A method, apparatus, electronic device, and storage medium for determining the state of an object.

Abstract

This disclosure relates to a method, apparatus, electronic device, and storage medium for determining the state of an object. The method includes: acquiring detection data of a currently detected object; the detection data includes dimensional information, attribute information, historical data, and current data; determining an anomaly threshold corresponding to the currently detected object based on the dimensional information and attribute information; determining an evaluation value based on the historical data and current data; if the evaluation value is greater than or equal to the anomaly threshold, determining that the currently detected object has an anomaly within the time period corresponding to the current data. Thus, embodiments of this application can obtain an evaluation value based on historical data and current data, and subsequently determine whether the detected object has an anomaly based on the evaluation value and the anomaly threshold, which is simple, convenient, and saves manpower.

Description

Technical Field

[0001] This disclosure relates to the field of Internet technology, and in particular to a method, apparatus, electronic device and storage medium for determining the state of an object. Background Technology

[0002] With the rapid development of the mobile internet, information dissemination via mobile devices has become increasingly sophisticated. Typically, information can be embedded in various browsers or applications on the device for dissemination. These applications generally include social networking apps, video playback apps, games, and so on.

[0003] However, with the continuous development of information consulting services, the probability of anomalies occurring during information dissemination also increases. For example, when the system hosting information consulting malfunctions, it may affect the information consulting service. However, due to the complexity of the systems hosting information consulting, it is difficult to manually determine whether an anomaly has occurred in practice. Therefore, an object state determination method is needed to determine whether an anomaly has occurred in the information consulting service, and thus, to determine whether the system itself has malfunctioned. Summary of the Invention

[0004] This disclosure provides a method, apparatus, electronic device, and storage medium for determining the state of an object. The technical solution of this disclosure is as follows:

[0005] According to a first aspect of the present disclosure, a method for determining the state of an object is provided, comprising:

[0006] Obtain the detection data of the current object; the detection data includes dimensional information, attribute information, historical data, and current data;

[0007] Determine the anomaly threshold corresponding to the current detection object based on dimensional and attribute information;

[0008] The assessment value is determined based on historical and current data;

[0009] If the evaluation value is greater than or equal to the anomaly threshold, it is determined that the current detected object has an anomaly within the time period corresponding to the current data.

[0010] In some possible embodiments, determining the evaluation value based on historical and current data includes:

[0011] Determine forecast data based on historical data;

[0012] The assessment value is determined based on both forecast and current data.

[0013] The duration corresponding to historical data, the duration corresponding to current data, and the duration corresponding to predicted data are all the same detection duration.

[0014] In some possible embodiments, obtaining the detection data of the current detection object includes:

[0015] Obtain the dimensional information, historical data, and current data of the current detection object; the dimensional information includes the recommended display page of the current detection object, its industry, and traffic entry information;

[0016] The attribute information of the current detection object is determined based on the dimensional information; the attribute information includes the consumption data and exposure data of the current detection object.

[0017] In some possible embodiments, determining the anomaly threshold corresponding to the current detection object based on dimensional information and attribute information includes:

[0018] Determine the first display time information of the currently detected object on the recommended display page within the time period corresponding to the current data;

[0019] Determine the second display time information of the currently detected object on the recommended display page within the time period corresponding to the historical data;

[0020] Determine the temporal correlation information between the first display time information and the second display time information;

[0021] The anomaly threshold corresponding to the current detection object is determined based on dimensional information, attribute information, and time correlation information.

[0022] In some possible embodiments, determining the predicted data based on historical data includes:

[0023] Input historical data into the anomaly prediction model to obtain the trend component data of the historical data at a preset timestamp within the detection period;

[0024] The trend data of historical data within the detection period is determined based on the trend component data;

[0025] Trend data carrying timestamp information is identified as forecast data.

[0026] In some possible embodiments, historical data is input into the anomaly prediction model to obtain trend component data of the historical data at preset timestamps within the detection period, including:

[0027] Perform anomaly detection and processing on historical data, delete abnormal data, and obtain processed historical data;

[0028] The processed historical data is input into the anomaly prediction model to obtain the trend component data of the historical data at preset timestamps within the detection period.

[0029] In some possible embodiments, determining the evaluation value based on predicted data and current data includes:

[0030] The predicted data and current data are segmented and processed to obtain a sub-predicted data set and a sub-current data set; the sub-predicted data in the sub-predicted data set and the sub-current data in the sub-current data set are in a one-to-one correspondence based on timestamp information;

[0031] The sub-evaluation value is determined based on sub-predicted data and sub-current data with the same timestamp information;

[0032] The evaluation value is determined based on multiple sub-evaluation values.

[0033] In some possible embodiments, obtaining historical and current data of the currently detected object includes:

[0034] Historical data is obtained based on historical time periods;

[0035] Get the current data based on the current time period;

[0036] The historical time period is within the first time cycle; the current time period is within the second time cycle; the start time of the historical time period and the start time of the current time period are the same within the time cycle; the end time of the historical time period and the end time of the current time period are the same within the time cycle.

[0037] In some possible embodiments, before acquiring the detection data of the current detection object, the method further includes:

[0038] If the evaluation value corresponding to the historical detection object is greater than or equal to the abnormal threshold corresponding to the historical detection object, the abnormality of the historical detection object is determined based on the current data of the historical detection object.

[0039] If no anomalies are found in the historically detected objects, the updated anomaly threshold is obtained based on the dimensional information, attribute information, and anomaly threshold of the historically detected objects.

[0040] According to a second aspect of the present disclosure, an object state determination apparatus is provided, comprising:

[0041] The data acquisition module is configured to acquire the detection data of the current detection object; the detection data includes dimensional information, attribute information, historical data, and current data.

