System change time determination method and device, equipment, medium and program product

Abstract

The invention provides a system change time determination method which can be applied to the technical field of big data. The system change time determination method comprises the following steps: acquiring N historical business data sets corresponding to N historical time windows; for each historical service data set, performing clustering processing on data in the M historical service data subsets to obtain N clustering results corresponding to the N historical service data sets; for each historical service data set, under the condition that at least one class cluster comprises an abnormal class cluster and a non-abnormal class cluster, determining a historical time sub-window corresponding to the non-abnormal class cluster, and obtaining N groups of candidate historical time sub-windows corresponding to the N historical service data sets; and determining system change time based on the N groups of candidate historical time sub-windows. The invention further provides a system change time determination device and equipment, a storage medium and a program product.

Description

Technical Field

[0001] The present disclosure relates to the field of big data, and more specifically to a method, device, equipment, medium and program product for determining system change time. Background Art

[0002] The system change process is an important part of enterprise information construction, which usually includes change application, review, testing, implementation and verification. In the change application stage, the change initiator needs to submit a change application form, specifying the time, content, reason, scope of impact and implementation plan of the change. The system administrator will review the change application, assess the impact and risk of the change on the system, and decide whether to approve the change.

[0003] In the process of implementing the concept of the present disclosure, the inventors found that there are at least the following problems in the related technology: (1) The method for determining the time of change of the relevant system relies on the personal experience of the system administrator, which leads to personal subjectivity in the determination of the system change time, and thus makes the accuracy of the system change time poor, and even the situation where novice system administrators provide wrong information. (2) The method for determining the time of change of the relevant system may have delays. When the system business changes slowly, it is difficult for the system administrator to provide timely and effective feedback, resulting in a certain deviation between the system change time and the actual system business requirements. (3) The method for determining the time of change of the relevant system relies on the manual collection of the system change time of each system, which leads to low efficiency. Summary of the invention

[0004] In view of the above problems, the present disclosure provides a method, apparatus, device, medium and program product for determining system change time.

[0005] According to a first aspect of the present disclosure, a method for determining a system change time is provided, comprising: obtaining N historical business data sets corresponding to N historical time windows, wherein each historical business data set includes M historical business data corresponding to M historical time sub-windows; for each historical business data set, clustering the M historical business data to obtain N clustering results corresponding to the N historical business data sets, wherein the clustering results include at least one cluster; for each historical business data set, when at least one cluster includes an abnormal cluster and a non-abnormal cluster, determining the historical time sub-window corresponding to the non-abnormal cluster to obtain N groups of candidate historical time sub-windows corresponding to the N historical business data sets; and determining the system change time based on the N groups of candidate historical time sub-windows.

[0006] According to an embodiment of the present disclosure, when at least one cluster includes abnormal clusters and non-abnormal clusters, determining the historical time subwindow corresponding to the non-abnormal cluster includes: performing the Nth data replacement process on the abnormal cluster to obtain the Nth replacement cluster; performing the Nth clustering process on the data in the Nth replacement cluster and the non-abnormal cluster to generate the Nth clustering result; iterative execution: when the Nth clustering result includes abnormal clusters and non-abnormal clusters, performing the N+1th data replacement process on the abnormal cluster to obtain the N+1th replacement cluster; performing the N+1th clustering process on the data in the N+1 replacement cluster and the non-abnormal cluster to generate the N+1th clustering result, until the clustering result does not include abnormal clusters, and a target clustering result is obtained, and the target clustering result includes at least one non-abnormal cluster; determining the historical time subwindow corresponding one-to-one to at least one non-abnormal cluster.

[0007] According to an embodiment of the present disclosure, performing the N+1th data replacement process on the abnormal clusters includes: sorting the non-abnormal clusters; determining the target non-abnormal clusters from the sorted non-abnormal clusters; and replacing the abnormal clusters with the target non-abnormal clusters.

[0008] According to an embodiment of the present disclosure, determining the historical time sub-window corresponding to the non-abnormal cluster also includes: when the Nth clustering result does not include the abnormal cluster, taking the Nth clustering result as the target clustering result, and the target clustering result includes at least one non-abnormal cluster; determining the historical time sub-window corresponding one-to-one to at least one non-abnormal cluster.

[0009] According to an embodiment of the present disclosure, the data volume of the data points included in the abnormal cluster meets the first preset threshold condition, and the numerical values ​​of the data points included in the abnormal cluster meet the second preset threshold condition.

[0010] According to an embodiment of the present disclosure, obtaining N historical business data sets corresponding to N historical time windows includes: obtaining N groups of original historical business data corresponding to N historical time windows; performing data screening processing on the N groups of original historical business data to obtain N groups of initial historical business data; and normalizing the N groups of initial historical business data to obtain N historical business data sets.

[0011] According to an embodiment of the present disclosure, data screening processing of N groups of original historical business data includes: screening out historical business data that meets preset screening conditions from the N groups of original historical business data, wherein the preset screening conditions include one of the following: a correlation coefficient greater than a first preset threshold, a data missing ratio less than a second preset threshold, a data zero value ratio less than a third preset threshold, and a data quantile value greater than a fourth preset threshold.

[0012] According to an embodiment of the present disclosure, determining the system change time based on N groups of candidate historical time sub-windows includes: sorting the N groups of candidate historical time sub-windows to obtain a historical time sub-window sequence; screening out at least one group of historical time sub-windows that are continuous in the time dimension from the historical time sub-window sequence; and determining the system change time based on at least one group of historical time sub-windows that are continuous in the time dimension.

[0013] According to an embodiment of the present disclosure, clustering processing of M historical business data includes: determining a neighborhood radius and a minimum number of neighbors; sorting the M historical business data; and clustering data in the sorted M historical business data based on the neighborhood radius and the minimum number of neighbors.

[0014] According to an embodiment of the present disclosure, the system change time determination method further includes: acquiring system change time execution result data; and updating the neighborhood radius and the minimum number of neighbors based on the system change time execution result data.

[0015] According to an embodiment of the present disclosure, updating the neighborhood radius and the minimum number of neighbors based on the system change time execution result data includes: obtaining a reference system change time; and updating the neighborhood radius and the minimum number of neighbors based on the reference system change time and the system change time execution result data.

