Data processing methods, apparatus, equipment and storage media

Abstract

This application discloses a data processing method, apparatus, device, and storage medium, belonging to the field of computer technology. The method includes: acquiring state data of a target object; performing multi-target identification on the state data based on the characteristics of anomalies of the target object at multiple development stages, obtaining identification results for multiple development stages; determining the confidence level of the multiple identification results based on the obtained multiple identification results, wherein the confidence level is positively correlated with the completeness of the identification of the anomaly event; and determining the event identification result based on the multiple identification results and the confidence level. In this way, the identification accuracy of each development stage can compensate for each other, weakening the influence of noise and improving the overall accuracy of anomaly event identification.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a data processing method, apparatus, device, and storage medium. Background Technology

[0002] In daily life, abnormal events may occur to the target object. When such events occur, they need to be detected and handled promptly to prevent them from escalating and causing worse consequences. The target object can refer to various entities, such as equipment or human beings. Abnormal events can be events such as malfunctions of the target object.

[0003] Currently, it's possible to collect state data of a target object. If the target object's state data differs from its normal state data, it can be determined that an abnormal event has occurred. However, noise is often introduced when collecting the target object's state data, and due to the noise issue, the accuracy of abnormal event identification is low. Summary of the Invention

[0004] This application provides a data processing method, apparatus, device, and storage medium that can improve the overall accuracy of abnormal event identification. The technical solution is as follows:

[0005] On the one hand, a data processing method is provided, the method comprising:

[0006] Obtain the state data of the target object, wherein the state data is used to represent the state of the target object;

[0007] Based on the characteristics of the abnormal events of the target object in multiple progress stages, multi-target identification is performed on the state data to obtain the identification results of the multiple progress stages. The identification results of the progress stages are used to indicate whether the characteristics of the progress stage appear in the state data and the time period of the appearance of the characteristics of the progress stage.

[0008] Based on the obtained multiple identification results, the confidence level of the multiple identification results is determined, and the confidence level is positively correlated with the completeness of the identification of the abnormal event;

[0009] Based on the multiple identification results and the confidence level, an event identification result is determined. The event identification result is used to indicate whether the abnormal event has occurred in the status data and the time period of the abnormal event.

[0010] On one hand, a data processing apparatus is provided, the apparatus comprising:

[0011] The first acquisition module is used to acquire the state data of the target object, wherein the state data is used to represent the state of the target object;

[0012] The first identification module is used to perform multi-target identification on the state data based on the characteristics of the abnormal events of the target object in multiple progress stages, and obtain the identification results of the multiple progress stages. The identification result of each progress stage is used to indicate whether the characteristics of the progress stage appear in the state data and the time period of the appearance of the characteristics of the progress stage.

[0013] The first determining module is used to determine the confidence level of the multiple identification results based on the obtained multiple identification results, wherein the confidence level is positively correlated with the completeness of the identification of the abnormal event;

[0014] The second determining module is used to determine an event identification result based on the multiple identification results and the confidence level. The event identification result is used to indicate whether the abnormal event has occurred in the status data and the time period of the abnormal event.

[0015] In one possible implementation, the second determining module is configured to, when the confidence level reaches a first confidence level threshold, determine a third occurrence time period of the abnormal event based on a first occurrence time period of the first identified progress stage and a second occurrence time period of the last identified progress stage.

[0016] The second determining module is used to determine, based on the first occurrence time period and the second occurrence time period, a first sub-data corresponding to the candidate abnormal event in the status data when the confidence level reaches the second confidence level threshold but does not reach the first confidence level threshold; and to determine a third occurrence time period of the abnormal event based on the first occurrence time period and the second occurrence time period when the first sub-data satisfies the filtering conditions that match the confidence level. The confidence level is proportional to the strictness of the filtering conditions.

[0017] If the confidence level does not reach the second confidence threshold, it is determined that the abnormal event has not occurred in the status data;

[0018] Wherein, the first confidence threshold is higher than the second confidence threshold.

[0019] In one possible implementation, the filtering criteria matching the confidence level include at least one of the following:

[0020] The duration range that matches the confidence level, wherein the duration range is used to indicate the conditions that the duration of the abnormal event needs to meet;

[0021] A range of dispersion that matches the confidence level, the range of dispersion being used to indicate the conditions that the dispersion of the RR intervals in the anomalous event needs to meet;

[0022] The range of state change counts that matches the confidence level, wherein the range of state change counts is used to indicate the conditions that the number of state changes of the target object in the abnormal event needs to meet.

[0023] In one possible implementation, the plurality of progress stages include a start stage, a main stage, and an end stage. The plurality of identification results indicate that at least one of the features of the start stage and the features of the end stage, as well as the features of the main stage, appear in the state data. The filtering conditions include a first duration range, which is a duration range greater than a first duration threshold.

[0024] The second determining module is configured to determine a fourth occurrence time period of the candidate abnormal event based on the first occurrence time period and the second occurrence time period; and if the fourth occurrence time period indicates that the duration of the candidate abnormal event is within the first duration range, determine the fourth occurrence time period as the third occurrence time period of the abnormal event.

[0025] In one possible implementation, the plurality of progress stages include a start stage, a main stage, and an end stage. The plurality of identification results indicate that the features of the start stage and the end stage appear in the state data, but the features of the main stage do not appear. The filtering conditions include a second duration range, which is a duration range that is greater than a first duration threshold and less than a second duration threshold.

[0026] The second determining module is used to determine a fourth occurrence time period of the candidate abnormal event based on the first occurrence time period and the second occurrence time period; and to determine the fourth occurrence time period as the third occurrence time period of the abnormal event if the fourth occurrence time period indicates that the duration of the candidate abnormal event is within the range of the second duration.

[0027] In one possible implementation, the first identification module is used to perform multi-target identification on the state data based on the characteristics of the multiple progress stages using an event identification model, and obtain the identification results of the multiple progress stages, wherein the event identification model is used to identify the multiple progress stages of the abnormal event.

[0028] In one possible implementation, the device further includes:

[0029] The first acquisition module is used to acquire sample data, which includes sample status data and sample identification results of the multiple progress stages. The sample identification result of any progress stage is used to indicate whether the feature of the progress stage appears in the sample status data and the time period of the appearance of the feature of the progress stage.

[0030] The first identification module is used to process the sample state data through the event identification model before training to obtain the prediction identification results of the multiple progress stages. The prediction identification result of any progress stage is used to indicate whether the event identification model before training has identified the progress stage and the time period of the occurrence of the progress stage.

[0031] The training module is used to train the event recognition model based on the difference between the sample recognition results and the predicted recognition results corresponding to the same progress stage, so as to obtain the trained event recognition model.

[0032] In one possible implementation, the device further includes:

[0033] The second acquisition module is used to acquire the second sub-data of the abnormal event from the status data based on the time period in which the abnormal event occurs;

[0034] The second identification module is used to identify noise in the second sub-data and determine the noise data in the second sub-data.

[0035] The elimination module is used to eliminate the noise data from the second sub-data to obtain the third sub-data;

[0036] The statistics module is used to perform statistical processing on the third sub-data to obtain the number of consecutive state changes, where the number of consecutive state changes represents the number of times the state of the target object in the third sub-data changes consecutively.

[0037] An update module is used to update the event identification result when the number of consecutive state changes is less than a first state change threshold. The updated event identification result indicates that the abnormal event has not appeared in the state data.

[0038] On one hand, a computer device is provided, the computer device including one or more processors and one or more memories, the one or more memories storing at least one piece of program code, the at least one piece of program code being loaded and executed by the one or more processors to perform the operations performed by the data processing method as described in any of the above possible implementations.

[0039] On one hand, a computer-readable storage medium is provided, which stores at least one piece of program code that is loaded and executed by a processor to perform the operations performed by the data processing method as described in any of the possible implementations above.

[0040] On one hand, a computer program or computer program product is provided, the computer program or computer program product comprising: computer program code, which, when executed by a computer, causes the computer to perform the operations performed by the data processing method as described in any of the above possible implementations.