[0042] The anomaly threshold determination module is configured to determine the anomaly threshold corresponding to the current detection object based on dimension information and attribute information.

[0043] The evaluation determination module is configured to determine an evaluation value based on historical and current data.

[0044] The anomaly detection module is configured to determine if the evaluated value is greater than or equal to the anomaly threshold, indicating that the currently detected object has an anomaly within the time period corresponding to the current data.

[0045] In some possible embodiments, the evaluation and determination module is configured to perform:

[0046] Determine forecast data based on historical data;

[0047] The assessment value is determined based on both forecast and current data.

[0048] The duration corresponding to historical data, the duration corresponding to current data, and the duration corresponding to predicted data are all the same detection duration.

[0049] In some possible embodiments, the data acquisition module is configured to perform:

[0050] Obtain the dimensional information, historical data, and current data of the current detection object; the dimensional information includes the recommended display page of the current detection object, its industry, and traffic entry information;

[0051] The attribute information of the current detection object is determined based on the dimensional information; the attribute information includes the consumption data and exposure data of the current detection object.

[0052] In some possible embodiments, the anomaly threshold determination module is configured to perform:

[0053] Determine the first display time information of the currently detected object on the recommended display page within the time period corresponding to the current data;

[0054] Determine the second display time information of the currently detected object on the recommended display page within the time period corresponding to the historical data;

[0055] Determine the temporal correlation information between the first display time information and the second display time information;

[0056] The anomaly threshold corresponding to the current detection object is determined based on dimensional information, attribute information, and time correlation information.

[0057] In some possible embodiments, the evaluation and determination module is configured to perform:

[0058] Input historical data into the anomaly prediction model to obtain the trend component data of the historical data at a preset timestamp within the detection period;

[0059] The trend data of historical data within the detection period is determined based on the trend component data;

[0060] Trend data carrying timestamp information is identified as forecast data.

[0061] In some possible embodiments, the evaluation and determination module is configured to perform:

[0062] Perform anomaly detection and processing on historical data, delete abnormal data, and obtain processed historical data;

[0063] The processed historical data is input into the anomaly prediction model to obtain the trend component data of the historical data at preset timestamps within the detection period.

[0064] In some possible embodiments, the evaluation and determination module is configured to perform:

[0065] The predicted data and current data are segmented and processed to obtain a sub-predicted data set and a sub-current data set; the sub-predicted data in the sub-predicted data set and the sub-current data in the sub-current data set are in a one-to-one correspondence based on timestamp information;

[0066] The sub-evaluation value is determined based on sub-predicted data and sub-current data with the same timestamp information;

[0067] The evaluation value is determined based on multiple sub-evaluation values.

[0068] In some possible embodiments, the data acquisition module is configured to perform:

[0069] Historical data is obtained based on historical time periods;

[0070] Get the current data based on the current time period;

[0071] The historical time period is within the first time cycle; the current time period is within the second time cycle; the start time of the historical time period and the start time of the current time period are the same within the time cycle; the end time of the historical time period and the end time of the current time period are the same within the time cycle.

[0072] In some possible embodiments, the apparatus further includes:

[0073] The anomaly detection module is configured to determine the anomaly status of the historical detection object based on the current data of the historical detection object if the evaluation value corresponding to the historical detection object is greater than or equal to the anomaly threshold corresponding to the historical detection object.

[0074] The anomaly threshold update module is configured to perform an update of the anomaly threshold based on the dimension information, attribute information, and anomaly threshold of the historically detected objects if no anomalies are found.

[0075] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute instructions to implement the method as described in any of the first aspects above.

[0076] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided such that, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the method of any one of the first aspects of the present disclosure.

[0077] According to a fifth aspect of the present disclosure, a computer program product is provided, the computer program product including a computer program stored in a readable storage medium, wherein at least one processor of a computer device reads from the readable storage medium and executes the computer program, causing the computer device to perform the method of any one of the first aspects of the present disclosure.

[0078] The technical solutions provided by the embodiments of this disclosure have at least the following beneficial effects:

[0079] The detection data of the current detection object is acquired. This data includes dimensional information, attribute information, historical data, and current data. An anomaly threshold corresponding to the current detection object is determined based on the dimensional and attribute information. An evaluation value is then determined based on the historical and current data. If the evaluation value is greater than or equal to the anomaly threshold, it is determined that the current detection object has an anomaly within the time period corresponding to the current data. Thus, this embodiment of the application can obtain an evaluation value based on historical and current data, and subsequently determine whether the detection object has an anomaly based on the evaluation value and the anomaly threshold. This method is simple, convenient, and saves manpower.

[0080] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0081] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0082] Figure 1 This is a schematic diagram illustrating an application environment according to an exemplary embodiment;

[0083] Figure 2 This is a flowchart illustrating an object state determination method according to an exemplary embodiment;

[0084] Figure 3 This is a flowchart illustrating an object state determination method according to an exemplary embodiment;

[0085] Figure 4 This is a flowchart illustrating a method for determining an anomaly threshold according to an exemplary embodiment;

[0086] Figure 5This is a flowchart illustrating a method for determining prediction data according to an exemplary embodiment;

[0087] Figure 6 This is a flowchart illustrating a method for determining an evaluation value according to an exemplary embodiment;

[0088] Figure 7 This is a block diagram illustrating an object state determination apparatus according to an exemplary embodiment;

[0089] Figure 8 This is a block diagram illustrating an electronic device for determining the state of an object according to an exemplary embodiment. Detailed Implementation

[0090] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0091] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar first objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0092] Please see Figure 1 , Figure 1 This is a schematic diagram illustrating an application environment for an object state determination method according to an exemplary embodiment, such as... Figure 1 As shown, the application environment may include client 01 and server 02.