[0016] A second aspect of the present disclosure provides a system change time determination device, comprising: a first acquisition module, used to acquire N historical business data sets corresponding to N historical time windows, wherein each historical business data set includes M historical business data corresponding to M historical time sub-windows; a clustering module, used to cluster data in the M historical business data for each historical business data set, and obtain N clustering results corresponding to the N historical business data sets, wherein the clustering results include at least one cluster; a first determination module, used to determine, for each historical business data set, the historical time sub-window corresponding to the non-abnormal cluster when at least one cluster includes an abnormal cluster and a non-abnormal cluster, and obtain N groups of candidate historical time sub-windows corresponding to the N historical business data sets; a second determination module, used to determine the system change time based on the N groups of candidate historical time sub-windows.

[0017] A third aspect of the present disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0018] The fourth aspect of the present disclosure further provides a computer-readable storage medium having a computer program or instructions stored thereon, which implements the steps of the above method when the above computer program or instructions are executed by a processor.

[0019] The fifth aspect of the present disclosure further provides a computer program product, including a computer program or instructions, which implement the steps of the above method when the above computer program or instructions are executed by a processor.

[0020] According to the embodiment of the present disclosure, the system change time is determined based on N historical business data sets corresponding to N historical time windows, rather than directly manually configured based on past experience, which reduces the dependence on personal experience, reduces the personal subjectivity in the process of determining the system change time, improves the accuracy of the system change time, makes the determined system change time more realistic, and improves the efficiency of the system change time determination process. When the system business changes, the system change time can also be adjusted in time. By clustering the data in the historical business data, and in the case where the clustering results include abnormal clusters, the system change time is determined only based on the historical time sub-window corresponding to the non-abnormal cluster, the business peak period can be accurately screened out, and the business peak period is avoided in the system change time, thereby ensuring the stable operation of the system and reducing the impact of the system change on the business. Therefore, at least partially solves the problem of low accuracy and low efficiency of the system change time caused by manually configuring the system change time based on past experience, and achieves the technical effect of improving the accuracy of the system change time, improving the efficiency of the system change time determination process, and ensuring the stable operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above contents and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0022] Figure 1 The application scenario diagram of the system change time determination method, apparatus, device, medium and program product according to the embodiment of the present disclosure is schematically shown;

[0023] Figure 2 A flowchart of a method for determining system change time according to an embodiment of the present disclosure is schematically shown;

[0024] Figure 3 A flowchart of a method for determining a system change time according to another embodiment of the present disclosure is schematically shown;

[0025] Figure 4 A flowchart of a method for acquiring N historical business data sets corresponding to N historical time windows according to an embodiment of the present disclosure is schematically shown;

[0026] Figure 5 A structural block diagram of a device for determining system change time according to an embodiment of the present disclosure is schematically shown; and

[0027] Figure 6 A block diagram of an electronic device suitable for implementing a method for determining a system change time according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0028] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0029] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0030] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0031] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0032] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0033] In the technical solution of the present disclosure, the user information (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0034] In the scenario of using personal information for automated decision-making, the methods, devices, and systems provided by the embodiments of the present disclosure provide users with corresponding operation portals for users to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered. The expression "automated decision-making" here refers to the activity of automatically analyzing and evaluating a person's behavioral habits, interests and hobbies, or economic, health, credit status, etc. through computer programs, and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by people who specialize in a certain field, have specialized experience, knowledge and skills, and have reached a certain level of professionalism.

[0035] It should be noted that the system change time determination method and device disclosed herein can be used in the field of big data technology, and can also be used in any field other than the field of big data technology. The application field of the system change time determination method and device disclosed herein is not limited.

[0036] For reasons such as technology upgrades, business development, and risk management, the system needs to be changed. System changes can be, for example, system upgrades and system migrations. For example, financial institutions may need to upgrade their systems after expanding new businesses. In order to reduce the impact of system changes on the normal operation of the business and improve the reliability and stability of the system, it is necessary to ensure that the system change time avoids the peak business volume period. For example, the system change time can be from 00:00 to 6:00 on Tuesday. The method for determining the relevant system change time is mostly based on manual experience or based on the previously configured system change time. For example, the system administrator may collect the previously configured system change time and compare it with the system change time filled in by the change initiator during the change application process. If the system change time filled in by the change initiator is not within the previously determined system change time, the change risk is prompted. However, with the growth of business volume, continuous development of technology, and changes in business needs, system business and system construction have become increasingly complex. Determining the system change time based solely on the previously configured system change time or manual experience will result in an unreasonable system change time, and there may be erroneous information in past experience, resulting in low accuracy of the system change time. In addition, when the system business changes slowly, it is difficult for system managers to provide effective feedback in a timely manner, resulting in certain deviations in the determination of the system change time.

[0037] In order to at least partially solve the above technical problems, an embodiment of the present disclosure provides a method for determining a system change time, comprising: obtaining N historical business data sets corresponding to N historical time windows, wherein each historical business data set includes M historical business data corresponding to M historical time sub-windows; for each historical business data set, clustering the data in the M historical business data to obtain N clustering results corresponding to the N historical business data sets, wherein the clustering results include at least one cluster; for each historical business data set, when at least one cluster includes an abnormal cluster and a non-abnormal cluster, determining the historical time sub-window corresponding to the non-abnormal cluster to obtain N groups of candidate historical time sub-windows corresponding to the N historical business data sets; and determining the system change time based on the N groups of candidate historical time sub-windows.

[0038] Figure 1 The application scenario diagram of the system change time determination method, apparatus, device, medium and program product according to the embodiments of the present disclosure is schematically shown.

[0039] like Figure 1As shown, the application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0040] The user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0041] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0042] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0043] In the application scenario of the embodiment of the present disclosure, the user can use at least one of the first terminal device 101, the second terminal device 102, and the third terminal device 103 to initiate a request to the server 105 to determine the system change time. In response to the user request, the server 105 can be used to execute the system change time determination method of the embodiment of the present disclosure, including: obtaining N historical business data sets corresponding to N historical time windows, wherein each historical business data set includes M historical business data corresponding to M historical time sub-windows; for each historical business data set, clustering the M historical business data to obtain N clustering results corresponding to the N historical business data sets, wherein the clustering results include at least one cluster; for each historical business data set, when at least one cluster includes an abnormal cluster and a non-abnormal cluster, determining the historical time sub-window corresponding to the non-abnormal cluster to obtain N groups of candidate historical time sub-windows corresponding to the N historical business data sets; and determining the system change time based on the N groups of candidate historical time sub-windows.

[0044] It should be noted that the system change time determination method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the system change time determination device provided in the embodiment of the present disclosure can generally be set in the server 105. The system change time determination method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the system change time determination device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0045] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0046] The following will be based on Figure 1 The scene described by Figure 2~Figure 4 The method for determining the system change time according to an embodiment of the present disclosure is described in detail.