[0041] The data processing method, apparatus, device, and storage medium provided in this application divides an abnormal event into multiple development stages. These stages are then identified within the state data of the target object. By assessing the completeness of the identification of each development stage, the accuracy of multiple identification results can be determined. Based on these multiple identification results and their accuracy, an event identification result is determined. In this way, the identification accuracy of each development stage can compensate for each other, weakening the influence of noise and improving the overall accuracy of abnormal event identification. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of an implementation environment provided in an embodiment of this application;

[0044] Figure 2 This is a flowchart of a data processing method provided in an embodiment of this application;

[0045] Figure 3 This is a flowchart of a data processing method provided in an embodiment of this application;

[0046] Figure 4 This is a flowchart of an atrial fibrillation identification method provided in an embodiment of this application;

[0047] Figure 5 This is a schematic diagram of sample data provided in an embodiment of this application;

[0048] Figure 6 This is a schematic diagram of a data processing device structure provided in an embodiment of this application;

[0049] Figure 7 This is a schematic diagram of another data processing device structure provided in an embodiment of this application;

[0050] Figure 8 This is a schematic diagram of the terminal structure provided in the embodiments of this application;

[0051] Figure 9This is a schematic diagram of the server structure provided in an embodiment of this application. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0053] It is understood that the terms "first," "second," etc., used in this application may be used to describe various concepts herein, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of this application, a first confidence level may be referred to as a second confidence level, and similarly, a second confidence level may be referred to as a first confidence level.

[0054] As used in this application, the terms "at least one", "multiple", "each", and "any" are used in the following ways: at least one includes one, two, or more; multiple includes two or more; each refers to each of the corresponding multiple; and any refers to any one of the multiple. For example, multiple development stages include three development stages, each refers to each of the three development stages, and any refers to any one of the three development stages, which can be the first, the second, or the third.

[0055] It should be noted that all information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.), and signals involved in this application have been authorized by the user or fully authorized by all parties, and the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. For example, the status data involved in this application was obtained with full authorization.

[0056] The data processing method provided in this application is executed by a computer device. In one possible implementation, the computer device is a terminal, such as a wearable device, mobile phone, tablet computer, desktop computer, or any other type of terminal. In some embodiments, the terminal acquires the status data of a target object, analyzes the status data, determines whether an abnormal event has occurred in the status data, and, if an abnormal event is determined to have occurred in the status data, determines the time period of the abnormal event.

[0057] In another possible implementation, the computer device is a server. For example, the server can be a single server, a server cluster consisting of several servers, or a cloud computing service center. In some embodiments, the server acquires the status data of the target object, analyzes the status data, determines whether an abnormal event has occurred in the status data, and, if it is determined that an abnormal event has occurred, determines the time period in which the abnormal event occurred.

[0058] In another possible implementation, the computer device includes a terminal and a server.

[0059] Figure 1 This is a schematic diagram of an implementation environment provided in an embodiment of this application, such as... Figure 1 As shown, the implementation environment includes a terminal 101 and a server 102. The terminal 101 and the server 102 are connected via a wireless or wired network.

[0060] Optionally, a target application provided by server 102 is installed on terminal 101, enabling terminal 101 to perform functions such as data transmission and message interaction. Optionally, the target application is an application within the operating system of terminal 101, or an application provided by a third party. For example, the target application is a status analysis application, which has the function of analyzing the running status of a target object. Of course, this status analysis application may also have other functions, such as data acquisition and sharing functions.

[0061] In some embodiments, terminal 101 collects the status data of the target object and uploads the status data to server 102. Server 102 receives the status data, analyzes the status data, obtains the event identification result, and returns the event identification result to terminal 101. Terminal 101 receives the event identification result and displays the event identification result.

[0062] The data processing method provided in this application can be applied to multiple scenarios such as equipment monitoring and medical scenarios.

[0063] For example, it can be applied in medical settings.

[0064] In this medical scenario, the target object can be a user, and the status data can be the user's physiological status data, used to represent the user's physiological state, such as the user's electrocardiogram (ECG). Abnormal events can be arrhythmia events, atrial fibrillation events, etc. Taking a user as the target object, an ECG as the status data, and an atrial fibrillation event as the abnormal event as an example, when identifying atrial fibrillation events based on ECG, dividing the atrial fibrillation event into multiple progression stages allows the accuracy of each progression stage to compensate for each other, thereby improving the overall accuracy of atrial fibrillation event identification.

[0065] For example, it can be applied to equipment monitoring scenarios.

[0066] In this device monitoring scenario, the target object can be any device, and the status data can be the device's operational status data, used to represent the device's operating state, such as voltage data, current data, and interaction data with other devices. Abnormal events can be fault events, etc. Taking the target device as the target object, operational status data as the status data, and fault events as the abnormal events as an example, when identifying whether a fault event has occurred in the target device based on operational status data, the fault event is divided into multiple development stages for identification. This allows the identification accuracy of each development stage to compensate for each other, thereby improving the overall accuracy of fault event identification.

[0067] It should be noted that the embodiments of this application are only examples of equipment monitoring scenarios and medical scenarios to illustrate the application scenarios of this application, and do not limit the application scenarios of this application.

[0068] Figure 2 This is a flowchart illustrating a data processing method provided in an embodiment of this application. This embodiment uses a computer device as an example for illustrative purposes, and includes:

[0069] 201. The computer device acquires the state data of the target object, which is used to represent the state of the target object.

[0070] In this embodiment, the target object can refer to various entities, such as devices, human bodies, etc. This embodiment does not limit the target object. State data is used to represent the state of the target object. Taking a device as an example, the state data can be data representing the operating state of the target object. Taking a human body as an example, the state data can be data representing the physiological state of the target object. Different types of state data can be configured for different types of target objects according to the needs of actual applications. This embodiment does not limit the target object or its state data.

[0071] 202. Based on the characteristics of abnormal events of the target object in multiple development stages, the computer equipment performs multi-target identification on the state data to obtain the identification results of multiple development stages. The identification results of the development stages are used to indicate whether the characteristics of the development stage appear in the state data and the time period of the appearance of the characteristics of the development stage.

[0072] Abnormal events can refer to various types of abnormal events, such as any type of equipment failure, abnormal heart rhythm events, atrial fibrillation events, etc. This application does not limit the definition of abnormal events.

[0073] Abnormal events typically last for a certain period of time, and their characteristics change during this time. Therefore, abnormal events can be divided into multiple development stages based on these changes, such as an initial stage, a main stage, and an ending stage; or, for example, an initial stage, a deterioration stage, a calming stage, and an ending stage. It should be noted that the embodiments in this application are merely illustrative examples of multiple development stages and do not limit the number of development stages. These multiple development stages can be divided according to the changes in the characteristics of the abnormal event or actual application requirements. The embodiments in this application do not limit the stage division of abnormal events.

[0074] The identification result of the progress stage indicates whether the characteristics of that progress stage appear in the state data and the time period during which those characteristics appear. Specifically, if the identification result indicates the presence of a certain progress stage characteristic in the state data, it means that the computer device has identified that progress stage from the state data. For example, if the identification result indicates the presence of the initiation stage characteristic in the state data, it means that the computer device has identified the initiation stage of the abnormal event from the state data.

[0075] 203. The computer device determines the confidence level of the multiple identification results based on the obtained results. The confidence level is positively correlated with the completeness of the identification of the abnormal event.

[0076] In some embodiments, after a computer device performs multi-target identification on the state data, it may identify each stage of the abnormal event from the state data, or it may identify some stages of the abnormal event from the state data, or it may identify no stage of the abnormal event at all.

[0077] Understandably, when a computer device identifies every stage of an abnormal event from the state data, or fails to identify any stage, the multiple identification results obtained by the computer device are relatively accurate. However, when the computer device identifies only some stages of an abnormal event from the state data, some identification results may be erroneous. Therefore, embodiments of this application determine the confidence level of multiple identification results based on the completeness of the identified abnormal event.

[0078] 204. The computer device determines the event identification result based on multiple identification results and confidence levels. The event identification result is used to indicate whether an abnormal event has occurred in the status data and the time period during which the abnormal event occurred.

[0079] In this embodiment of the application, when determining the event identification result, not only are multiple identification results considered, but also the confidence levels of the multiple identification results are considered. For example, when a computer device identifies a partial progress stage of an abnormal event from status data, if the confidence levels of multiple identification results are high, it can be determined that an abnormal event has occurred in the status data; if the confidence levels of multiple identification results are low, it can be determined that no abnormal event has occurred in the status data.

[0080] The data processing method provided in this application divides an abnormal event into multiple development stages. These stages are then identified within the state data of the target object. By assessing the completeness of the identification of each development stage, the accuracy of multiple identification results can be determined. Based on these multiple identification results and their accuracy, the final event identification result is determined. In this way, the identification accuracy of each development stage can compensate for each other, thereby improving the overall accuracy of abnormal event identification.

[0081] Figure 3 This is a flowchart illustrating a data processing method provided in an embodiment of this application. This embodiment uses a computer device as an example for illustrative purposes, and includes:

[0082] 301. The computer device acquires the state data of the target object, which is used to represent the state of the target object.