[0093] In an optional embodiment, the client 01 can receive an identifier of the detection information from the server 02, and then obtain the detection data of the current detection object. The detection data includes dimension information, attribute information, historical data, and current data. An anomaly threshold corresponding to the current detection object is determined based on the dimension information and attribute information. An evaluation value is determined based on the historical data and current data. If the evaluation value is greater than or equal to the anomaly threshold, it is determined that the current detection object has an anomaly within the time period corresponding to the current data. The client 01 can be, but is not limited to, electronic devices such as smartphones, desktop computers, tablets, laptops, smart speakers, digital assistants, augmented reality (AR) / virtual reality (VR) devices, and smart wearable devices. It can also be software running on the aforementioned electronic devices, such as applications or mini-programs. Optionally, the operating system running on the electronic device can be, but is not limited to, Android, iOS, Linux, Windows, and Unix systems.

[0094] In an optional embodiment, server 02 may be a system that hosts the detected object. Optionally, server 02 may be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Optionally, the operating system running on the server may include, but is not limited to, Android, iOS, Linux, Windows, Unix, etc.

[0095] Optionally, the client 01 and server 02 mentioned above can be connected via a wired link or a wireless link, depending on the actual application scenario.

[0096] In addition, it should be noted that, Figure 1 The illustration shows only one application environment for the object state determination method provided in this disclosure. In practical applications, other application environments may also be included. For example, in practical applications, there may only be a server hosting the object to be detected, and the determination of whether the object to be detected has experienced an anomaly may also be performed by this server.

[0097] Figure 2 This is a flowchart illustrating an object state determination method according to an exemplary embodiment, such as... Figure 2 As shown, the object state determination method can be applied to both servers and clients, and includes the following steps:

[0098] In step S201, the detection data of the current detection object is obtained; the detection data includes dimension information, attribute information, historical data and current data.

[0099] This application will describe the implementation using the client as the subject of execution. For other subjects of execution, such as independent physical servers or distributed servers, please refer to the client-side implementation method, which will not be repeated here.

[0100] In this embodiment of the application, the current detection object can be a piece of information (such as advertising information), a piece of music, or a video, etc.

[0101] In one alternative embodiment, the client can determine the detection data of the current object being detected, wherein the detection data may include dimensional information, attribute information, historical data, and current data.

[0102] Taking information consulting as an example, the client can send the identification information of the currently detected object to the server, so that the server can determine the dimension information, attribute information, historical data, and current data of the currently detected object based on the identification information. For example, assuming the currently detected object is a certain piece of information, the detected data would be the dimension information, attribute information, historical data, and current data of that information. After the server obtains the detected data corresponding to the identification information, it can send the dimension information, attribute information, historical data, and current data of the detected object to the client.

[0103] Correspondingly, the client can directly obtain the dimensional information, attribute information, historical data, and current data of the object being detected based on the website or application where the object is located, so that the client can determine whether there are any anomalies in the object being detected based on the currently obtained detection data.

[0104] In this embodiment, the dimension information may include the recommended display page of the currently detected object, its industry, and traffic entry information. Taking information A as an example, after receiving the detection data, the client can first determine the website or application where information A is located, and then determine the dimension information of information A on the website or application. Optionally, the client can determine the recommended display page of information A, including the following page, discovery page, or local page, etc. Optionally, the client can determine the industry to which information A belongs, such as the education industry, finance industry, medical industry, technology industry, or beauty industry, etc. Optionally, the client can also determine the traffic entry information of information A. In this embodiment, the traffic entry refers to the website or application that must be entered before entering the website or application where the currently detected object is located. The website that must be entered first, i.e., the traffic entry, can be a social networking website or social application, a short video website or short video application, or an information website or information application.

[0105] Specifically, the user can first open a first website or application. This first website or application may include a page redirection control to a second website or application where the information A is located. When the page redirection control of the second website or application is triggered, the user can be redirected to the second website or application. The currently detected object, i.e., information A, can be displayed on the recommended display page of the second website or application.

[0106] In this embodiment, the attribute information may include the consumption data and exposure data of the currently detected object. Optionally, the server may use the Flink distributed data stream engine to obtain the attribute information of the detected object in real time, or it may use the Kafka high-throughput, low-latency tool to obtain the attribute information of the detected object, or it may use the Druid efficient and manageable database connection pool to obtain the attribute information of the detected object, and then send the attribute information to the client.

[0107] In this embodiment, the consumption data can be the resources consumed by the current detection object, including manpower and money. The exposure data can refer to the statistical data of the number of times the current detection object can be seen by the user.

[0108] In another alternative embodiment for determining the dimensional and attribute information of the object to be detected, the client may first determine the dimensional information and then determine the attribute information. Figure 3 This is a flowchart illustrating an object state determination method according to an exemplary embodiment, such as... Figure 3 As shown, it includes the following steps:

[0109] In step S2011, the dimensional information, historical data, and current data of the current detection object are obtained; the dimensional information includes the recommended display page of the current detection object, its industry, and traffic entry information.

[0110] The client can send the identification information of the currently detected object to the server, allowing the server to determine the dimensional information, historical data, and current data of the object based on this identification information. Continuing with the example of information A, the detection data would then consist of the dimensional information, historical data, and current data of information A. Once the server obtains the detection data corresponding to this identification information, it can send the dimensional information, historical data, and current data of the detected object back to the client.

[0111] Correspondingly, the client can directly obtain the dimensional information, historical data, and current data of the object being detected based on the website or application where the object is located, so that the client can determine whether there are any anomalies in the object being detected based on the currently obtained detection data.

