Method, apparatus, and device for determining a data processing strategy

Abstract

The present application provides a method, apparatus, and device for determining a data processing strategy. The method includes: constructing a first data set based on the acquired data, where the first data set includes at least one piece of data related to the problem source, or includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal manifestation; respectively determining the similarity between the elements in a plurality of second data sets pre-stored in the first database and the elements in the first data set, where any second data set includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal manifestation; and determining the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored second data set and the processing strategy and the similarity. Through this method, the data processing strategy can be automatically determined, improving the determination efficiency.

Description

Technical Field

[0001] This application relates to data processing technologies, and in particular to a method, apparatus, and device for determining a data processing strategy. Background Art

[0002] With the development of society and the technological innovation of communication, communication services carried by mobile phone terminals have become an indispensable part of people's lives. While pursuing stable communication quality, users also put forward higher requirements for service experience. Only by ensuring the security and accuracy of user data can user experience be effectively guaranteed.

[0003] In the prior art, usually, staff members regularly verify user data, such as data on charging situations, package situations, etc., and process the data with faults based on experience.

[0004] However, this way of determining a processing strategy for telecom abnormal data relying on manual experience has low efficiency. Summary of the Invention

[0005] This application provides a method, apparatus, and device for determining a data processing strategy to solve the problem of low efficiency in the prior art of determining a processing strategy for telecom abnormal data.

[0006] On the one hand, this application provides a method for determining a data processing strategy, including: constructing a first data set according to the obtained data, where the first data set includes at least one piece of data related to the problem source, or includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal manifestation. The data related to the problem source includes one or more of the following data: user bill data, service data subscribed by the user. The data related to the abnormal manifestation includes one or more of the following data: communication abnormal data, tariff abnormal data, abnormal change of user information; respectively determining the similarity between the elements in a plurality of second data sets pre-stored in the first database and the elements in the first data set, where any second data set includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal manifestation; determining the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored second data set and the processing strategy and the similarity.

[0007] Optionally, constructing a first data set according to the obtained data includes: obtaining at least one third data corresponding to any second data pre-stored in the second database within a first preset time period, where the second data is data related to the problem source or data related to the abnormal manifestation; determining whether there is abnormal data in at least one third data according to the value of each third data, and if so, constructing the first data set with the abnormal data as an element.

[0008] Optionally, determining the similarity between the elements in a plurality of pre-stored second data sets and the elements in the first data set respectively includes: determining a first similarity between the data related to the problem source in each pre-stored second data set and the data related to the problem source in the first data set; determining a second similarity between the data related to the abnormal performance in each pre-stored second data set and the data related to the abnormal performance in the first data set.

[0009] Correspondingly, determining the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored second data set and the processing strategy, and the similarity includes: determining the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored second data set and the processing strategy, and each first similarity and each second similarity.

[0010] Optionally, determining the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored second data set and the processing strategy, and each first similarity and each second similarity includes: comparing each first similarity and each second similarity with a first preset threshold respectively, and determining the second data set corresponding to the first similarity greater than or equal to the first preset threshold, or the second data set corresponding to the second similarity greater than or equal to the first preset threshold as the first preselected data set; counting the number of the first preselected data sets; if the number of the first preselected data sets is greater than or equal to a second preset threshold, determining the target data set from the first preselected data sets; determining the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored target data set and the processing strategy.

[0011] Optionally, determining the target data set from the first preselected data sets includes: determining the first preselected data sets with the first similarity greater than or equal to a third preset threshold, or the second similarity greater than or equal to the third preset threshold as the second preselected data sets; counting the number of the second preselected data sets; if the number of the first preselected data sets is greater than or equal to a fourth preset threshold, determining the second preselected data sets as the target data sets.

[0012] Optionally, determining the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored target data set and the processing strategy includes: determining the frequency of occurrence of each processing strategy according to the corresponding relationship between each pre-stored target data set and the processing strategy; determining the processing strategy with the highest frequency of occurrence as the target processing strategy corresponding to the first data set.

[0013] Optionally, it further includes: obtaining at least one third data set, where the third data set is the first data set for which the corresponding target processing policy has not changed within the second preset duration; determining the support degree and confidence degree corresponding to any third data set according to the elements in each third data set; determining the third data set with a support degree greater than or equal to the fifth preset threshold and a confidence degree greater than or equal to the sixth preset threshold as the fourth data set; and updating the second database according to the elements in the fourth data set.