[0083] In some embodiments, the computer device is a data acquisition device, and the computer device acquires the state data of the target object, including: the computer device acquiring the state data of the target object. In some embodiments, the state data is acquired by the computer device from other devices. This application does not limit the source of the state data.

[0084] In some embodiments, the state data of the target object is the state data of the target object within a target time period. The computer device can analyze the state data for a specific time period, enabling the computer device to analyze the state data within a suspicious time period, thereby reducing the amount of data that the computer device needs to process.

[0085] In some embodiments, in order to expedite the determination of whether an abnormal event has occurred in a target object, the computer device can segment the status data, thereby obtaining the event identification results of the status data as quickly as possible.

[0086] 302. The computer device uses an event recognition model to perform multi-target recognition on the state data based on the characteristics of the abnormal event of the target object in multiple development stages, and obtains the recognition results of multiple development stages. This event recognition model is used to identify multiple development stages of abnormal events.

[0087] This event recognition model is used to identify multiple stages of an abnormal event; that is, it is a multi-classification model. The event recognition model can be a KNN (K-nearest neighbors) model, a logistic regression model, an LDA (Linear discriminant analysis) model, a QDA (Quadratic discriminant analysis) model, etc. This application does not limit the event recognition model used in its embodiments.

[0088] In some embodiments, the computer device uses an event recognition model to perform multi-target recognition on the state data based on the characteristics of abnormal events of the target object at multiple stages of development, and obtains recognition results for multiple stages of development. This includes: the computer device uses an event recognition model to obtain sub-data corresponding to multiple time windows in the state data based on time windows; extracts features from the obtained sub-data respectively to obtain features of the sub-data; classifies the features of the sub-data based on the characteristics of abnormal events at multiple stages of development to obtain categories of the sub-data; and performs statistical processing on the categories of the sub-data based on the time corresponding to each of the sub-data to obtain recognition results for multiple stages of development.

[0089] For example, if the status data is the target object's data within 30 seconds, after inputting this status data into an event recognition model, the model extracts features from the sub-data corresponding to each second of the status data, obtaining features for 30 sub-data. Based on the features of the abnormal event at multiple stages of its development, the features of these 30 sub-data are classified, resulting in multiple categories for each sub-data. For instance, the first and second sub-data are classified as non-abnormal events, the third and fourth sub-data as the initial stage of an abnormal event, the fifth to twenty-eighth sub-data as the main stage of an abnormal event, and the twenty-ninth to thirtieth sub-data as non-abnormal events. Statistical processing of these 30 sub-data categories determines that the status data shows the initial and main stages of an abnormal event; the initial stage occurs between the 3rd and 4th seconds, and the main stage occurs between the 5th and 28th seconds.

[0090] In some embodiments, when determining the category of a certain sub-data, the event recognition model can determine the probability that the sub-data belongs to that category, and determine the category corresponding to the highest probability as the category of the sub-data, or determine the category whose probability exceeds the probability threshold as the category of the sub-data. This application embodiment does not limit this.

[0091] In one possible implementation, the training process of the event recognition model includes: acquiring sample data, which includes sample state data and sample recognition results for multiple progress stages, wherein the sample recognition result for any progress stage is used to indicate whether the feature of that progress stage appears in the sample state data and the time period during which the feature of that progress stage appears; processing the sample state data using the event recognition model before training to obtain predicted recognition results for multiple progress stages, wherein the predicted recognition result for any progress stage is used to indicate whether the event recognition model before training has identified that progress stage and the time period during which that progress stage appears; and training the event recognition model based on the difference between the sample recognition result and the predicted recognition result corresponding to the same progress stage to obtain the trained event recognition model.

[0092] In some embodiments, the computer device includes a terminal and a server. The server trains the event recognition model, and the trained event recognition model is deployed to the terminal. For example, the terminal has a target application installed, which includes the model file of the event recognition model. This target application is used to identify abnormal events, and the terminal can call the model file of the event recognition model through the target application to run the event recognition model.

[0093] In other embodiments, the computer device includes a first terminal and a second terminal, where the first terminal is a technician's terminal and the second terminal is a user's terminal. The event recognition model is trained using the first terminal, and the trained event recognition model is then deployed to the second terminal.

[0094] It should be noted that the embodiments of this application do not limit the devices used to train the event recognition model or the devices used to use the event recognition model.

[0095] It should be noted that the embodiments of this application are merely illustrative examples of "using an event recognition model as an example to perform multi-target recognition on state data based on the characteristics of an abnormal event of a target object at multiple development stages, and obtaining recognition results for multiple development stages." In another embodiment, the computer device can also perform multi-target recognition on state data using multiple event recognition models, wherein the multiple event recognition models are used to identify different development stages of the abnormal event. In yet another embodiment, the computer device may not use an event recognition model, but instead use other target recognition algorithms; the embodiments of this application do not limit the algorithm used for multi-target recognition.

[0096] 303. The computer device determines the confidence level of multiple identification results based on the obtained results, and the confidence level is positively correlated with the completeness of the identification of the abnormal event.

[0097] Understandably, when a computer device identifies every stage of an abnormal event from the state data, or fails to identify any stage, the multiple identification results obtained by the computer device are relatively accurate. However, when the computer device identifies only some stages of an abnormal event from the state data, some identification results may be erroneous. Therefore, embodiments of this application determine the confidence level of multiple identification results based on the completeness of the identified abnormal event.

[0098] Understandably, when computer equipment identifies partial stages of an abnormal event's progression from status data, the reliability of the identification results varies depending on the specific stages identified. For example, an abnormal event may have multiple progression stages, including an initiation stage, a main stage, and an end stage. The initiation and end stages are relatively short, while the main stage is a longer and more stable stage. Therefore, if the computer equipment only identifies the initiation or end stage, the accuracy of the identification result will be lower; conversely, if it only identifies the main stage, the accuracy will be higher. Thus, the computer equipment can determine the confidence level of the identification result based on the combination of identified progression stages.

[0099] For example, taking the multiple development stages of an abnormal event, including the initiation stage, the main stage, and the termination stage, as an example, the determination of the confidence level of multiple identification results is explained.

[0100] If multiple identification results indicate that the start stage, main stage, and end stage have been identified from the state data, then these multiple identification results can be determined as the first level. The level of the identification results is used to represent the confidence level of the multiple identification results, with the first level representing the highest confidence level.

[0101] If multiple identification results indicate that the initial stage and the main stage or the main stage and the ending stage have been identified from the state data, then these multiple identification results can be determined to be of the second level.

[0102] If multiple identification results indicate that the main stage has been identified from the state data, then these multiple identification results can be determined to be at the third level.

[0103] If multiple identification results indicate that the start and end stages have been identified from the state data, then these multiple identification results can be determined to be at the fourth level.

[0104] If multiple identification results indicate that the start or end stage has been identified from the state data, we can determine that the multiple identification results are at level 5.

[0105] The confidence levels represented by the first, second, third, fourth, and fifth levels decrease sequentially.

[0106] It should be noted that if the computer device does not identify any progress stage from the status data, then there is no need to determine the confidence level of the identification result. In other words, if the computer device does not identify any progress stage from the status data, then no subsequent steps need to be performed.

[0107] 304. When the confidence level reaches the first confidence threshold, the computer device determines the third occurrence time of the abnormal event based on the first occurrence time of the first identified progress stage and the second occurrence time of the last identified progress stage.

[0108] In this embodiment of the application, if the confidence level of the identification result reaches the first confidence level threshold, it indicates that the identification result is credible and it can be determined that an abnormal event has occurred in the status data. At this time, it is also necessary to determine the time period of the abnormal event. Therefore, the computer device can determine the third time period of the abnormal event based on the first time period of the first identified progress stage and the second time period of the last identified progress stage.

[0109] In some embodiments, the computer device determines the third occurrence time of the anomalous event based on a first occurrence time period of the first identified progress phase and a second occurrence time period of the last identified progress phase, including: the computer device using the start time of the first identified progress phase as the start time of the anomalous event and the end time of the last identified progress phase as the end time of the anomalous event.

[0110] In some embodiments, a confidence level reaching a first confidence threshold means that multiple identification results are at the first level.

[0111] 305. When the confidence level reaches the second confidence threshold but does not reach the first confidence threshold, the computer device determines the first sub-data corresponding to the candidate abnormal event in the status data based on the first occurrence time period of the first identified progress stage and the second occurrence time period of the last identified progress stage.