[0112] In this embodiment of the application, the dimensional information may include the recommended display page of the currently detected object, its industry, and traffic entry information.

[0113] In step S2012, the attribute information of the current detection object is determined based on the dimension information; the attribute information includes the consumption data and exposure data of the current detection object.

[0114] In this embodiment of the application, the client can determine the attribute information of the current detection object based on the dimension information. The attribute information may include the consumption data and exposure data of the current detection object.

[0115] In one optional embodiment, the description continues with the detection object as information A, consumption data as money, and exposure data as statistical data on the number of times users can see the information. Since different recommended display pages in an application have varying abilities to attract users to stay or be opened by users immediately, and the number of users following the industry to which the detection object belongs also differs, and different traffic entry points have varying abilities to attract users, the client can determine the consumption data and exposure data of the detection object based on any one of the recommended display pages, industry, and traffic entry information in the dimension information. Optionally, the client can determine the consumption data and exposure data of the detection object based on any two of the recommended display pages, industry, and traffic entry information in the dimension information. Optionally, to more comprehensively consider the impact of all dimension information on attribute information, the client can determine the consumption data and exposure data of the detection object based on the recommended display pages, industry, and traffic entry information in the dimension information.

[0116] For example, the stronger the recommended display page's ability to attract users to stay within the application, or the more likely it is to be opened by users first, the higher the ranking of the recommended display page. Similarly, the more users follow or work in the industry to which the currently detected object belongs, the higher the ranking of that industry. Likewise, the stronger the ability of a traffic entry point to attract users, the higher the ranking of that traffic entry point's information.

[0117] In one optional embodiment, the client can categorize recommended display pages into multiple levels according to preset rules, with each level corresponding to a consumption data A and an exposure data A. Similarly, the client can categorize industries into multiple levels according to preset rules, with each industry level corresponding to a consumption data B and an exposure data B. Likewise, the client can categorize traffic entry information into multiple levels according to preset rules, with each industry level of traffic entry information corresponding to a consumption data C and an exposure data C.

[0118] Optionally, the client can process the three consumption data points according to preset rules to obtain the final consumption data. Similarly, it can process the three exposure data points according to preset rules to obtain the final exposure data. For example, the client can multiply consumption data A, consumption data B, and consumption data C by their corresponding percentages, and then add the three results together to obtain the final consumption data. Likewise, the client can multiply exposure data A, exposure data B, and exposure data C by their corresponding percentages, and then add the three results together to obtain the final exposure data.

[0119] In this way, the client can obtain attribute information through dimensional information, enabling the client to take a holistic approach and consider all aspects when determining the core data of the object being detected. This ensures that the client obtains comprehensive and complete core data of the current object being detected, preparing for subsequent judgments on whether the current data of the object being detected is abnormal.

[0120] In this embodiment of the application, the duration corresponding to the historical data of the current detection object and the duration corresponding to the current data are the same, which is the same detection duration.

[0121] Specifically, the client can retrieve historical data based on a historical time period, and current data based on the current time period. That is, the time span of the historical data and the time span of the current data are the same, for example, one week. The historical time period can be within a first time cycle, and the current time period can be within a second time cycle. For example, the first time cycle refers to the entire week before last, and the second time cycle refers to the entire week before last. The start time of the historical time period and the start time of the current time period are the same within the time cycle, for example, 00:00:00 on Monday. The end time of the historical time period and the end time of the current time period are also the same within the time cycle, for example, 23:59:59 on Sunday.

[0122] For example, a client can retrieve historical data from the week before last for the currently monitored object, and also retrieve current data from the previous week for the same object. The historical data and current data are data from the same dimension. For instance, the historical data might be the overall reach of information A two weeks ago, and the current data might be the overall reach of information A last week. Alternatively, the historical data might be the overall revenue of information A two weeks ago, and the current data might be the overall revenue of information A last week. Or, the historical data might be the revenue of information A for each time period two weeks ago, and the current data might be the revenue of information A for each time period last week.

[0123] This ensures that historical data and current data can be compared on the same time dimension.

[0124] In step S202, the anomaly threshold corresponding to the current detection object is determined based on the dimension information and attribute information.

[0125] In this embodiment of the application, the anomaly threshold is an important metric used in subsequent implementation steps to measure whether the evaluation value between the current data and the predicted data based on historical data is reasonable, and thus to determine whether the current data of the detected object is abnormal.

[0126] In one optional embodiment, the anomaly threshold corresponding to the currently detected object can be determined based on empirical values. Different currently detected objects can correspond to different anomaly thresholds, or they can correspond to the same anomaly threshold. For example, the empirical value of the anomaly threshold can be obtained by labeling past anomalies using annotated data.

[0127] In another alternative embodiment, a relationship mapping table can exist between dimensional information, attribute information, and anomaly threshold. The client can determine the anomaly threshold corresponding to the current detection object based on the dimensional information and attribute information by looking up the table.

[0128] Figure 4This is a flowchart illustrating a method for determining an anomaly threshold according to an exemplary embodiment, such as... Figure 4 As shown, it includes the following steps:

[0129] In step S2021, the first display time information of the currently detected object on the recommended display page within the time period corresponding to the current data is determined.

[0130] In one optional embodiment, the client can determine the first display time information of the currently detected object on the recommendation display page within the time period corresponding to the current data, that is, within the current time period. For example, the display time information of the currently detected object on the recommendation display page last week is that it is displayed once every 5 seconds.

[0131] In step S2022, the second display time information of the currently detected object on the recommended display page is determined within the time period corresponding to the historical data.