[0047] Figure 2 The flowchart of the method for determining system change time according to an embodiment of the present disclosure is schematically shown.

[0048] like Figure 2 As shown, the system change time determination method of this embodiment includes operations S210 to S240.

[0049] In operation S210, N historical business data sets corresponding to N historical time windows are acquired, wherein each historical business data set includes M historical business data corresponding to M historical time sub-windows.

[0050] In operation S220, for each historical business data set, clustering processing is performed on the data in the M historical business data to obtain N clustering results corresponding to the N historical business data sets, wherein the clustering results include at least one cluster.

[0051] In operation S230, for each historical business data set, when at least one cluster includes an abnormal cluster and a non-abnormal cluster, a historical time subwindow corresponding to the non-abnormal cluster is determined to obtain N groups of candidate historical time subwindows corresponding to the N historical business data sets.

[0052] In operation S240 , the system change time is determined based on the N groups of candidate historical time sub-windows.

[0053] According to an embodiment of the present disclosure, in operation S210, in order to reduce the impact of system changes on the normal operation of the business, it is necessary to ensure that the system change time avoids the peak period of business volume, so the business volume in the predetermined historical time interval can be referenced to screen out the peak period of business volume, so as to accurately determine the system change time. The predetermined historical time interval can correspond to N historical time windows, for example: the predetermined historical time interval can be two weeks, 1 historical time window can correspond to 1 day, so the predetermined historical time interval can correspond to 14 historical time windows.

[0054] For each historical time window, it can be divided into M historical time sub-windows according to the preset duration. For example, one historical time window can correspond to one day, and the preset duration can be 30 minutes. One historical time window can be divided into 48 historical time sub-windows, and the duration of each historical time sub-window is 30 minutes.

[0055] For each historical time sub-window, the mean of multiple business volume data corresponding to the historical time sub-window can be obtained to obtain the historical business data corresponding to the historical time sub-window. The business volume data can characterize the size of the business volume generated in each predetermined time period. Specifically, for each historical time sub-window, the business volume data in each predetermined time period included in the historical time sub-window can be determined, and the average value of all business volume data corresponding to the historical time sub-window can be calculated to obtain the historical business data. By calculating the average value of all business volume data corresponding to the historical time sub-window, the influence of noise can be reduced, and the data granularity can be increased, so as to better determine the data distribution law of the historical business data.

[0056] For example, the predetermined period may be one minute. The traffic volume data generated every minute may be obtained. When the duration of a historical time sub-window is 30 minutes, the traffic volume per minute may be determined, and 30 traffic volume data may be obtained. The average value of the above 30 traffic volume data may be calculated to obtain a historical traffic data.

[0057] According to an embodiment of the present disclosure, in operation S220, in order to determine whether each historical time window includes a business peak period, clustering processing can be performed on M historical business data corresponding to the historical time window, so that historical business data with similar business volume are divided into a cluster. A clustering result may include one or more clusters.

[0058] According to an embodiment of the present disclosure, in operation S230, for each clustering result, it can be determined whether the clustering result contains abnormal clusters. For example, the abnormal cluster may be a cluster in which the proportion of the number of historical business data contained in all clusters is less than a certain threshold, and the total business volume corresponding to all the historical business data included is higher than a preset total business volume threshold.

[0059] One cluster may correspond to one or more historical business data, and the total amount of all historical business data included in a cluster can be determined. Since historical business data can represent the business volume generated in a historical time subwindow, the total business volume corresponding to all historical business data included in a cluster can be determined.

[0060] For peak business volume, it may correspond to a period when users occasionally handle more business, resulting in an occasional surge in business volume, or it may correspond to a period that is obviously different from normal business, such as batch execution. Peak business volume usually presents a higher business volume in a shorter period of time. For each cluster, when the proportion of the number of historical business data contained in it to the number of all historical business data contained in all clusters is less than the preset number threshold, it means that the number of historical time sub-windows corresponding to the cluster is small, that is, it corresponds to a shorter period; when the total business volume corresponding to all historical business data included in a cluster is higher than the preset total business volume threshold, it means that the business volume corresponding to the cluster is too high. That is, the period corresponding to the abnormal cluster is shorter and the business volume is too high, so the historical time sub-window corresponding to the abnormal cluster may be a peak business volume period. In addition, the historical business data in the abnormal cluster may also be noise, which will interfere with the accuracy of determining the system change time.

[0061] For example: the preset total business volume threshold is 400. For a certain historical business data set, it includes 40 historical business data corresponding to 40 historical time sub-windows. The above 40 historical business data are clustered, and the clustering results obtained include 5 clusters. The number of historical business data contained in the above 5 clusters is 10, 9, 12, 2, and 7 respectively. The total business volume corresponding to all the historical business data included in the above 5 clusters is 80, 99, 230, 453, and 56 respectively. For the fourth cluster, its corresponding total business volume is 453, which is higher than the preset total business volume threshold, and it only includes 2 historical business data, accounting for the proportion of the number of all historical business data contained in all clusters (i.e. 40) is: 2 / 40=0.05, which is less than the preset number threshold 0.08, so the fourth cluster is an abnormal cluster.

[0062] Since abnormal clusters may correspond to peak traffic or noise, there is no need to consider the historical time sub-windows corresponding to abnormal clusters when determining the system change time. Only the historical time sub-windows corresponding to non-abnormal clusters need to be considered. Therefore, the historical time sub-windows corresponding to non-abnormal clusters can be determined to obtain candidate historical time sub-windows.

[0063] According to an embodiment of the present disclosure, when at least one cluster does not include an abnormal cluster, it means that the N historical time windows do not contain traffic peaks or noise. Therefore, the historical time sub-windows corresponding to all clusters in the clustering results can be directly used as candidate historical time sub-windows.

[0064] According to an embodiment of the present disclosure, in operation S240, since the candidate historical time subwindows corresponding to each historical business data set may be different, N groups of candidate historical time subwindows corresponding to N historical business data sets can be aggregated to determine the system change time.

[0065] According to the embodiments of the present disclosure, the system change time is determined based on N historical business data sets corresponding to N historical time windows, rather than directly manually configured based on past experience, which reduces the reliance on personal experience, reduces personal subjectivity in the process of determining the system change time, improves the accuracy of the system change time, makes the determined system change time more realistic, and improves the efficiency of the system change time determination process. By clustering the historical business data and including abnormal clusters in the clustering results, the system change time is determined only based on the historical time sub-window corresponding to the non-abnormal cluster, which can accurately screen out the business peak period and avoid including the business peak period in the system change time, thereby ensuring the stable operation of the system and reducing the impact of system changes on the business.