[0112] Wherein, the first confidence threshold is higher than the second confidence threshold. It should be noted that the first and second confidence thresholds can be any values. Optionally, the first and second confidence thresholds are empirical values. Optionally, the first and second confidence thresholds are values ​​set by a technician. This application embodiment does not limit the first and second confidence thresholds.

[0113] If multiple identification results reach the second confidence threshold but fail to reach the first confidence threshold, it indicates that the computer device has identified a partial stage of an abnormal event. In this case, the multiple identification results may be accurate or incorrect. To more accurately determine whether an abnormal event has occurred in the status data, the identified partial stages of progress can be identified as candidate abnormal events. Further judgment can then be made based on appropriate filtering criteria.

[0114] In some embodiments, the computer device determines the first sub-data corresponding to the candidate abnormal event in the state data based on the first occurrence time period of the first identified progress stage and the second occurrence time period of the last identified progress stage, including: determining a fourth time period based on the start time of the first time period and the end time of the second time period, and determining the data corresponding to the fourth time period from the state data as the first sub-data corresponding to the candidate abnormal event.

[0115] In some embodiments, the confidence level reaching the second confidence threshold but not reaching the first confidence threshold means that the levels of multiple recognition results are the second level, the third level, or the fourth level.

[0116] 306. When the first sub-data satisfies the filtering conditions that match the confidence level, the computer device determines the third occurrence time of the abnormal event based on the first occurrence time period and the second occurrence time period, wherein the confidence level is proportional to the strictness of the filtering conditions.

[0117] In this embodiment, the lower the confidence level of the identification result, the lower the accuracy of the identification result. To accurately determine whether an abnormal event has occurred in the status data, the computer device can first identify certain stages of the identified process as candidate abnormal events. Further judgment is then made based on corresponding screening conditions, the strictness of which is directly proportional to the confidence level of the identification result.

[0118] In one possible implementation, the filtering criteria matching this confidence level include at least one of the following:

[0119] (1) The duration range that matches the confidence level, which indicates the conditions that the duration of the abnormal event needs to meet.

[0120] Abnormal events typically last for a certain duration; therefore, a time range can be set to verify candidate abnormal events.

[0121] (2) The range of dispersion that matches the confidence level, which indicates the conditions that the dispersion of the RR interval in the anomalous event needs to meet.

[0122] In some embodiments, the state data is represented as a signal wave, and the RR interval is the distance between the peaks in the signal wave. Under normal circumstances, the distances between the peaks are equal; when an abnormal event occurs, the distances between the peaks are unequal. The RR interval dispersion is used to represent the degree of unequal peak distances. The lower the confidence level, the higher the degree of dispersion represented by the discrete range in the screening criteria matching that confidence level.

[0123] (3) The range of state changes that matches the confidence level, which indicates the conditions that the number of state changes of the target object in the abnormal event needs to meet.

[0124] The number of state changes refers to the number of times the state of a target object changes. Taking the human heart as an example, each heartbeat can be considered as one state change; that is, the number of state changes refers to the number of times the heart beats.

[0125] It should be noted that the embodiments of this application are merely illustrative of the screening conditions and do not limit the screening conditions. In another embodiment, the screening conditions may also include other content.

[0126] 307. If the computer equipment does not meet the filtering conditions that match the confidence level for the first sub-data, it determines that the abnormal event has not occurred in the status data.

[0127] If the first sub-data does not meet the screening criteria that match the confidence level, the identification result can be considered incorrect, and it can be directly determined that no abnormal event has occurred in the status data.

[0128] 308. The computer equipment determines that no abnormal events have occurred in the status data if the confidence level does not reach the second confidence level threshold.

[0129] In some embodiments, the confidence level not reaching the second confidence threshold means that the level of multiple identification results is the fifth level.

[0130] 309. The computer device obtains the second sub-data of the abnormal event from the status data based on the third occurrence time period of the abnormal event.

[0131] In this embodiment of the application, the event identification result can be obtained through the above steps 301 to 308. The event identification result is used to indicate whether an abnormal event has occurred in the status data and the time period of the abnormal event.

[0132] However, considering that noise may be introduced during the state data acquisition process, in order to avoid the state abnormalities caused by noise being mistaken for abnormal events, this application embodiment can further verify the event identification results after determining the event identification results.

[0133] Among them, the second sub-data of the abnormal event is the data belonging to the third time period in the status data.

[0134] 310. The computer equipment performs noise identification on the second sub-data to determine the noisy data in the second sub-data.

[0135] In this embodiment, the computer device can employ any noise identification algorithm to perform noise identification on the second sub-data, such as the generalized least squares algorithm, multi-level least squares algorithm, etc. This embodiment does not limit the noise identification algorithm.

[0136] In some embodiments, the state data is represented as a signal wave. Each time the state of the target object changes, the collected state data forms a peak. However, introduced noise may also form a peak. Therefore, the computer device performs noise identification on the second sub-data. Determining the noise data in the second sub-data can be done by identifying which peaks are generated by introduced noise, and identifying the data corresponding to that peak in the state data as noise data.

[0137] 311. The computer equipment removes noisy data from the second sub-data to obtain the third sub-data.

[0138] Noisy data is at least one data segment in the second sub-data. After removing the noisy data from the second sub-data, the resulting third sub-data includes at least one data segment.

[0139] 312. The computer equipment performs statistical processing on the third sub-data to obtain the number of consecutive state changes. This number of consecutive state changes represents the number of times the target state in the third sub-data changes continuously.

[0140] It should be noted that state data is used to represent the state of the target object. Therefore, based on state data, it is possible to determine whether the state of the target object has changed and how many times the state of the target object has changed. Since the third sub-data is the data belonging to the third time period in the state data, it is also possible to determine whether the state of the target object has changed and how many times the state of the target object has changed based on the third sub-data.

[0141] In some embodiments, the state data is represented as a signal wave. Each time the state of the target object changes, the collected state data will form a peak. Therefore, the computer device performs statistical processing on the third sub-data to obtain the number of consecutive state changes, including: the computer device counts the number of peaks in each data segment of the third sub-data and determines the number of peaks as the number of consecutive state changes.

[0142] The computer device can determine the maximum number of peaks as the number of consecutive state changes, or it can determine the number of each peak as the number of consecutive state changes. This application embodiment does not limit this.

[0143] 313. When the number of consecutive state changes of a computer device is less than the first state change threshold, the event identification result is updated. The updated event identification result indicates that no abnormal events have occurred in the state data.

[0144] The first state change threshold can be any value, such as 0.5, 0.6, etc. This first state change threshold can be an empirical value or a value set by a technician; the embodiments of this application do not limit the first state change threshold.

[0145] If the number of consecutive state changes is less than the first state change threshold, it indicates that the state data that is not disturbed by noise is not continuous enough. The accuracy of the event identification result determined based on this state data is low. Therefore, the event identification result is updated.

[0146] The data processing method provided in this application divides an abnormal event into multiple development stages. These stages are then identified within the state data of the target object. By assessing the completeness of the identification of each development stage, the accuracy of multiple identification results can be determined. Based on these multiple identification results and their accuracy, the final event identification result is determined. In this way, the identification accuracy of each development stage can compensate for each other, weakening the influence of noise and improving the overall accuracy of abnormal event identification.

[0147] Furthermore, this application embodiment sets different levels of strictness in the screening conditions based on the confidence level of the identification results. When the identification results are not accurate enough, the screening conditions assist in the judgment, thereby improving the accuracy of the event identification results.

[0148] Furthermore, after determining the event identification result, this embodiment of the application will also perform noise detection on the second sub-data corresponding to the abnormal event. If multiple state data in the second sub-data are affected by noise, the obtained event identification result will be inaccurate. Therefore, the event identification result is updated to indicate that the abnormal event has not occurred, thereby improving the accuracy of the event identification result.

[0149] It should be noted that the data processing method provided in this application embodiment can be applied to atrial fibrillation identification scenarios. This application embodiment uses... Figure 4The illustrated embodiments provide an exemplary description of a data processing method applied to atrial fibrillation (AF) identification scenarios. It should be noted that in the AF identification scenario, the target object is the user, the status data is the user's electrocardiogram (ECG) signal, the abnormal event is the AF event, and the number of status changes is the heart rate. Furthermore, this application embodiment uses multiple progression stages of an AF event as an example, representing the start stage, the main stage, and the end stage, for illustrative purposes.

[0150] Figure 4 This is a flowchart of an atrial fibrillation identification method provided in an embodiment of this application. The method is illustrated using a computer device as the execution subject. See [link to relevant documentation]. Figure 4 This embodiment includes:

[0151] 401. Computer equipment acquires the user's electrocardiogram (ECG) signal, which is used to represent the user's physiological state.