[0132] In one optional embodiment, the client can determine the second display time information of the currently detected object on the recommended display page within the time period corresponding to the historical data, that is, within the historical time period. For example, the display time information of the currently detected object on the recommended display page last week is that it is displayed once every 10 seconds.

[0133] In step S2023, the time correlation information between the first display time information and the second display time information is determined.

[0134] Therefore, the client can determine the time correlation information between the first display time information and the second display time information. The time correlation information can be the ratio 2 between the second display time information and the first display time information.

[0135] In step S2024, the anomaly threshold corresponding to the current detection object is determined based on the dimension information, attribute information, and time correlation information.

[0136] Finally, the client can determine the anomaly threshold corresponding to the current detection object based on the dimension information, attribute information, and time correlation information.

[0137] The subsequent steps of this application require using historical data as a benchmark to obtain predicted data. Then, an evaluation value is determined based on the predicted data and the current data. This evaluation value is then compared with the anomaly threshold to determine whether the current data exhibits any anomalies within the corresponding time period. Successfully completing these steps presupposes that certain conditions (such as dimension and attribute information) are identical between the week before last corresponding to the historical data and the week before last corresponding to the current data. Similarly, other information should be kept as consistent as possible; for example, both should be displayed scrolling every 5 seconds on the recommendation page. In other words, the time-related information should ideally be 1. If external factors (such as operational activities or promotional interactions) cause the time-related information to deviate from 1, then the impact of time-related information on the anomaly threshold can be considered in addition to dimension and attribute information. For instance, it might ultimately lead to different overall revenue, and the anomaly threshold can be reasonably adjusted based on the time-related information.

[0138] In this embodiment, the detection duration is time information used in subsequent implementation steps to determine the length of time spanned by the acquired historical or current data. Different current detection objects can correspond to the same detection duration or different detection durations.

[0139] In some optional embodiments, in addition to determining the anomaly threshold corresponding to the current detection object based on dimensional and attribute information, the client can also determine the detection duration corresponding to the current detection object based on the dimensional and attribute information. The client can determine a first detection duration based on the dimensional information, a second detection duration based on the attribute information, and then determine the final detection duration based on the first and second detection durations. For example, a detection duration could be one week or one day. Adjusting the granularity of the detection duration can also determine the accuracy of subsequent judgments on whether the current data is abnormal.

[0140] Furthermore, the client can also determine the detection duration corresponding to the current detection object based on dimension information, attribute information, and time correlation information.

[0141] In step S203, the evaluation value is determined based on historical data and current data.

[0142] In some possible embodiments, the client can directly determine the evaluation value based on historical data and current data. Optionally, the client can determine the evaluation value based on the overall historical data and the overall current data; alternatively, the evaluation value can be the difference between historical data and current data.

[0143] In some possible embodiments, the client can determine the predicted data based on historical data and determine the evaluation value based on the predicted data and the current data, wherein the duration corresponding to the predicted data is the same as the detection duration of the historical data and the current data.

[0144] In an optional embodiment, the client can determine the predicted data based on an anomaly prediction model. Optionally, the anomaly prediction model can be constructed based on a time series decomposition algorithm (Seasonal and Trend decomposition using Loess, STL). Alternatively, the anomaly prediction model can be constructed based on a neural convolutional network.

[0145] Figure 5 This is a flowchart illustrating a method for determining prediction data according to an exemplary embodiment, such as... Figure 5 As shown, it includes the following steps:

[0146] In step S501, historical data is input into the anomaly prediction model to obtain the trend component data of the historical data at preset timestamps within the detection period.

[0147] In this embodiment of the application, the anomaly prediction model determined by the time series decomposition algorithm mainly includes two component data: one is time component data (which satisfies the same detection duration in terms of the time information of historical data and current data), and the other is trend component data. When the client inputs historical data into the anomaly prediction model, the anomaly prediction model can analyze the trend component data of the historical data at a preset timestamp within the detection duration. In this way, the client can obtain the trend component data of the historical data at a preset timestamp within the detection duration.

[0148] In an optional embodiment, the historical data described above may contain obvious errors. For example, if the return corresponding to each timestamp of the previous two weeks in the historical data is negative, it can be identified as an obvious error. Based on this, the client can perform anomaly detection processing on the historical data, deleting abnormal data to obtain processed historical data. The processed historical data is then input into an anomaly prediction model to obtain the trend component data of the historical data at preset timestamps within the detection period.

[0149] In this way, by processing erroneous data in advance, the client ensures that the historical data input into the abnormal prediction model is relatively reasonable, thereby ensuring the reasonableness of the obtained prediction data.

[0150] In step S502, the trend data of historical data within the detection period is determined based on the trend component data.

[0151] Optionally, the client can perform fitting processing on the trend component data of historical data at preset timestamps within the detection period to obtain the trend data of the historical data within the detection period. Each sub-data point in this trend data can contain timestamp information; for example, the timestamps between two adjacent sub-data points can be one minute apart.

[0152] In step S503, the trend data carrying timestamp information is determined as the prediction data.

[0153] Optionally, the client can determine the trend data carrying timestamp information as the prediction data. For example, if the historical data is the revenue corresponding to each timestamp information of the week before last, then the prediction data is the predicted revenue corresponding to each timestamp information of the week before last.

[0154] In this way, the client can determine the predicted data corresponding to the current data based on the anomaly prediction model, thus preparing for the subsequent judgment of whether the current data of the detected object is abnormal.

[0155] Optionally, the client can determine the evaluation value based on the overall predicted data and the overall current data. Optionally, the evaluation value can be the difference between the predicted data and the current data.

[0156] Optionally, the client can divide the predicted data and current data into different sub-data according to actual needs, and then obtain the evaluation value based on the processing of the sub-data.