[0066] According to an embodiment of the present disclosure, in the case where at least one cluster includes an abnormal cluster and a non-abnormal cluster, determining a historical time sub-window corresponding to the non-abnormal cluster includes operations 11 to 15.

[0067] Operation 11: Perform the Nth data replacement process on the abnormal cluster to obtain the Nth replacement cluster.

[0068] Operation 12: Perform an N-th clustering process on the data in the N-th replacement cluster and the non-abnormal cluster to generate an N-th clustering result.

[0069] Operation 13, iterative execution: when the Nth clustering result includes abnormal clusters and non-abnormal clusters, perform the N+1th replacement process on the abnormal cluster to obtain the N+1th replacement cluster.

[0070] Operation 14: perform the N+1th clustering process on the data in the N+1th replacement cluster and the non-abnormal cluster to generate the N+1th clustering result, until the clustering result does not include abnormal clusters, and obtains the target clustering result, which includes at least one non-abnormal cluster.

[0071] Operation 15: Determine a historical time sub-window that corresponds one-to-one to at least one non-abnormal cluster.

[0072] According to an embodiment of the present disclosure, in operation 11, since the abnormal cluster corresponds to the business peak or noise, it is not necessary to consider the historical time sub-window corresponding to the abnormal cluster when determining the system change time. By replacing the abnormal cluster, the historical time sub-window corresponding to the abnormal cluster can be eliminated to avoid the historical time sub-window corresponding to the abnormal cluster from interfering with the determination process of the system change time.

[0073] According to an embodiment of the present disclosure, in operation 12, after obtaining the replacement cluster, the historical business data in the replacement cluster and the non-abnormal cluster can be clustered again to determine again whether the clustering result contains a cluster in which the proportion of the number of historical business data contained in all clusters is less than a preset number threshold, and the total business volume corresponding to all the historical business data included is higher than the preset total business volume threshold, that is, to determine again whether the clustering result contains abnormal clusters.

[0074] According to an embodiment of the present disclosure, the iterative execution operation corresponds to operation 13 and operation 14. Specifically, in operation 13, when it is determined that the clustering result contains an abnormal cluster, the newly determined abnormal cluster is replaced again, thereby eliminating the historical time sub-window corresponding to the abnormal cluster.

[0075] According to an embodiment of the present disclosure, in operation 14, the data in the replacement cluster and the non-abnormal cluster are clustered again for the N+1th time to generate the N+1th clustering result, and it is determined again whether the N+1th clustering result includes abnormal clusters and non-abnormal clusters. If the N+1th clustering result includes abnormal clusters and non-abnormal clusters, the abnormal cluster is replaced for the N+2th time to obtain the N+2th replacement cluster; the data in the N+2th replacement cluster and the non-abnormal cluster are clustered for the N+1th time to generate the N+2th clustering result, and if the N+2th clustering result includes abnormal clusters and non-abnormal clusters, the abnormal cluster is replaced for the N+3th time... until the clustering result does not include abnormal clusters and the target clustering result is obtained. In this way, all abnormal clusters can be thoroughly screened out.

[0076] It should be noted that the above-mentioned embodiment of performing iterative operations is only for the purpose of illustration and is not intended to limit the scope of the present disclosure.

[0077] According to an embodiment of the present disclosure, in operation 15, since the target clustering result does not include an abnormal cluster, the historical time subwindow corresponding to the target clustering result no longer corresponds to the business peak period, so the historical time subwindow corresponding to the target clustering result can be determined, and the system change time is determined based on the historical time subwindow corresponding to the target clustering result. Therefore, the system change time finally determined will not include the business peak period.

[0078] According to an embodiment of the present disclosure, performing the N+1th data replacement process on the abnormal clusters includes: sorting the non-abnormal clusters; determining the target non-abnormal clusters from the sorted non-abnormal clusters; and replacing the abnormal clusters with the target non-abnormal clusters.

[0079] According to an embodiment of the present disclosure, the total business volume corresponding to each non-abnormal cluster can be determined, and the non-abnormal clusters can be sorted according to the total business volume. For example, the non-abnormal clusters can be sorted from large to small according to the total business volume, or they can be sorted from small to large according to the total business volume.

[0080] According to an embodiment of the present disclosure, determining a target non-abnormal cluster from the sorted non-abnormal clusters may be: selecting a cluster whose total business volume is greater than a preset business volume threshold from the sorted non-abnormal clusters to obtain a target non-abnormal cluster.

[0081] According to an embodiment of the present disclosure, the replacement process of the abnormal cluster with the target non-abnormal cluster may be: determining the business volume corresponding to each historical business data in the target non-abnormal cluster, and selecting the historical business data with the largest business volume to obtain the target historical business data. All data in the abnormal cluster may be replaced with the target historical business data.

[0082] For example: the preset business volume threshold is 200. For a certain historical business data set, a total of 5 clusters are obtained. Among them, the 4th cluster is an abnormal cluster, and the other 4 clusters are non-abnormal clusters. The total business volumes corresponding to the above 4 non-abnormal clusters are 80 (Cluster 1), 99 (Cluster 2), 230 (Cluster 3), and 56 (Cluster 5), respectively. The non-abnormal clusters are sorted from large to small according to the total business volume, and the sorting results are: Cluster 3, Cluster 2, Cluster 1, Cluster 5. The cluster whose total business volume is greater than the preset business volume threshold is Cluster 3, that is, Cluster 3 is the target non-abnormal cluster.

[0083] For the above cluster 3, there are 4 historical business data in total. The business volume corresponding to each historical business data is determined, and the historical business data with the largest business volume is selected: historical business data A. Since the abnormal cluster contains 2 historical business data in total, the 2 historical business data contained in the abnormal cluster can be replaced by historical business data A.

[0084] It should be noted that the above embodiment of performing data replacement processing on abnormal clusters is only for illustrative purposes and is not intended to limit the scope of the present disclosure.

[0085] According to an embodiment of the present disclosure, by replacing the abnormal cluster with the target non-abnormal cluster, the historical time sub-window corresponding to the abnormal cluster can be eliminated, thereby reducing the probability that the system change time includes the business peak period.

[0086] According to an embodiment of the present disclosure, determining the historical time sub-window corresponding to the non-abnormal cluster also includes: when the Nth clustering result does not include the abnormal cluster, taking the Nth clustering result as the target clustering result, and the target clustering result includes at least one non-abnormal cluster; determining the historical time sub-window corresponding one-to-one to at least one non-abnormal cluster.