[0152] In some embodiments, after a computer device acquires a user's electrocardiogram (ECG) signal, it can directly perform the following steps on the ECG signal. Alternatively, after acquiring the user's ECG signal, it can divide the ECG signal into multiple segments and perform the following steps for each ECG signal segment. This application embodiment does not limit this.

[0153] In some embodiments, after acquiring the user's electrocardiogram (ECG) signal, the computer device may also filter and normalize the ECG signal, and perform the following steps on the processed ECG signal.

[0154] For example, a computer device acquires a user's long-term electrocardiogram (ECG) signal and divides it into multiple segments, each segment lasting 30 seconds. Furthermore, the resulting ECG signals exhibit a certain degree of overlap during this division. For instance, the first 3 seconds of one segment may be identical to the last 3 seconds of the previous segment. In other words, seconds 1 to 30 are divided into one segment, seconds 27 to 57 into another, and seconds 54 to 84 into yet another.

[0155] It should be noted that the overlap of ECG signals can also be determined by the number of heartbeats. For example, an ECG signal may overlap with the previous ECG signal by 6 heartbeats. One heartbeat can be considered as one heartbeat, and each heartbeat will produce a peak in the ECG signal. Therefore, the number of heartbeats can be used to determine the number of heartbeats.

[0156] 402. The computer device uses an event recognition model to perform multi-target recognition on the electrocardiogram signal based on the characteristics of atrial fibrillation events in multiple stages of development, and obtains the recognition results of multiple stages of development. The recognition results of the stages of development are used to indicate whether the characteristics of the stage of development appear in the electrocardiogram signal and the time period of the stage of development.

[0157] Step 402 is the same as step 302 above, and will not be described in detail here.

[0158] It should be noted that the sample data used to train the event recognition model is manually labeled or obtained by other methods. This application embodiment does not limit this, but only provides an illustrative example of the sample data.

[0159] The annotation method for sample data may include: dividing an electrocardiogram (ECG) signal into background signal, initial phase, main phase, and final phase, using a single heartbeat as the unit, such as... Figure 5 As shown. The background signal may include sinus ECG signal, non-atrial fibrillation arrhythmia signal, and noise signal. The initial phase is centered on the first heartbeat of the atrial fibrillation event, taking one heartbeat before and after it. The final phase is centered on the last heartbeat of the atrial fibrillation event, taking one heartbeat before and after it. The main phase is from the first heartbeat of the atrial fibrillation event to the last heartbeat of the atrial fibrillation event.

[0160] Another point to note is that in step 401, the computer device can divide the electrocardiogram (ECG) signal into multiple segments. Therefore, in step 402 above, the recognition results corresponding to multiple ECG signals can be obtained. Since multiple ECG signals have overlapping parts, the recognition results of multiple ECG signals can be merged according to time. During the merging, the start and end stages can cover the main stage.

[0161] 403. When multiple identification results indicate that the initial stage, the main stage, and the final stage have been identified, the computer equipment determines the identification level of the multiple identification results as the first level. When the identification level is the first level, the third occurrence time of the atrial fibrillation event is determined based on the first occurrence time period of the initial stage and the second occurrence time period of the final stage.

[0162] When multiple identification results are classified as Level 1, it can be assumed that an atrial fibrillation event has occurred in the electrocardiogram signal. The occurrence time of the atrial fibrillation event can be determined directly based on the occurrence time of multiple progression stages.

[0163] 404. When multiple identification results indicate that the initial stage and the main stage or the main stage and the end stage have been identified, the computer equipment determines the identification level of the multiple identification results to be the second level. When the identification level is the second level, the computer equipment determines the first sub-ECG signal corresponding to the candidate atrial fibrillation event in the ECG signal, determines the RR interval dispersion and duration of the first sub-ECG signal, and determines the occurrence time of the candidate atrial fibrillation event as the occurrence time of the atrial fibrillation event when the RR interval dispersion is greater than the first dispersion and the duration is greater than the first duration.

[0164] In some embodiments, the first dispersion is 0.5 and the first duration is 5 seconds. If the RR interval dispersion is greater than the first dispersion and the duration is greater than the first duration, an atrial fibrillation event can be considered to have occurred in the electrocardiogram signal.

[0165] 405. When multiple recognition results indicate that the main body stage has been identified, the computer equipment determines the recognition level of the multiple recognition results as the third level. When the recognition level is the third level, it determines the first sub-ECG signal corresponding to the candidate atrial fibrillation event in the ECG signal, determines the RR interval dispersion and duration of the first sub-ECG signal, and determines the occurrence time of the candidate atrial fibrillation event as the occurrence time of the atrial fibrillation event when the RR interval dispersion is greater than the second dispersion and the duration is greater than the second duration.

[0166] In some embodiments, the second dispersion is 0.6 and the second duration is 5 seconds. When the RR interval dispersion is greater than the second dispersion and the duration is greater than the second duration, an atrial fibrillation event can be considered to have occurred in the electrocardiogram signal.

[0167] 406. When multiple recognition results indicate that the recognition has reached the start and end stages, the computer device determines the recognition level of the multiple recognition results to be the fourth level. When the recognition level is the fourth level, it determines the first sub-ECG signal corresponding to the candidate atrial fibrillation event in the ECG signal, determines the RR interval dispersion and duration of the first sub-ECG signal, and determines the occurrence time of the candidate atrial fibrillation event as the occurrence time of the atrial fibrillation event when the RR interval dispersion is greater than the third dispersion and the duration is greater than the third duration and less than the fourth duration.

[0168] In some embodiments, the third dispersion is 0.6, the third duration is 5 seconds, and the fourth duration is 10 seconds. If the RR interval dispersion is greater than the third dispersion and the duration is greater than the third duration but less than the fourth duration, an atrial fibrillation event can be considered to have occurred in the electrocardiogram signal.

[0169] 407. The computer device obtains the second sub-ECG signal corresponding to the atrial fibrillation event from the ECG signal, performs noise identification on the second sub-ECG signal, and determines the noise signal in the second sub-ECG signal.

[0170] 408. The computer equipment removes noise signals from the second sub-ECG signal to obtain the third sub-ECG signal.

[0171] 409. The computer equipment performs statistical processing on the third sub-electrocardiogram signal to obtain the number of consecutive state changes. This number of consecutive state changes represents the number of times the state of the target object in the third sub-electrocardiogram signal changes continuously.

[0172] 410. If the number of consecutive state changes of a computer device is less than the first state change threshold, it is determined that no atrial fibrillation event has occurred in the second sub-electrocardiogram signal.

[0173] In some embodiments, the first state change threshold is any value such as 6, 8, or 10.

[0174] The atrial fibrillation detection method provided in this application divides the atrial fibrillation event into multiple progression stages, identifies these stages in the electrocardiogram signal, and determines the accuracy of multiple identification results based on the completeness of the identification of these multiple progression stages. Based on these multiple identification results and their accuracy, the event identification result is determined. In this way, the identification accuracy of each progression stage can compensate for each other, weakening the influence of noise and improving the overall accuracy of atrial fibrillation event identification.

[0175] It should be noted that the atrial fibrillation detection method provided in this application can be used to identify both paroxysmal atrial fibrillation events and persistent atrial fibrillation events. Paroxysmal atrial fibrillation events are characterized by short duration and susceptibility to noise interference in the electrocardiogram (ECG) signal. The atrial fibrillation identification method provided in this application not only identifies multiple progression stages of the paroxysmal atrial fibrillation event, allowing the accuracy of each stage to compensate for each other, thereby improving the overall accuracy of paroxysmal atrial fibrillation event identification and reducing noise interference, but also allows for the selection of different stringency screening conditions based on combinations of identified progression stages for further identification, further improving the accuracy of paroxysmal atrial fibrillation event identification and reducing noise interference. Furthermore, this application embodiment can also determine the degree of noise interference in the ECG signal through noise detection, using this as a basis for identifying paroxysmal atrial fibrillation events, further reducing noise interference.

[0176] Furthermore, in this embodiment, the time period of the atrial fibrillation event is determined based on the first and last identified progression stages. Therefore, even if noise exists between the first and last progression stages, affecting the identification result, it will not affect the duration of the atrial fibrillation event. For example, if the atrial fibrillation event occurs between the 7th and 15th seconds of the electrocardiogram (ECG) signal, the ECG signal at the 9th second is not identified as a progression stage of the atrial fibrillation event because it is interfered with by noise. However, since the start and end stages at the 7th and 15th seconds are identified, the duration of the atrial fibrillation event can still be accurately obtained, improving the accuracy of identifying the duration of paroxysmal and persistent atrial fibrillation events.