[0157] Figure 6 This is a flowchart illustrating a method for determining an evaluation value according to an exemplary embodiment, such as... Figure 6 As shown, it includes the following steps:

[0158] In step S601, the predicted data and the current data are processed in segments to obtain a sub-predicted data set and a sub-current data set; the sub-predicted data in the sub-predicted data set and the sub-current data in the sub-current data set correspond one-to-one in terms of timestamp information.

[0159] Optionally, the predicted data and current data can be segmented to obtain 7 sub-predicted data and 7 sub-current data, which are the data for each day from Monday to Sunday.

[0160] In step S602, the sub-evaluation value is determined based on the sub-predicted data and sub-current data with the same timestamp information.

[0161] The client can determine the first sub-evaluation value based on the sub-predicted data and sub-current data for Monday, the second sub-evaluation value based on the sub-predicted data and sub-current data for Tuesday, the third sub-evaluation value based on the sub-predicted data and sub-current data for Wednesday, the fourth sub-evaluation value based on the sub-predicted data and sub-current data for Thursday, the fifth sub-evaluation value based on the sub-predicted data and sub-current data for Friday, the sixth sub-evaluation value based on the sub-predicted data and sub-current data for Saturday, and the seventh sub-evaluation value based on the sub-predicted data and sub-current data for Sunday.

[0162] In step S603, the evaluation value is determined based on multiple sub-evaluation values.

[0163] Optionally, the client can determine the average evaluation value based on the above 7 sub-evaluation values, and use the average evaluation value as the final evaluation value.

[0164] In step S204, if the evaluation value is greater than or equal to the abnormal threshold, it is determined that the current detection object has an abnormality in the time period corresponding to the current data.

[0165] Optionally, the client can compare the final evaluation value with the anomaly threshold. If the evaluation value is greater than or equal to the anomaly threshold, it is determined that the detected object has an anomaly within the time period corresponding to the current data.

[0166] Alternatively, the client can compare each of the seven sub-evaluation values ​​with the anomaly threshold. If any sub-evaluation value is greater than or equal to the anomaly threshold, it is determined that the detected object has an anomaly within the time period corresponding to the sub-evaluation value.

[0167] The anomaly threshold and detection duration in steps S201-S204 above can refer to relatively reasonable thresholds and durations obtained after continuous updates and adjustments based on a large number of anomaly feedbacks. Therefore, in subsequent steps, as long as the evaluation value is greater than or equal to the anomaly threshold, it can be determined that the detected object has an anomaly within the time period corresponding to the current data. That is, steps S201-S204 are the application of object state determination. Before this, there will inevitably be unreasonable stages in the anomaly threshold and / or detection duration. In other words, there can be temporary anomaly thresholds and temporary detection durations. Therefore, during the stage of temporary anomaly thresholds and temporary detection durations, the operation can also be carried out according to steps S201-S204. However, when the evaluation value is greater than or equal to the anomaly threshold, the current data and historical data can be manually verified to continuously adjust the anomaly threshold and / or detection duration.

[0168] Optionally, the client can obtain detection data for historical detection objects. These historical and current detection objects can be the same object from different time periods, such as both being information A. They can also be detection objects of the same type, such as information A and information B. The detection data for historical objects can include their dimension information, attribute information, historical data, and current data. Next, the client can determine the anomaly threshold corresponding to the historical detection object based on the dimension and attribute information, determine the predicted data for the historical detection object based on the historical data, and determine the evaluation value of the historical detection object based on the predicted data and current data. If the evaluation value of the historical detection object is greater than or equal to the anomaly threshold, the anomaly status of the historical detection object can be determined based on the current data. If the historical detection object does not have any anomalies, the anomaly threshold can be determined to be inaccurate, and an updated anomaly threshold can be obtained based on the dimension information, attribute information, and anomaly threshold of the historical detection object.

[0169] In summary, the client can obtain a relatively scientific and rigorous prediction based on historical data, then compare the current data with the prediction data to obtain an evaluation value, and then determine whether the detected object has an anomaly based on the evaluation value. This is simple, convenient, and saves human resources.

[0170] Figure 7 This is a block diagram illustrating an object state determination apparatus according to an exemplary embodiment. (Refer to...) Figure 7 The device includes a data acquisition module 701, an anomaly threshold determination module 702, an evaluation determination module 703, and an anomaly judgment module 704.

[0171] The data acquisition module 701 is configured to acquire the detection data of the current detection object; the detection data includes dimension information, attribute information, historical data, and current data;

[0172] The anomaly threshold determination module 702 is configured to determine the anomaly threshold corresponding to the current detection object based on dimension information and attribute information.

[0173] The evaluation determination module 703 is configured to determine an evaluation value based on historical and current data.

[0174] The anomaly detection module 704 is configured to determine that if the evaluation value is greater than or equal to the anomaly threshold, the current detection object has an anomaly in the time period corresponding to the current data.

[0175] In some possible embodiments, the evaluation and determination module is configured to perform:

[0176] Determine forecast data based on historical data;

[0177] The assessment value is determined based on both forecast and current data.

[0178] The duration corresponding to historical data, the duration corresponding to current data, and the duration corresponding to predicted data are all the same detection duration.

[0179] In some possible embodiments, the data acquisition module is configured to perform:

[0180] Obtain the dimensional information, historical data, and current data of the current detection object; the dimensional information includes the recommended display page of the current detection object, its industry, and traffic entry information;

[0181] The attribute information of the current detection object is determined based on the dimensional information; the attribute information includes the consumption data and exposure data of the current detection object.