[0014] Optionally, updating the second database according to the elements in the fourth data set includes: for any fourth data set: determining the third similarity between the third subset included in the fourth data set and each second data set according to the elements in the fourth data set and the elements in each second data set; determining the fourth similarity between the fourth subset included in the fourth data set and each second data set according to the elements in the fourth data set and the elements in each second data set; determining the data set with only the third similarity greater than or equal to the seventh preset threshold, or the data set with only the fourth similarity greater than or equal to the seventh preset threshold as the risk data set; and updating the second database according to the elements in the risk data set.

[0015] Optionally, updating the second database according to the elements in the risk data set includes: obtaining a new third data set generated within the third preset duration; for the first element in the risk data set, determining whether the first element exists in the new third data set; and if the first element exists, supplementing the first element into the second database.

[0016] Optionally, constructing the first data set according to the obtained data includes: constructing the first data set according to the data in the obtained user input that is consistent with the preset element type.

[0017] Optionally, determining whether there is abnormal data in at least one third data according to the values of each third data includes: for any third data, calculating the standard deviation corresponding to the third data according to the value of the third data; determining the value range according to the standard deviation; and if at least one of the multiple values corresponding to the third data does not belong to the value range, determining that the third data is abnormal.

[0018] In a second aspect, the apparatus for determining a data processing policy provided by this application includes:

[0019] An acquisition module, configured to construct a first data set according to the acquired data. The first data set includes at least one piece of data related to the problem source, or includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal performance. The data related to the problem source includes one or more of the following types of data: user bill data, service data subscribed by the user. The data related to the abnormal performance includes one or more of the following types of data: communication anomaly data, tariff anomaly data, abnormal change of user profile.

[0020] A processing module, configured to respectively determine the similarity between the elements in a plurality of second data sets pre-stored in the first database and the elements in the first data set. Any second data set includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal performance.

[0021] The processing module is further configured to determine the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored second data set and the processing strategy and the similarity.

[0022] Optionally, the acquisition module is specifically configured to acquire at least one third data corresponding to any second data pre-stored in the second database within a first preset time period. The second data is data related to the problem source or data related to the abnormal performance. According to the values of each third data, determine whether there is abnormal data in at least one third data. If there is, construct the first data set with the abnormal data as elements.

[0023] Optionally, the processing module is specifically configured to determine the first similarity between the data related to the problem source in each pre-stored second data set and the data related to the problem source in the first data set. Determine the second similarity between the data related to the abnormal performance in each pre-stored second data set and the data related to the abnormal performance in the first data set. According to the corresponding relationship between each pre-stored second data set and the processing strategy, and each first similarity and each second similarity, determine the target processing strategy corresponding to the first data set.

[0024] Optionally, the processing module is specifically configured to compare each first similarity and each second similarity with a first preset threshold respectively, and determine the second data set corresponding to when the first similarity is greater than or equal to the first preset threshold, or the second data set corresponding to when the second similarity is greater than or equal to the first preset threshold as the first preselected data set. Count the number of the first preselected data sets. If the number of the first preselected data sets is greater than or equal to a second preset threshold, determine the target data set from the first preselected data sets. According to the corresponding relationship between each pre-stored target data set and the processing strategy, determine the target processing strategy corresponding to the first data set.

[0025] Optionally, the processing module is specifically configured to determine, as a second preselected data set, a first preselected data set in which the first similarity is greater than or equal to a third preset threshold, or the second similarity is greater than or equal to the third preset threshold; count the number of the second preselected data set; if the number of the first preselected data set is greater than or equal to a fourth preset threshold, determine the second preselected data set as the target data set.

[0026] Optionally, the processing module is specifically configured to determine the frequency of occurrence of each processing strategy according to the corresponding relationship between each pre-stored target data set and the processing strategy; determine the processing strategy with the highest frequency of occurrence as the target processing strategy corresponding to the first data set.

[0027] Optionally, the processing module is further configured to obtain at least one third data set, where the third data set is a first data set in which the corresponding target processing strategy has not changed within a second preset duration; determine the support degree and confidence degree corresponding to any one of the third data sets according to the elements in each third data set; determine, as a fourth data set, a third data set in which the support degree is greater than or equal to a fifth preset threshold and the confidence degree is greater than or equal to a sixth preset threshold; update the second database according to the elements in the fourth data set.