[0087] According to an embodiment of the present disclosure, when no abnormal cluster is included in the Nth clustering result, no replacement processing is required, and a historical time sub-window corresponding to at least one non-abnormal cluster can be directly determined.

[0088] Figure 3 The flowchart of a method for determining system change time according to another embodiment of the present disclosure is schematically shown.

[0089] like Figure 3 As shown, the system change time determination method of this embodiment includes operations S310 to S350.

[0090] In operation S310, the Nth data replacement process is performed on the abnormal cluster.

[0091] In operation S320 , an N-th clustering process is performed on the data in the N-th replacement cluster and the non-abnormal cluster.

[0092] In operation S330 , it is determined whether the Nth clustering result includes an abnormal cluster. If the Nth clustering result does not include an abnormal cluster, the Nth clustering result is used as a target clustering result.

[0093] In operation S340 , when the Nth clustering result includes an abnormal cluster, an N+1th replacement process is performed on the abnormal cluster to obtain an N+1th replacement cluster.

[0094] In operation S350 , the data in the N+1th replacement cluster and the non-abnormal cluster are clustered for the N+1th time to generate an N+1th clustering result.

[0095] According to the embodiments of the present disclosure, data replacement processing is performed on abnormal clusters so that the final clustering result no longer contains abnormal clusters, thereby ensuring that the historical time subwindow corresponding to the final clustering result no longer contains business peaks or noise. In this way, the interference of business peaks or noise can be eliminated, making the final system change time more accurate and reasonable, and effectively avoiding the situation where the system change time contains business peaks.

[0096] According to an embodiment of the present disclosure, the data volume of the data points included in the abnormal cluster meets the first preset threshold condition, and the numerical values ​​of the data points included in the abnormal cluster meet the second preset threshold condition.

[0097] According to an embodiment of the present disclosure, the data points included in the abnormal cluster may be the historical business data included in the abnormal cluster. The data volume of the data points included in the abnormal cluster may satisfy the first preset threshold condition: the ratio of the number of historical business data included in the abnormal cluster to the number of all historical business data included in all clusters is less than the preset quantity threshold. The value of the data points included in the abnormal cluster may satisfy the second preset threshold condition: the total business volume corresponding to all historical business data included in the abnormal cluster is higher than the preset business volume total threshold.

[0098] According to the embodiments of the present disclosure, abnormal clusters are determined simultaneously from two dimensions: the amount of data points included in the abnormal clusters and the total amount of business volume corresponding to the abnormal clusters. The distribution pattern of business volume data corresponding to business volume peaks or noise is fully considered, so that abnormal clusters can be accurately screened out, and the screened abnormal clusters are more in line with actual business scenarios.

[0099] Figure 4 A flowchart of a method for acquiring N historical business data sets corresponding to N historical time windows according to an embodiment of the present disclosure is schematically shown.

[0100] like Figure 4 As shown, the system change time determination method of this embodiment includes operations S410 to S430.

[0101] In operation S410, N groups of original historical business data corresponding to N historical time windows are obtained.

[0102] In operation S420, data screening is performed on the N groups of original historical business data to obtain N groups of initial historical business data.

[0103] In operation S430, normalization processing is performed on the N groups of initial historical business data to obtain N historical business data sets.

[0104] According to an embodiment of the present disclosure, the original historical business data may be data characterizing the size of the business volume at each moment in each historical time window.

[0105] According to an embodiment of the present disclosure, the normalization process may be a maximum-minimum normalization process, and the method of the maximum-minimum normalization process is shown in formula (1).

[0106] (1)

[0107] Among them, i is the initial historical business data, max and min are the maximum and minimum values ​​in N groups of initial historical business data, and value is the final historical business data. Multiple historical business data can form a historical business data set.

[0108] According to an embodiment of the present disclosure, specifically, performing data screening processing on N groups of original historical business data includes: screening out historical business data that meets a preset screening condition from the N groups of original historical business data.

[0109] According to an embodiment of the present disclosure, the preset screening conditions include one of the following: the correlation coefficient is greater than a first preset threshold, the data missing ratio is less than a second preset threshold, the data zero value ratio is less than a third preset threshold, and the data quantile value is greater than a fourth preset threshold.

[0110] According to an embodiment of the present disclosure, the correlation coefficient may be a Pearson correlation coefficient, and a first preset threshold may be set based on past experience, and the first preset threshold may be, for example, 0.8. The calculation method of the Pearson correlation coefficient is shown in formula (2).

[0111] (2)

[0112] Among them, r is the Pearson correlation coefficient, x i and i are the values ​​of any two original historical business data. and are the sample means of the two original historical business data. Formula (2) represents the covariance of two variables divided by their respective standard deviations. The value range of the Pearson correlation coefficient is between -1 and 1, where 0 indicates no linear relationship, a positive value indicates a positive correlation, and a negative value indicates a negative correlation.

[0113] According to the embodiments of the present disclosure, by performing correlation analysis on the original historical business data, it can be ensured that the original historical business data has not changed significantly within a certain time range, and that multiple original historical business data have similar distributions.

[0114] According to an embodiment of the present disclosure, the data missing ratio can be determined by the number of original historical business data that should be obtained in theory and the number of original historical business data that should be obtained in practice. Data zero value can be data whose original historical business data is 0. The data quantile value can be, for example, the 15% quantile of the data. The data quantile value is greater than the fourth preset threshold value, which can ensure that the business volume of the original historical business data is avoided from being too low, thereby avoiding inaccurate clustering results when the business volume level is too low. Each preset threshold value can be set based on past experience. For example, the second preset threshold value can be, for example, 50%; the third preset threshold value can be, for example, 50%, and the fourth preset threshold value can be, for example, 30, but it is not limited thereto. According to actual needs, the first preset threshold value, the second preset threshold value, the third preset threshold value, and the fourth preset threshold value can also be other values.

[0115] It should be noted that the above embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure.

[0116] According to the embodiments of the present disclosure, by performing data screening processing on N groups of original historical business data, problematic original historical business data can be excluded, thereby improving the accuracy of the ultimately determined system change time.

[0117] According to an embodiment of the present disclosure, determining the system change time based on N groups of candidate historical time sub-windows includes operations 21 to 23.

[0118] In operation 21, the N groups of candidate historical time sub-windows are sorted to obtain a historical time sub-window sequence.

[0119] In operation 22, at least one group of historical time sub-windows that are continuous in the time dimension is screened out from the historical time sub-window sequence.

[0120] In operation 23 , the system change time is determined based on at least one set of historical time sub-windows that are continuous in the time dimension.