[0182] In some possible embodiments, the anomaly threshold determination module is configured to perform:

[0183] Determine the first display time information of the currently detected object on the recommended display page within the time period corresponding to the current data;

[0184] Determine the second display time information of the currently detected object on the recommended display page within the time period corresponding to the historical data;

[0185] Determine the temporal correlation information between the first display time information and the second display time information;

[0186] The anomaly threshold corresponding to the current detection object is determined based on dimensional information, attribute information, and time correlation information.

[0187] In some possible embodiments, the evaluation and determination module is configured to perform:

[0188] Input historical data into the anomaly prediction model to obtain the trend component data of the historical data at a preset timestamp within the detection period;

[0189] The trend data of historical data within the detection period is determined based on the trend component data;

[0190] Trend data carrying timestamp information is identified as forecast data.

[0191] In some possible embodiments, the evaluation and determination module is configured to perform:

[0192] Perform anomaly detection and processing on historical data, delete abnormal data, and obtain processed historical data;

[0193] The processed historical data is input into the anomaly prediction model to obtain the trend component data of the historical data at preset timestamps within the detection period.

[0194] In some possible embodiments, the evaluation and determination module is configured to perform:

[0195] The predicted data and current data are segmented and processed to obtain a sub-predicted data set and a sub-current data set; the sub-predicted data in the sub-predicted data set and the sub-current data in the sub-current data set are in a one-to-one correspondence based on timestamp information;

[0196] The sub-evaluation value is determined based on sub-predicted data and sub-current data with the same timestamp information;

[0197] The evaluation value is determined based on multiple sub-evaluation values.

[0198] In some possible embodiments, the data acquisition module is configured to perform:

[0199] Historical data is obtained based on historical time periods;

[0200] Get the current data based on the current time period;

[0201] The historical time period is within the first time cycle; the current time period is within the second time cycle; the start time of the historical time period and the start time of the current time period are the same within the time cycle; the end time of the historical time period and the end time of the current time period are the same within the time cycle.

[0202] In some possible embodiments, the apparatus further includes:

[0203] The anomaly detection module is configured to determine the anomaly status of the historical detection object based on the current data of the historical detection object if the evaluation value corresponding to the historical detection object is greater than or equal to the anomaly threshold corresponding to the historical detection object.

[0204] The anomaly threshold update module is configured to, if no anomalies are found in historically detected objects, obtain an updated anomaly threshold based on the dimensional information, attribute information, and anomaly threshold of the historically detected objects. Regarding the apparatus in the above embodiments, the specific methods by which each module performs its operations have been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0205] Figure 8 This is a block diagram illustrating an electronic device 800 for determining the state of an object according to an exemplary embodiment.

[0206] This electronic device can be a server or a terminal device, and its internal structure diagram can be as follows: Figure 8As shown, the electronic device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for determining the state of an object.

[0207] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present disclosure and does not constitute a limitation on the electronic device to which the present disclosure is applied. A specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0208] In an exemplary embodiment, an electronic device is also provided, including: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the object state determination method as described in the embodiments of this disclosure.

[0209] In an exemplary embodiment, a computer-readable storage medium is also provided, which, when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the object state determination method of the present disclosure embodiments.

[0210] In an exemplary embodiment, a computer program product is also provided, comprising a computer program stored in a readable storage medium, wherein at least one processor of a computer device reads from the readable storage medium and executes the computer program, causing the computer device to perform the object state determination method of the present disclosure embodiments.

[0211] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0212] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0213] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for determining the state of an object, characterized in that, include: Obtain the detection data of the current object; the detection data includes dimensional information, attribute information, historical data, and current data; The current detection object is information; the dimension information includes the recommended display page of the current detection object, its industry, and traffic entry information. The attribute information includes the consumption data and exposure data of the currently detected object; The anomaly threshold corresponding to the current detection object is determined based on the dimension information and the attribute information; The anomaly threshold is a metric used to determine whether the current data is abnormal; An evaluation value is determined based on the historical data and the current data; the evaluation value is the difference between the historical data and the current data, or the difference between the predicted data obtained based on the historical data and the current data; If the evaluation value is greater than or equal to the anomaly threshold, it is determined that the currently detected object has an anomaly within the time period corresponding to the current data; Determining the anomaly threshold corresponding to the current detection object based on the dimension information and the attribute information includes: Determine the first display time information of the currently detected object on the recommended display page within the time period corresponding to the current data; Determine the second display time information of the currently detected object on the recommended display page within the time period corresponding to the historical data; Determine the time correlation information between the first display time information and the second display time information; The anomaly threshold corresponding to the current detection object is determined based on the dimensional information, the attribute information, and the time correlation information.

2. The object state determination method according to claim 1, characterized in that, The process of determining the evaluation value based on the historical data and the current data includes: The predicted data is determined based on the historical data. The evaluation value is determined based on the predicted data and the current data; The duration corresponding to the historical data, the duration corresponding to the current data, and the duration corresponding to the predicted data are all the same detection duration.

3. The object state determination method according to claim 2, characterized in that, The acquisition of detection data for the current detection object includes: Obtain the dimensional information, historical data, and current data of the currently detected object; The attribute information of the currently detected object is determined based on the dimensional information.

4. The object state determination method according to claim 2, characterized in that, The step of determining the prediction data based on the historical data includes: The historical data is input into the anomaly prediction model to obtain the trend component data of the historical data within the preset timestamp of the detection period; The trend data of the historical data within the detection period is determined based on the trend component data; The trend data carrying timestamp information is determined as the predicted data.

