Access Memory), etc.

For the sleep prediction device of the embodiment of the present application, its functional modules may be integrated in one processing chip, or each module may exist alone physically, or two or more modules may be integrated in one module. The above-mentioned integrated modules can be implemented in the form of hardware or software functional modules. If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer readable storage medium, such as a read-only memory, a magnetic disk, or an optical disk.

The above describes in detail a sleep prediction method, device, storage medium, and electronic equipment provided by the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementations of the present application. The description of the above embodiments is only It is used to help understand the method and core idea of this application; at the same time, for those skilled in the art, according to the idea of this application, there will be changes in the specific implementation and scope of application. In summary, the content of this specification does not It should be understood as a limitation of this application.

Claims

A sleep prediction method applied to electronic equipment, including:

Determine whether the preset sleep prediction conditions are currently met;

If yes, obtain the screen on and off data of the electronic device, and obtain the user's work and rest behavior data and work and rest plan data;

Obtain a pre-trained sleep prediction model;

Perform sleep prediction on the user according to the screen on and off data, the work and rest behavior data, the work and rest plan data, and the sleep prediction model to obtain a prediction result.
The sleep prediction method according to claim 1, wherein said acquiring the user’s work and rest behavior data and work and rest plan data comprises:

Acquiring motion sensor data corresponding to the user recorded by the motion application, and generating the work and rest behavior data according to the motion sensor data;

Obtain the event schedule data configured by the user in the memo application, and obtain the alarm clock data configured by the user in the alarm clock application, and generate the schedule data according to the event schedule data and the alarm clock data.
The sleep prediction method according to claim 1, wherein the sleep prediction is performed on the user according to the on-off screen data, the work and rest behavior data, the work and rest plan data, and the sleep prediction model to obtain Forecast results, including:

Preprocessing the on-off screen data, the work and rest behavior data, and the work and rest plan data;

The preprocessed screen on and off data, the work and rest behavior data, and the work and rest plan data are input into the sleep prediction model to obtain a prediction result output by the sleep prediction model for the user's sleep prediction.
The sleep prediction method according to claim 3, wherein the preprocessing of the on-off screen data, the work and rest behavior data, and the work and rest plan data comprises:

Perform data cleaning processing, data integration processing, data transformation processing, and data reduction processing on the on-off screen data, the work and rest behavior data, and the work and rest plan data.
The sleep prediction method according to claim 1, wherein the sleep prediction condition comprises:

The duration of the static state reaches the preset duration;

Or, reach the preset sleep prediction time.
The sleep prediction method according to claim 1, wherein said obtaining a pre-trained sleep prediction model comprises:

Determine the current usage scenario of the electronic device;

Obtain a sleep prediction model corresponding to the current use scene from a plurality of pre-trained sleep prediction models, where different sleep prediction models in the plurality of sleep prediction models correspond to different use scenarios.
The sleep prediction method according to claim 6, wherein the determining the current usage scenario of the electronic device comprises:

Acquiring current state information of the electronic device, where the current state information includes information describing the current use state, location state, and environmental state of the electronic device;

According to the pre-stored state information of the multiple use scenarios, a use scenario whose state information matches the current state information is determined from the multiple use scenarios as the current use scenario of the electronic device.
The sleep prediction method according to claim 1, wherein the prediction result includes a user's sleep interval, and the sleep prediction model is based on the on-off screen data, the work and rest behavior data, the work and rest plan data, and the sleep prediction model , Perform sleep prediction on the user, and after obtaining the prediction result, it also includes:

If the sleep interval is reached, a preset operation is performed.
The sleep prediction method according to claim 8, wherein the preset operation includes at least one of a system update operation, an application update operation, and a power consumption control operation.
A sleep prediction device applied to electronic equipment, including:

The condition judgment module is used to judge whether the preset sleep prediction condition is currently met;

The data acquisition module is used to acquire the screen on and off data of the electronic device when the judgment result of the condition judgment module is yes, and acquire the user's work and rest behavior data and work and rest plan data;

A model acquisition module, configured to select a target sleep prediction model corresponding to the current usage scenario from a set of sleep prediction models;

The sleep prediction module is configured to perform sleep prediction on the user according to the screen on and off data, the work and rest behavior data, the work and rest plan data, and the sleep prediction model to obtain a prediction result.
A storage medium having a computer program stored thereon, wherein when the computer program is executed on a computer, the computer is caused to execute the sleep prediction method according to any one of claims 1 to 9.
An electronic device comprising a memory and a processor, wherein the processor is configured to execute the sleep prediction method according to any one of claims 1 to 9 by calling a computer program stored in the memory.