[0121] According to an embodiment of the present disclosure, in operation 21, sorting N groups of candidate historical time sub-windows may include two methods. For the first method, the N groups of candidate historical time sub-windows may be sorted directly in chronological order to obtain a historical time sub-window sequence; for the second method, the N groups of candidate historical time sub-windows may be first divided into a first category of candidate historical time sub-windows and a second category of candidate historical time sub-windows. Among them, the first category of candidate historical time sub-windows and the second category of candidate historical time sub-windows respectively include multiple groups of candidate historical time sub-windows, the first category of candidate historical time sub-windows may correspond to working days, and the second category of candidate historical time sub-windows may correspond to non-working days. Among them, after the division is completed, the first category of candidate historical time sub-windows and the second category of candidate historical time sub-windows may be sorted in chronological order, respectively, to obtain two historical time sub-window sequences: a first category of historical time sub-window sequence and a second category of historical time sub-window sequence.

[0122] According to an embodiment of the present disclosure, in operation 22, in a first manner, at least one group of historical time sub-windows that are continuous in the time dimension can be directly screened out from the historical time sub-window sequence; in a second manner, at least one group of historical time sub-windows that are continuous in the time dimension can be respectively screened out from the first type of historical time sub-window sequence and the second type of historical time sub-window sequence.

[0123] According to an embodiment of the present disclosure, in operation 23, the start time and the end time corresponding to each historical time sub-window may be determined, and the start time and the end time corresponding to each historical time sub-window may be summarized to obtain the system change time.

[0124] For example, data from Monday to Friday is collected for a total of 5 working days, and Monday to Friday corresponds to a total of 5 historical time windows. Furthermore, Monday to Friday corresponds to a total of N groups of candidate historical time sub-windows. Among them, the candidate historical time sub-windows corresponding to Monday include: 00:00-00:30, 00:30-01:00; the candidate historical time sub-windows corresponding to Tuesday include: 00:30-01:00, 01:00-01:30, 01:30-02:00; the candidate historical time sub-windows corresponding to Wednesday include: 01:00-01:30, 01:30-02:00; the candidate historical time sub-windows corresponding to Thursday include: 00:30-01:00, 01:30-02:00; the candidate historical time sub-windows corresponding to Friday include: 00:00-00:30, 00:30-01:00, 01:00-01:30, 01:30-02:00, 03:30-04:00.

[0125] The above N groups of candidate historical time sub-windows can be sorted, and the resulting historical time sub-window sequence is: 00:00-00:30, 00:30-01:00, 01:00-01:30, 01:30-02:00, 03:30-04:00. The historical time sub-windows that are continuous in the time dimension selected from the above historical time sub-window sequence are: 00:00-00:30, 00:30-01:00, 01:00-01:30, 01:30-02:00. The start and end times of the above historical time sub-windows are summarized, and the system change time is: 00:00-02:00.

[0126] According to an embodiment of the present disclosure, after determining the candidate historical time sub-windows, by sorting the candidate historical time sub-windows, historical time sub-windows that are continuous in the time dimension can be accurately and efficiently screened out, so that the system change time can be determined simply, efficiently and accurately.

[0127] According to an embodiment of the present disclosure, clustering processing of M historical business data includes: determining a neighborhood radius and a minimum number of neighbors; sorting the M historical business data; and clustering the sorted M historical business data based on the neighborhood radius and the minimum number of neighbors.

[0128] According to an embodiment of the present disclosure, the values ​​of the neighborhood radius and the minimum number of neighbors can be set based on past experience. For example, the neighborhood radius can be 0.05 and the minimum number of neighbors can be 1, but it is not limited to this. The specific values ​​of the neighborhood radius and the minimum number of neighbors can also be adjusted according to actual business needs.

[0129] According to an embodiment of the present disclosure, M historical business data may be sorted based on the business volume corresponding to each historical business data. For example, the M historical business data may be sorted from large to small based on the business volume corresponding to each historical business data.

[0130] According to an embodiment of the present disclosure, clustering processing of the M historical business data after sorting based on the neighborhood radius and the minimum number of neighbors can be: calculating the distance between each historical business data and the adjacent historical business data, marking the historical business data whose distance is less than or equal to a preset distance threshold as a cluster, and setting the unmarked historical business data as noise.

[0131] According to the embodiments of the present disclosure, the historical business data are all one-dimensional data. By sorting the historical business data according to the business volume and then clustering them, the clustering speed can be significantly improved, thereby improving the efficiency of determining the system change time.

[0132] According to an embodiment of the present disclosure, after determining the system change time, the system change time execution result data may also be obtained; based on the system change time execution result data, the neighborhood radius and the minimum number of neighbors are updated.

[0133] According to an embodiment of the present disclosure, after the system change time is determined, the system change time can be applied to an actual business scenario, and user feedback on the system change time can be obtained to obtain system change time execution result data.

[0134] According to the embodiments of the present disclosure, after determining the system change time, the system change time execution result data fed back by users is collected to update the neighborhood radius and the minimum number of neighbors, thereby timely updating the system change time, so that the determined system change time can be more in line with user needs. Moreover, when the system business slowly changes, the system change time can also be updated in time, avoiding the deviation between the system change time and the actual system business needs.

[0135] According to an embodiment of the present disclosure, specifically, updating the neighborhood radius and the minimum number of neighbors based on the system change time execution result data includes: obtaining a reference system change time; and updating the neighborhood radius and the minimum number of neighbors based on the reference system change time and the system change time execution result data.

[0136] According to an embodiment of the present disclosure, the reference system change time may be a system change time reported by users that is more in line with actual business needs. Based on the reference system change time and the system change time execution result data, updating the neighborhood radius and the minimum number of neighbors may be: comparing the system change time execution result data with the reference system change time to determine the overlap between the reference system change time and the system change time. When the overlap between the reference system change time and the system change time is low, it means that the system change time is not reasonable enough, and therefore the system change time needs to be readjusted.

[0137] According to an embodiment of the present disclosure, the system change time may be readjusted as follows: when the duration of the system change time is much shorter than the duration of the reference system change time, the neighborhood radius and the minimum number of neighbors may be increased to obtain an updated system change time, so that the overlap between the updated system change time and the reference system change time is greater than a preset overlap threshold; when the duration of the system change time is much longer than the duration of the reference system change time, the neighborhood radius and the minimum number of neighbors may be reduced to obtain an updated system change time, so that the overlap between the updated system change time and the reference system change time is greater than a preset overlap threshold.