5. The object state determination method according to claim 4, characterized in that, The step of inputting the historical data into the anomaly prediction model to obtain the trend component data of the historical data within the preset timestamps of the detection period includes: The historical data is subjected to anomaly detection processing, and the abnormal data is deleted to obtain the processed historical data; The processed historical data is input into the anomaly prediction model to obtain the trend component data of the historical data at a preset timestamp within the detection period.

6. The object state determination method according to any one of claims 4-5, characterized in that, The process of determining the evaluation value based on the predicted data and the current data includes: The predicted data and the current data are segmented and processed to obtain a sub-predicted data set and a sub-current data set; the sub-predicted data in the sub-predicted data set and the sub-current data in the sub-current data set correspond one-to-one based on the timestamp information; A sub-evaluation value is determined based on sub-predicted data and sub-current data with the same timestamp information; The evaluation value is determined based on multiple sub-evaluation values.

7. The object state determination method according to claim 3, characterized in that, The process of obtaining the historical and current data of the currently detected object includes: The historical data is obtained based on historical time periods; The current data is obtained based on the current time period; The historical time period is within a first time cycle; the current time period is within a second time cycle; the start time of the historical time period and the start time of the current time period are the same within the time cycle; the end time of the historical time period and the end time of the current time period are the same within the time cycle.

8. The object state determination method according to claim 1, characterized in that, Before acquiring the detection data of the current detection object, the process also includes: If the evaluation value corresponding to a historical detection object is greater than or equal to the abnormal threshold corresponding to the historical detection object, the abnormality of the historical detection object is determined based on the current data of the historical detection object. If the historical detection object does not have any anomalies, the updated anomaly threshold is obtained based on the dimension information, attribute information and anomaly threshold of the historical detection object.

9. An object state determination device, characterized in that, include: The data acquisition module is configured to acquire the detection data of the current detection object; the detection data includes dimensional information, attribute information, historical data, and current data; the current detection object is information; the dimensional information includes the recommended display page of the current detection object, its industry, and traffic entry information; The attribute information includes the consumption data and exposure data of the currently detected object; An anomaly threshold determination module is configured to determine the anomaly threshold corresponding to the current detection object based on the dimension information and the attribute information. The anomaly threshold is a metric used to determine whether the current data is abnormal; The evaluation determination module is configured to determine an evaluation value based on the historical data and the current data; the evaluation value is the difference between the historical data and the current data or the difference between predicted data obtained based on the historical data and the current data; The anomaly detection module is configured to determine if the evaluated value is greater than or equal to the anomaly threshold, and if so, that the currently detected object has an anomaly in the time period corresponding to the current data. The abnormal threshold determination module is configured to determine the first display time information of the currently detected object on the recommended display page within the time period corresponding to the current data. Determine the second display time information of the currently detected object on the recommended display page within the time period corresponding to the historical data; Determine the temporal correlation information between the first display time information and the second display time information; determine the anomaly threshold corresponding to the current detection object based on the dimension information, the attribute information, and the temporal correlation information.

10. The object state determination device according to claim 9, characterized in that, The evaluation and determination module is configured to execute: The predicted data is determined based on the historical data. The evaluation value is determined based on the predicted data and the current data; The duration corresponding to the historical data, the duration corresponding to the current data, and the duration corresponding to the predicted data are all the same detection duration.

11. The object state determination device according to claim 10, characterized in that, The data acquisition module is configured to execute: Obtain the dimensional information, historical data, and current data of the currently detected object; The attribute information of the currently detected object is determined based on the dimensional information.

12. The object state determination device according to claim 10, characterized in that, The evaluation and determination module is configured to execute: The historical data is input into the anomaly prediction model to obtain the trend component data of the historical data within the preset timestamp of the detection period; The trend data of the historical data within the detection period is determined based on the trend component data; The trend data carrying timestamp information is determined as the predicted data.

13. The object state determination device according to claim 12, characterized in that, The evaluation and determination module is configured to execute: The historical data is subjected to anomaly detection processing, and the abnormal data is deleted to obtain the processed historical data; The processed historical data is input into the anomaly prediction model to obtain the trend component data of the historical data at a preset timestamp within the detection period.

14. The object state determination device according to any one of claims 12-13, characterized in that, The evaluation and determination module is configured to execute: The predicted data and the current data are segmented and processed to obtain a sub-predicted data set and a sub-current data set; the sub-predicted data in the sub-predicted data set and the sub-current data in the sub-current data set correspond one-to-one based on the timestamp information; A sub-evaluation value is determined based on sub-predicted data and sub-current data with the same timestamp information; The evaluation value is determined based on multiple sub-evaluation values.

15. The object state determination device according to claim 11, characterized in that, The data acquisition module is configured to execute: The historical data is obtained based on historical time periods; The current data is obtained based on the current time period; The historical time period is within a first time cycle; the current time period is within a second time cycle; the start time of the historical time period and the start time of the current time period are the same within the time cycle; the end time of the historical time period and the end time of the current time period are the same within the time cycle.

16. The object state determination device according to claim 9, characterized in that, The device further includes: The anomaly detection module is configured to determine the anomaly status of the historical detection object based on the current data of the historical detection object if the evaluation value corresponding to the historical detection object is greater than or equal to the anomaly threshold corresponding to the historical detection object. The anomaly threshold update module is configured to perform an update of the anomaly threshold based on the dimension information, attribute information, and anomaly threshold of the historically detected object if no anomaly is found.

17. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the object state determination method as described in any one of claims 1 to 8.

18. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the object state determination method as described in any one of claims 1 to 8.

19. A computer program product, characterized in that, The computer program product includes a computer program stored in a readable storage medium, wherein at least one processor of a computer device reads from the readable storage medium and executes the computer program, causing the computer device to perform the object state determination method as described in any one of claims 1 to 8.

Images