[0177] Figure 6 This is a schematic diagram of a data processing device structure provided in an embodiment of this application. See also... Figure 6 The device includes:

[0178] The first acquisition module 601 is used to acquire the state data of the target object, which is used to represent the state of the target object.

[0179] The first identification module 602 is used to perform multi-target identification on the state data based on the characteristics of the abnormal events of the target object in multiple progress stages, and obtain the identification results of the multiple progress stages. The identification result of each progress stage is used to indicate whether the characteristics of the progress stage appear in the state data and the time period of the appearance of the characteristics of the progress stage.

[0180] The first determining module 603 is used to determine the confidence level of the multiple identification results based on the obtained multiple identification results, and the confidence level is positively correlated with the completeness of the identification of the abnormal event;

[0181] The second determining module 604 is used to determine the event identification result based on the multiple identification results and the confidence level. The event identification result is used to indicate whether the abnormal event has occurred in the status data and the time period of the abnormal event.

[0182] like Figure 7 As shown, in one possible implementation, the second determining module 604 is used to determine the third occurrence time of the abnormal event based on the first occurrence time of the first identified progress stage and the second occurrence time of the last identified progress stage when the confidence level reaches the first confidence level threshold.

[0183] The second determining module 604 is used to determine the first sub-data corresponding to the candidate abnormal event in the status data based on the first occurrence time period and the second occurrence time period when the confidence level reaches the second confidence level threshold but does not reach the first confidence level threshold; and to determine the third occurrence time period of the abnormal event based on the first occurrence time period and the second occurrence time period when the first sub-data satisfies the filtering conditions that match the confidence level. The confidence level is proportional to the strictness of the filtering conditions.

[0184] If the confidence level does not reach the second confidence threshold, it is determined that the abnormal event has not occurred in the state data;

[0185] The first confidence threshold is higher than the second confidence threshold.

[0186] In one possible implementation, the filtering criteria matching this confidence level include at least one of the following:

[0187] The duration range that matches the confidence level, which indicates the conditions that the duration of the anomalous event must meet;

[0188] The range of dispersion that matches the confidence level, which indicates the conditions that the dispersion of the RR intervals in the anomaly event needs to meet;

[0189] The range of state change counts that matches the confidence level, which indicates the conditions that the target object needs to meet in the anomaly event.

[0190] In one possible implementation, the plurality of progress stages include a start stage, a main stage, and an end stage. The plurality of identification results indicate that at least one of the features of the start stage and the features of the end stage, as well as the features of the main stage, appear in the state data. The filtering conditions include a first duration range, which is a duration range greater than a first duration threshold.

[0191] The second determining module 604 is used to determine the fourth occurrence time of the candidate abnormal event based on the first occurrence time period and the second occurrence time period; and when the fourth occurrence time period indicates that the duration of the candidate abnormal event is within the first duration range, the fourth occurrence time period is determined as the third occurrence time period of the abnormal event.

[0192] In one possible implementation, the multiple progress stages include a start stage, a main stage, and an end stage. The multiple identification results indicate that the features of the start stage and the end stage appear in the state data, but the features of the main stage do not appear. The filtering condition includes a second duration range, which is a duration range that is greater than a first duration threshold and less than a second duration threshold.

[0193] The second determining module 604 is used to determine the fourth occurrence time of the candidate abnormal event based on the first occurrence time period and the second occurrence time period; and when the fourth occurrence time period indicates that the duration of the candidate abnormal event is within the range of the second duration, the fourth occurrence time period is determined as the third occurrence time period of the abnormal event.

[0194] In one possible implementation, the first identification module 602 is used to perform multi-target identification on the state data based on the characteristics of the multiple progress stages through an event identification model, and obtain the identification results of the multiple progress stages. The event identification model is used to identify the multiple progress stages of the abnormal event.

[0195] In one possible implementation, the device further includes:

[0196] The first acquisition module 601 is used to acquire sample data, which includes sample status data and sample identification results of the multiple progress stages. The sample identification result of any progress stage is used to indicate whether the feature of the progress stage appears in the sample status data and the time period of the appearance of the feature of the progress stage.

[0197] The first identification module 602 is used to process the sample state data through the event identification model before training to obtain the prediction identification results of the multiple progress stages. The prediction identification result of any progress stage is used to indicate whether the event identification model before training has identified the progress stage and the time period of the occurrence of the progress stage.

[0198] Training module 605 is used to train the event recognition model based on the difference between the sample recognition results and the predicted recognition results corresponding to the same progress stage, so as to obtain the trained event recognition model.

[0199] In one possible implementation, the device further includes:

[0200] The second acquisition module 606 is used to acquire the second sub-data of the abnormal event from the status data based on the time period in which the abnormal event occurs;

[0201] The second identification module 607 is used to identify noise in the second sub-data and determine the noise data in the second sub-data.

[0202] The elimination module 608 is used to eliminate the noisy data from the second sub-data to obtain the third sub-data.

[0203] The statistics module 609 is used to perform statistical processing on the third sub-data to obtain the number of consecutive state changes. The number of consecutive state changes represents the number of times the state of the target object in the third sub-data changes consecutively.

[0204] The update module 610 is used to update the event identification result when the number of consecutive changes in the state is less than the first state change threshold. The updated event identification result indicates that the abnormal event has not appeared in the state data.

[0205] It should be noted that the data processing apparatus provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the data processing apparatus and the data processing method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0206] In an exemplary embodiment, a computer device is provided, the computer device including one or more processors and one or more memories, the one or more memories storing at least one piece of program code, the at least one piece of program code being loaded and executed by the one or more processors to implement the data processing method as described in the above embodiments.

[0207] Optionally, the computer device is provided as a terminal. Figure 8 This illustration shows a structural block diagram of a terminal 800 provided in an exemplary embodiment of this application. The terminal 800 may be a wearable device, a smartphone, a tablet computer, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 player (Moving Picture Experts Group Audio Layer IV), a laptop computer, or a desktop computer. The terminal 800 may also be referred to as a user device, a portable terminal, a laptop terminal, a desktop terminal, or other names.

[0208] Terminal 800 includes a processor 801 and a memory 802.

[0209] Processor 801 may include one or more processing cores, such as a quad-core processor or an octa-core processor. Processor 801 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 801 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 801 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 801 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0210] The memory 802 may include one or more computer-readable storage media, which may be non-transitory. The memory 802 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 802 are used to store at least one program code, which is executed by the processor 801 to implement the data processing method provided in the method embodiments of this application.

[0211] In some embodiments, the terminal 800 may also optionally include a peripheral device interface 803 and at least one peripheral device. The processor 801, memory 802, and peripheral device interface 803 can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface 803 via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of the following: a radio frequency circuit 804, a display screen 805, a camera 806, an audio circuit 807, a positioning component 808, and a power supply 809.

[0212] Peripheral device interface 803 can be used to connect at least one I / O (Input / Output) related peripheral device to processor 801 and memory 802. In some embodiments, processor 801, memory 802 and peripheral device interface 803 are integrated on the same chip or circuit board; in some other embodiments, any one or two of processor 801, memory 802 and peripheral device interface 803 can be implemented on separate chips or circuit boards, which is not limited in this embodiment.

[0213] The radio frequency (RF) circuit 804 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The RF circuit 804 communicates with communication networks and other communication devices via electromagnetic signals. The RF circuit 804 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals back into electrical signals. Optionally, the RF circuit 804 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The RF circuit 804 can communicate with other terminals through at least one wireless communication protocol. This wireless communication protocol includes, but is not limited to: metropolitan area networks (MANs), various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks (WLANs), and / or WiFi (Wireless Fidelity) networks. In some embodiments, the RF circuit 804 may also include circuitry related to NFC (Near Field Communication), which is not limited in this application.

[0214] Display screen 805 is used to display a UI (User Interface). This UI may include graphics, text, icons, videos, and any combination thereof. When display screen 805 is a touch display screen, it also has the ability to collect touch signals on or above its surface. These touch signals can be input as control signals to processor 801 for processing. In this case, display screen 805 can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, there may be one display screen 805, which serves as the front panel of terminal 800; in other embodiments, there may be at least two display screens, respectively disposed on different surfaces of terminal 800 or in a folded design; in still other embodiments, display screen 805 may be a flexible display screen, disposed on a curved or folded surface of terminal 800. Furthermore, display screen 805 may be configured as a non-rectangular irregular shape, i.e., a non-rectangular screen. Display screen 805 may be made of materials such as LCD (Liquid Crystal Display) or OLED (Organic Light-Emitting Diode).