[0138] According to an embodiment of the present disclosure, by comparing the duration of the determined system change time to be much shorter than the duration of the reference system change time, thereby adjusting the neighborhood radius and the minimum number of neighbors, the adjustment range of the neighborhood radius and the minimum number of neighbors can be determined more accurately, so that the updated system change time is more in line with the actual needs of users.

[0139] It should be noted that the above embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure.

[0140] Based on the above system change time determination method, the present disclosure also provides a system change time determination device. Figure 5 The device is described in detail.

[0141] Figure 5 The structure block diagram of the device for determining system change time according to an embodiment of the present disclosure is schematically shown.

[0142] like Figure 5 As shown, the system change time determination device 500 of this embodiment includes a first acquisition module 510 , a clustering module 520 , a first determination module 530 and a second determination module 540 .

[0143] The first acquisition module 510 is used to acquire N historical business data sets corresponding to N historical time windows, wherein each historical business data set includes M historical business data corresponding to M historical time sub-windows. In one embodiment, the first acquisition module 510 can be used to perform the operation S210 described above, which will not be repeated here.

[0144] The clustering module 520 is used to perform clustering processing on the data in the M historical business data for each historical business data set, and obtain N clustering results corresponding to the N historical business data sets, wherein the clustering results include at least one cluster. In one embodiment, the clustering module 520 can be used to perform the operation S220 described above, which will not be repeated here.

[0145] The first determination module 530 is used to determine, for each historical business data set, a historical time subwindow corresponding to a non-abnormal cluster when at least one cluster includes an abnormal cluster and a non-abnormal cluster, and obtain N groups of candidate historical time subwindows corresponding to the N historical business data sets. In one embodiment, the first determination module 530 can be used to perform the operation S230 described above, which will not be repeated here.

[0146] The second determination module 540 is used to determine the system change time based on the N groups of candidate historical time sub-windows. In one embodiment, the second determination module 540 can be used to perform the operation S240 described above, which will not be described in detail here.

[0147] According to an embodiment of the present disclosure, the first determination module includes a replacement processing submodule, a first clustering processing submodule and an iterative execution submodule.

[0148] The replacement processing submodule is used to perform the Nth data replacement processing on the abnormal cluster to obtain the Nth replacement cluster; the first clustering processing submodule is used to perform the Nth clustering processing on the data in the Nth replacement cluster and the non-abnormal cluster to generate the Nth clustering result; the iterative execution submodule is used to iteratively perform the following operations: when the Nth clustering result includes abnormal clusters and non-abnormal clusters, perform the N+1th replacement processing on the abnormal cluster to obtain the N+1th replacement cluster; perform the N+1th clustering processing on the data in the N+1 replacement cluster and the non-abnormal cluster to generate the N+1th clustering result, until the initial clustering result does not include the abnormal cluster, and obtain the target clustering result, the target clustering result includes at least one non-abnormal cluster; determine the historical time subwindow corresponding one-to-one to at least one non-abnormal cluster.

[0149] According to an embodiment of the present disclosure, the iterative execution submodule includes a sorting unit, a first determining unit and a replacement processing unit.

[0150] The sorting unit is used to sort the non-abnormal clusters; the first determination unit is used to determine the target non-abnormal cluster from the sorted non-abnormal clusters; and the replacement processing unit is used to replace the abnormal cluster with the target non-abnormal cluster.

[0151] According to an embodiment of the present disclosure, the first determining module further includes a second determining unit and a third determining unit.

[0152] The second determination unit is used to use the Nth clustering result as the target clustering result when the Nth clustering result does not include an abnormal cluster, and the target clustering result includes at least one non-abnormal cluster; the third determination unit is used to determine a historical time sub-window corresponding one-to-one to at least one non-abnormal cluster.

[0153] According to an embodiment of the present disclosure, the first acquisition module includes a first acquisition submodule, a screening processing submodule and a normalization processing submodule.

[0154] The first acquisition submodule is used to obtain N groups of original historical business data corresponding to N historical time windows; the screening and processing submodule is used to perform data screening and processing on the N groups of original historical business data to obtain N groups of initial historical business data; the normalization processing submodule is used to perform normalization processing on the N groups of initial historical business data to obtain N historical business data sets.

[0155] According to an embodiment of the present disclosure, the screening and processing submodule includes screening out historical business data that meets preset screening conditions from N groups of historical business data, wherein the preset screening conditions include one of the following: a correlation coefficient greater than a first preset threshold, a data missing ratio less than a second preset threshold, a data zero value ratio less than a third preset threshold, and a data quantile value greater than a fourth preset threshold.

[0156] According to an embodiment of the present disclosure, the second determination module includes a first sorting processing submodule, a screening submodule and a first determination submodule.

[0157] The first sorting processing submodule is used to sort N groups of candidate historical time subwindows to obtain a historical time subwindow sequence; the screening submodule is used to screen out at least one group of historical time subwindows that are continuous in the time dimension from the historical time subwindow sequence; the first determination submodule is used to determine the system change time based on at least one group of historical time subwindows that are continuous in the time dimension.

[0158] According to an embodiment of the present disclosure, the clustering module includes a second determination submodule, a second sorting processing submodule and a second clustering processing submodule.

[0159] The second determination submodule is used to determine the neighborhood radius and the minimum number of neighbors; the second sorting processing submodule is used to sort the data in the M historical business data; the second clustering processing submodule is used to cluster the data in the sorted M historical business data based on the neighborhood radius and the minimum number of neighbors.

[0160] According to an embodiment of the present disclosure, the system change time determination device further includes a second acquisition module and an update module.

[0161] The second acquisition module includes acquiring system change time execution result data; the update module is used to update the neighborhood radius and the minimum number of neighbors based on the system change time execution result data.

[0162] According to an embodiment of the present disclosure, the update module includes a first acquisition submodule and an update submodule.

[0163] The first acquisition submodule includes acquiring the reference system change time; the update submodule is used to update the neighborhood radius and the minimum number of neighbors based on the reference system change time and the system change time execution result data.

[0164] According to an embodiment of the present disclosure, any multiple modules of the first acquisition module 510, the clustering module 520, the first determination module 530, and the second determination module 540 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the first acquisition module 510, the clustering module 520, the first determination module 530, and the second determination module 540 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware, and firmware, or in any appropriate combination of any of them. Alternatively, at least one of the first acquisition module 510 , the clustering module 520 , the first determination module 530 , and the second determination module 540 may be at least partially implemented as a computer program module, and when the computer program module is executed, a corresponding function may be performed.

[0165] Figure 6 A block diagram of an electronic device suitable for implementing a method for determining a system change time according to an embodiment of the present disclosure is schematically shown.

[0166] like Figure 6 As shown, the electronic device 600 according to an embodiment of the present disclosure includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 to a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a dedicated microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include an onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0167] In RAM 603, various programs and data required for the operation of electronic device 600 are stored. Processor 601, ROM 602 and RAM 603 are connected to each other via bus 604. Processor 601 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 602 and / or RAM 603. It should be noted that the program can also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in one or more memories.

[0168] According to an embodiment of the present disclosure, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to the bus 604. The electronic device 600 may further include one or more of the following components connected to the input / output (I / O) interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output (I / O) interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage portion 608 as needed.

[0169] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0170] According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present disclosure, the computer-readable storage medium may include the ROM 602 and / or RAM 603 described above and / or one or more memories other than ROM 602 and RAM 603.

[0171] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the system change time determination method provided by the embodiment of the present disclosure.

[0172] The above functions defined in the system / device of the embodiment of the present disclosure are performed when the computer program is executed by the processor 601. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0173] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 609, and / or installed from a removable medium 611. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0174] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, apparatus, module, unit, etc. described above can be implemented by a computer program module.

[0175] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0176] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0177] It will be appreciated by those skilled in the art that the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations or combinations are not explicitly described in the present disclosure. In particular, without departing from the spirit and teachings of the present disclosure, the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0178] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A method for determining system change time, characterized in that: The method comprises: Acquire N historical business data sets corresponding to the N historical time windows, wherein each of the historical business data sets includes M historical business data corresponding to the M historical time sub-windows; For each of the historical business data sets, clustering processing is performed on the M historical business data to obtain N clustering results corresponding to the N historical business data sets, wherein the clustering results include at least one cluster; For each of the historical business data sets, when the at least one cluster includes an abnormal cluster and a non-abnormal cluster, determining a historical time subwindow corresponding to the non-abnormal cluster, and obtaining N groups of candidate historical time subwindows corresponding to the N historical business data sets; Based on the N groups of candidate historical time sub-windows, a system change time is determined.

2. The method according to claim 1, characterized in that In the case where the at least one cluster includes an abnormal cluster and a non-abnormal cluster, determining the historical time subwindow corresponding to the non-abnormal cluster comprises: Performing an N-th data replacement process on the abnormal cluster to obtain an N-th replacement cluster; Performing an Nth clustering process on the data in the Nth replacement cluster and the non-abnormal cluster to generate an Nth clustering result; Iterative execution: when the Nth clustering result includes the abnormal cluster and the non-abnormal cluster, performing the N+1th data replacement process on the abnormal cluster to obtain the N+1th replacement cluster; Performing an N+1th clustering process on the data in the N+1th replacement cluster and the non-abnormal cluster to generate an N+1th clustering result, until the clustering result does not include the abnormal cluster, and obtaining a target clustering result, wherein the target clustering result includes at least one of the non-abnormal clusters; Determine a historical time sub-window that corresponds one-to-one to at least one of the non-abnormal clusters.

3. The method according to claim 2, characterized in that The N+1th data replacement process for the abnormal cluster includes: sorting the non-anomalous clusters; Determine a target non-abnormal cluster from the sorted non-abnormal clusters; The abnormal cluster is replaced by the target non-abnormal cluster.

4. The method according to claim 2, characterized in that: The determining of the historical time sub-window corresponding to the non-abnormal cluster further comprises: In the case that the Nth clustering result does not include the abnormal cluster, taking the Nth clustering result as the target clustering result, the target clustering result includes at least one of the non-abnormal clusters; Determine a historical time sub-window that corresponds one-to-one to at least one of the non-abnormal clusters.

5. The method according to claim 1, characterized in that The data volume of the data points included in the abnormal cluster meets the first preset threshold condition, and the value of the data points included in the abnormal cluster meets the second preset threshold condition.

6. The method according to claim 1, characterized in that The obtaining of N historical business data sets corresponding to N historical time windows includes: Obtaining N groups of original historical business data corresponding to the N historical time windows; Performing data screening processing on the N groups of original historical business data to obtain N groups of initial historical business data; The N groups of initial historical business data are normalized to obtain the N historical business data sets.

7. The method according to claim 6, characterized in that The data screening process for the N groups of original historical business data includes: Filter out historical business data that meets preset screening conditions from the N groups of original historical business data, wherein the preset screening conditions include one of the following: a correlation coefficient greater than a first preset threshold, a data missing ratio less than a second preset threshold, a data zero value ratio less than a third preset threshold, and a data quantile value greater than a fourth preset threshold.

8. The method according to claim 1, characterized in that The determining of the system change time based on the N groups of candidate historical time sub-windows includes: Sorting the N groups of candidate historical time sub-windows to obtain a historical time sub-window sequence; Filtering out at least one group of historical time sub-windows that are continuous in the time dimension from the historical time sub-window sequence; The system change time is determined based on the at least one group of historical time sub-windows that are continuous in the time dimension.

9. The method according to claim 1, characterized in that: The clustering process of the M historical service data includes: Determine the neighborhood radius and the minimum number of neighbors; Sorting the M historical business data; Based on the neighborhood radius and the minimum number of neighbors, clustering is performed on the M historical business data after sorting.

10. The method according to claim 9, characterized in that The method further comprises: Get the execution result data of system change time; Based on the system change time execution result data, the neighborhood radius and the minimum number of neighbors are updated.

11. The method according to claim 10, characterized in that The updating of the neighborhood radius and the minimum number of neighbors based on the system change time execution result data comprises: Get the reference system change time; Based on the reference system change time and the system change time execution result data, the neighborhood radius and the minimum number of neighbors are updated.

12. A system change time determination device, characterized in that: The device comprises: A first acquisition module, configured to acquire N historical business data sets corresponding to N historical time windows, wherein each of the historical business data sets includes M historical business data corresponding to M historical time sub-windows; A clustering module, configured to perform clustering processing on the data in the M historical business data sets for each of the historical business data sets to obtain N clustering results corresponding to the N historical business data sets, wherein the clustering results include at least one cluster; A first determination module is configured to determine, for each of the historical business data sets, a historical time subwindow corresponding to the non-abnormal cluster when the at least one cluster includes an abnormal cluster and a non-abnormal cluster, to obtain N groups of candidate historical time subwindows corresponding to the N historical business data sets; and The second determination module is used to determine the system change time based on the N groups of candidate historical time sub-windows.

13. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 11.

14. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 11 are implemented.

15. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 11 are implemented.

Images