[0215] The camera assembly 806 is used to acquire images or videos. Optionally, the camera assembly 806 includes a front-facing camera and a rear-facing camera. The front-facing camera is disposed on the front panel of the terminal, and the rear-facing camera is disposed on the back of the terminal. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusion of the main camera and the depth-sensing camera, panoramic shooting by fusion of the main camera and the wide-angle camera, VR (Virtual Reality) shooting, or other fusion shooting functions. In some embodiments, the camera assembly 806 may also include a flash. The flash may be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cool light flash, which can be used for light compensation at different color temperatures.

[0216] The audio circuit 807 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, converting the sound waves into electrical signals that are input to the processor 801 for processing, or input to the radio frequency circuit 804 to achieve voice communication. For stereo sound acquisition or noise reduction purposes, multiple microphones may be used, each located at a different part of the terminal 800. The microphone may also be an array microphone or an omnidirectional microphone. The speaker is used to convert the electrical signals from the processor 801 or the radio frequency circuit 804 into sound waves. The speaker may be a conventional diaphragm speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can convert electrical signals not only into audible sound waves but also into inaudible sound waves for purposes such as distance measurement. In some embodiments, the audio circuit 807 may also include a headphone jack.

[0217] The positioning component 808 is used to determine the current geographic location of the terminal 800 in order to enable navigation or LBS (Location Based Service). The positioning component 808 can be a positioning component based on the US GPS (Global Positioning System), China's BeiDou system, Russia's Granas system, or the European Union's Galileo system.

[0218] Power supply 809 is used to supply power to the various components in terminal 800. Power supply 809 can be AC ​​power, DC power, a disposable battery, or a rechargeable battery. When power supply 809 includes a rechargeable battery, the rechargeable battery can support wired or wireless charging. The rechargeable battery can also be used to support fast charging technology.

[0219] In some embodiments, the terminal 800 further includes one or more sensors 810. The one or more sensors 810 include, but are not limited to: an accelerometer 811, a gyroscope 812, a pressure sensor 813, a fingerprint sensor 814, an optical sensor 815, and a proximity sensor 816.

[0220] Those skilled in the art will understand that Figure 8 The structure shown does not constitute a limitation on terminal 800 and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0221] Alternatively, the computer device may be provided as a server. Figure 9This is a schematic diagram of a server structure provided in an embodiment of this application. The server 900 can vary considerably due to different configurations or performance. It may include one or more Central Processing Units (CPUs) 901 and one or more memories 902. The memory 902 stores at least one line of program code, which is loaded and executed by the processor 901 to implement the methods provided in the above-described method embodiments. Of course, the server may also have wired or wireless network interfaces, a keyboard, and input / output interfaces for input and output. The server may also include other components for implementing device functions, which will not be elaborated here.

[0222] The server 900 is used to execute the steps performed by the server in the above method embodiments.

[0223] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including program code that can be executed by a processor in a computer device to perform the data processing method described above. For example, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0224] In an exemplary embodiment, a computer program or computer program product is also provided, which includes computer program code that, when executed by a computer, causes the computer to implement the data processing method described above.

[0225] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0226] The above description is merely an optional embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A data processing method, characterized in that, The method includes: Obtain the state data of the target object, wherein the state data is used to represent the state of the target object; Based on the characteristics of the abnormal events of the target object in multiple progress stages, multi-target identification is performed on the state data to obtain the identification results of the multiple progress stages. The identification results of the progress stages are used to indicate whether the characteristics of the progress stage appear in the state data and the time period of the appearance of the characteristics of the progress stage. The multiple progress stages include the start stage, the main stage and the end stage. Based on the obtained multiple identification results, the confidence level of the multiple identification results is determined. The confidence level is positively correlated with the identification completeness of the abnormal event. The multiple identification results indicate that at least one of the features of the initial stage and the features of the final stage, as well as the features of the main stage, appear in the state data. If the confidence level reaches the first confidence threshold, the third occurrence time of the abnormal event is determined based on the first occurrence time period of the first identified progress stage and the second occurrence time period of the last identified progress stage. If the confidence level reaches a second confidence threshold but does not reach the first confidence threshold, based on the first occurrence time period and the second occurrence time period, a first sub-data corresponding to the candidate abnormal event in the status data is determined. If the first sub-data satisfies the filtering conditions that match the confidence level, based on the first occurrence time period and the second occurrence time period, a fourth occurrence time period of the candidate abnormal event is determined. If the fourth occurrence time period indicates that the duration of the candidate abnormal event is within a first duration range, the fourth occurrence time period is determined as the third occurrence time period of the abnormal event. The confidence level is proportional to the strictness of the filtering conditions, and the filtering conditions include the first duration range, which is a duration range greater than the first duration threshold. If the confidence level does not reach the second confidence threshold, it is determined that the abnormal event has not occurred in the status data; wherein the first confidence threshold is higher than the second confidence threshold.

2. The method according to claim 1, characterized in that, The filtering criteria that match the confidence level include at least one of the following: The duration range that matches the confidence level, wherein the duration range is used to indicate the conditions that the duration of the abnormal event needs to meet; A range of dispersion that matches the confidence level, the range of dispersion being used to indicate the conditions that the dispersion of the RR intervals in the anomalous event needs to meet; The range of state change counts that matches the confidence level, wherein the range of state change counts is used to indicate the conditions that the number of state changes of the target object in the abnormal event needs to meet.

3. The method according to claim 1, characterized in that, The method of performing multi-target identification on the state data based on the characteristics of the abnormal events of the target object at multiple progress stages, to obtain the identification results of the multiple progress stages, includes: By using an event recognition model, based on the characteristics of the multiple progress stages, multi-target recognition is performed on the state data to obtain the recognition results of the multiple progress stages. The event recognition model is used to identify the multiple progress stages of the abnormal event.

4. The method according to any one of claims 1 to 3, characterized in that, The method further includes: Based on the time period in which the abnormal event occurred, the second sub-data of the abnormal event is obtained from the status data; Noise identification is performed on the second sub-data to determine the noisy data in the second sub-data; The noisy data is removed from the second sub-data to obtain the third sub-data; The third sub-data is statistically processed to obtain the number of consecutive state changes, where the number of consecutive state changes represents the number of times the state of the target object in the third sub-data changes consecutively. If the number of consecutive state changes is less than the first state change threshold, the event identification result is updated. The updated event identification result indicates that the abnormal event has not occurred in the state data.

5. A data processing method, characterized in that, The method includes: Obtain the state data of the target object, wherein the state data is used to represent the state of the target object; Based on the characteristics of the abnormal events of the target object in multiple progress stages, multi-target identification is performed on the state data to obtain the identification results of the multiple progress stages. The identification results of the progress stages are used to indicate whether the characteristics of the progress stage appear in the state data and the time period of the appearance of the characteristics of the progress stage. The multiple progress stages include the start stage, the main stage and the end stage. Based on the obtained multiple identification results, the confidence level of the multiple identification results is determined. The confidence level is positively correlated with the completeness of the identification of the abnormal event. The multiple identification results indicate that the features of the initial stage and the features of the final stage appear in the state data, but the features of the main stage do not appear. If the confidence level reaches the first confidence threshold, the third occurrence time of the abnormal event is determined based on the first occurrence time of the first identified progress stage and the second occurrence time of the last identified progress stage. If the confidence level reaches the second confidence threshold but does not reach the first confidence threshold, based on the first occurrence time period and the second occurrence time period, the first sub-data corresponding to the candidate abnormal event in the status data is determined. If the first sub-data satisfies the filtering conditions that match the confidence level, based on the first occurrence time period and the second occurrence time period, the fourth occurrence time period of the candidate abnormal event is determined. If the fourth occurrence time period indicates that the duration of the candidate abnormal event is within the second duration range, the fourth occurrence time period is determined as the third occurrence time period of the abnormal event. The confidence level is proportional to the strictness of the filtering conditions, and the filtering conditions include the second duration range, which is a duration range greater than the first duration threshold and less than the second duration threshold. If the confidence level does not reach the second confidence threshold, it is determined that the abnormal event has not occurred in the status data; wherein the first confidence threshold is higher than the second confidence threshold.

6. The method according to claim 5, characterized in that, The filtering criteria that match the confidence level include at least one of the following: The duration range that matches the confidence level, wherein the duration range is used to indicate the conditions that the duration of the abnormal event needs to meet; A range of dispersion that matches the confidence level, the range of dispersion being used to indicate the conditions that the dispersion of the RR intervals in the anomalous event needs to meet; The range of state change counts that matches the confidence level, wherein the range of state change counts is used to indicate the conditions that the number of state changes of the target object in the abnormal event needs to meet.

7. The method according to claim 5, characterized in that, The method of performing multi-target identification on the state data based on the characteristics of the abnormal events of the target object at multiple progress stages, to obtain the identification results of the multiple progress stages, includes: By using an event recognition model, based on the characteristics of the multiple progress stages, multi-target recognition is performed on the state data to obtain the recognition results of the multiple progress stages. The event recognition model is used to identify the multiple progress stages of the abnormal event.

8. The method according to any one of claims 5 to 7, characterized in that, The method further includes: Based on the time period in which the abnormal event occurred, the second sub-data of the abnormal event is obtained from the status data; Noise identification is performed on the second sub-data to determine the noisy data in the second sub-data; The noisy data is removed from the second sub-data to obtain the third sub-data; The third sub-data is statistically processed to obtain the number of consecutive state changes, where the number of consecutive state changes represents the number of times the state of the target object in the third sub-data changes consecutively. If the number of consecutive state changes is less than the first state change threshold, the event identification result is updated. The updated event identification result indicates that the abnormal event has not occurred in the state data.

9. A data processing apparatus, characterized in that, The device includes: The first acquisition module is used to acquire the state data of the target object, wherein the state data is used to represent the state of the target object; The first identification module is used to perform multi-target identification on the state data based on the characteristics of the abnormal events of the target object in multiple progress stages, and obtain the identification results of the multiple progress stages. The identification results of the progress stages are used to indicate whether the characteristics of the progress stage appear in the state data and the time period of the appearance of the characteristics of the progress stage. The multiple progress stages include the start stage, the main stage and the end stage. The first determining module is used to determine the confidence level of the multiple identification results based on the obtained multiple identification results. The confidence level is positively correlated with the identification completeness of the abnormal event. The multiple identification results indicate that at least one of the features of the initial stage and the features of the final stage, as well as the features of the main stage, appears in the state data. The second determining module is configured to: when the confidence level reaches a first confidence level threshold, determine a third occurrence time period of the abnormal event based on a first occurrence time period of the first identified progress stage and a second occurrence time period of the last identified progress stage; when the confidence level reaches a second confidence level threshold but does not reach the first confidence level threshold, determine a first sub-data corresponding to the candidate abnormal event in the status data based on the first occurrence time period and the second occurrence time period; when the first sub-data satisfies a filtering condition matching the confidence level, determine a fourth occurrence time period of the candidate abnormal event based on the first occurrence time period and the second occurrence time period; when the fourth occurrence time period indicates that the duration of the candidate abnormal event is within a first duration range, determine the fourth occurrence time period as the third occurrence time period of the abnormal event, wherein the confidence level is proportional to the strictness of the filtering condition, the filtering condition includes the first duration range, the first duration range being a duration range greater than the first duration threshold; and when the confidence level does not reach the second confidence level threshold, determine that the abnormal event does not appear in the status data; wherein the first confidence level threshold is higher than the second confidence level threshold.

10. The apparatus according to claim 9, characterized in that, The filtering criteria that match the confidence level include at least one of the following: The duration range that matches the confidence level, wherein the duration range is used to indicate the conditions that the duration of the abnormal event needs to meet; A range of dispersion that matches the confidence level, the range of dispersion being used to indicate the conditions that the dispersion of the RR intervals in the anomalous event needs to meet; The range of state change counts that matches the confidence level, wherein the range of state change counts is used to indicate the conditions that the number of state changes of the target object in the abnormal event needs to meet.

11. The apparatus according to claim 9, characterized in that, The identification module is used to perform multi-target identification on the state data based on the characteristics of the multiple progress stages using an event identification model, and to obtain the identification results of the multiple progress stages. The event identification model is used to identify the multiple progress stages of the abnormal event.

12. The apparatus according to any one of claims 9-11, characterized in that, The device further includes: The second acquisition module is used to acquire the second sub-data of the abnormal event from the status data based on the time period in which the abnormal event occurs; The second identification module is used to identify noise in the second sub-data and determine the noise data in the second sub-data. The elimination module is used to eliminate the noise data from the second sub-data to obtain the third sub-data; The statistics module is used to perform statistical processing on the third sub-data to obtain the number of consecutive state changes, where the number of consecutive state changes represents the number of times the state of the target object in the third sub-data changes consecutively. The update module is used to update the event identification result when the number of consecutive state changes is less than a first state change threshold. The updated event identification result indicates that the abnormal event has not appeared in the state data.

13. A data processing apparatus, characterized in that, The device includes: The first acquisition module is used to acquire the state data of the target object, wherein the state data is used to represent the state of the target object; The first identification module is used to perform multi-target identification on the state data based on the characteristics of the abnormal events of the target object in multiple progress stages, and obtain the identification results of the multiple progress stages. The identification results of the progress stages are used to indicate whether the characteristics of the progress stage appear in the state data and the time period of the appearance of the characteristics of the progress stage. The multiple progress stages include the start stage, the main stage and the end stage. The first determining module is used to determine the confidence level of the multiple identification results based on the obtained multiple identification results. The confidence level is positively correlated with the identification completeness of the abnormal event. The multiple identification results indicate that the features of the initial stage and the features of the final stage appear in the state data, but the features of the main stage do not appear. The second determining module is configured to: when the confidence level reaches a first confidence level threshold, determine a third occurrence time period of the abnormal event based on a first occurrence time period of the first identified progress stage and a second occurrence time period of the last identified progress stage; when the confidence level reaches a second confidence level threshold but does not reach the first confidence level threshold, determine a first sub-data corresponding to the candidate abnormal event in the status data based on the first occurrence time period and the second occurrence time period; when the first sub-data satisfies the filtering conditions matching the confidence level, determine a fourth occurrence time period of the candidate abnormal event based on the first occurrence time period and the second occurrence time period; when the fourth occurrence time period indicates that the duration of the candidate abnormal event is within a second duration range, determine the fourth occurrence time period as the third occurrence time period of the abnormal event, wherein the confidence level is proportional to the strictness of the filtering conditions, the filtering conditions include the second duration range, the second duration range being a duration range greater than the first duration threshold and less than the second duration threshold; when the confidence level does not reach the second confidence level threshold, determine that the abnormal event does not appear in the status data; wherein the first confidence level threshold is higher than the second confidence level threshold.

14. The apparatus according to claim 13, characterized in that, The filtering criteria that match the confidence level include at least one of the following: The duration range that matches the confidence level, wherein the duration range is used to indicate the conditions that the duration of the abnormal event needs to meet; A range of dispersion that matches the confidence level, the range of dispersion being used to indicate the conditions that the dispersion of the RR intervals in the anomalous event needs to meet; The range of state change counts that matches the confidence level, wherein the range of state change counts is used to indicate the conditions that the number of state changes of the target object in the abnormal event needs to meet.

15. The apparatus according to claim 13, characterized in that, The identification module is used to perform multi-target identification on the state data based on the characteristics of the multiple progress stages using an event identification model, and to obtain the identification results of the multiple progress stages. The event identification model is used to identify the multiple progress stages of the abnormal event.

16. The apparatus according to any one of claims 13-15, characterized in that, The device further includes: The second acquisition module is used to acquire the second sub-data of the abnormal event from the status data based on the time period in which the abnormal event occurs; The second identification module is used to identify noise in the second sub-data and determine the noise data in the second sub-data. The elimination module is used to eliminate the noise data from the second sub-data to obtain the third sub-data; The statistics module is used to perform statistical processing on the third sub-data to obtain the number of consecutive state changes, where the number of consecutive state changes represents the number of times the state of the target object in the third sub-data changes consecutively. The update module is used to update the event identification result when the number of consecutive state changes is less than a first state change threshold. The updated event identification result indicates that the abnormal event has not appeared in the state data.

17. A computer device, characterized in that, The computer device includes one or more processors and one or more memories, wherein at least one piece of program code is stored in the one or more memories, and the at least one piece of program code is loaded and executed by the one or more processors to perform the operations performed by the data processing method as described in any one of claims 1 to 8.

18. A computer-readable storage medium, characterized in that, The storage medium stores at least one piece of program code, which is loaded and executed by a processor to perform the operations of the data processing method as described in any one of claims 1 to 8.

19. A computer program product, the computer program product comprising: Computer program code, when executed by a computer, causes the computer to perform the operations performed by the data processing method as described in any one of claims 1 to 8.

Images