[0028] Optionally, the processing module is specifically configured to, for any one of the fourth data sets: determine the third similarity between the third subset included in the fourth data set and each second data set according to the elements in the fourth data set and the elements in each second data set; determine the fourth similarity between the fourth subset included in the fourth data set and each second data set according to the elements in the fourth data set and the elements in each second data set; determine, as a risk data set, a fourth data set in which only the third similarity is greater than or equal to a seventh preset threshold, or a fourth data set in which only the fourth similarity is greater than or equal to the seventh preset threshold; update the second database according to the elements in the risk data set.

[0029] Optionally, the processing module is specifically configured to obtain a newly generated third data set within a third preset duration; for a first element in the risk data set, determine whether the first element exists in the newly generated third data set; if the first element exists, supplement the first element into the second database.

[0030] Optionally, the obtaining module is specifically configured to construct a first data set according to the data consistent with the preset element type in the obtained user input data.

[0031] Optionally, the obtaining module is specifically configured to calculate the standard deviation corresponding to any third data according to the value of the third data for any third data; determine the value range according to the standard deviation; if at least one of the multiple values corresponding to the third data does not belong to the value range, determine that the third data is abnormal.

[0032] In a third aspect, the present disclosure provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method provided in the first aspect and its optional manners.

[0033] In a fourth aspect, the present disclosure provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the method provided in the first aspect or the optional manners of the first aspect.

[0034] In a fifth aspect, the present disclosure provides a computer program product, including computer programs / instructions, and when the computer programs / instructions are executed by a processor, they implement the method provided in the first aspect or the optional manners of the first aspect.

[0035] The present application provides a method, apparatus, and device for determining a data processing policy. By constructing a first data set according to the obtained data, the first data set includes at least one piece of data related to the problem source, or includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal manifestation, and the data related to the problem source includes one or more of the following data: user bill data, service data subscribed by the user, and the data related to the abnormal manifestation includes one or more of the following data: communication anomaly data, tariff anomaly data, abnormal change of user profile; respectively determine the similarity between the elements in a plurality of second data sets pre-stored in the first database and the elements in the first data set, and any second data set includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal manifestation; according to the corresponding relationship between each pre-stored second data set and the processing policy, and the similarity, determine the target processing policy corresponding to the first data set, which can automatically determine the data processing policy and improve the determination efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings here are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0037] Figure 1 It is a schematic diagram of an application scenario of the method for determining a data processing policy provided by the present application;

[0038] Figure 2 A flowchart of a method for determining a data processing strategy provided by this application;

[0039] Figure 3 Another flowchart of a method for determining a data processing strategy provided by this application;

[0040] Figure 4 A structural diagram of a device for determining a data processing strategy provided by this application;

[0041] Figure 5 A structural diagram of an electronic device provided by this application.

[0042] Explanation of reference numerals:

[0043] 11 - Terminal device; 12 - Server.

[0044] Through the above - mentioned drawings, specific embodiments of this application have been shown, and there will be more detailed descriptions hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of this application in any way, but to illustrate the concept of this application to those skilled in the art by referring to specific embodiments. Detailed implementation manners

[0045] Here, exemplary embodiments will be described in detail, and their examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with this application. On the contrary, they are merely examples of devices and methods consistent with some aspects of this application as detailed in the appended claims.

[0046] With the development of society and the technological innovation of communication, communication services carried by mobile phone terminals have become an indispensable part of people's lives. While users pursue stable communication quality, they also put forward higher requirements for service experience. Only by ensuring the security and accuracy of user data can the user experience be effectively guaranteed. In the prior art, usually, auditors regularly verify user data, such as data on deduction situations, package situations, etc., and process the faulty data based on experience. With the accelerating pace of life, users' requirements for processing speed are also getting higher and higher. This way of relying on manual experience to determine the processing strategy is inefficient and has a lag on the one hand, affecting the user experience; on the other hand, this way places high requirements on auditors, and the accuracy and reliability of the determined processing strategy will vary depending on the auditors; in addition, since the processing methods are summarized by auditors based on historical experience, they cannot handle newly emerging problems.

[0047] To solve the above technical problems, the present application provides a method for determining a data processing strategy. By obtaining data related to the problem source or data related to abnormal manifestations, and respectively determining the similarity between the elements in a plurality of pre-stored second data sets and the obtained data, and then according to the corresponding relationship between each pre-stored second data set and the processing strategy, as well as the determined similarity, automatically determine the target processing strategy corresponding to the obtained data, which can improve the determination efficiency of the processing strategy, reduce manual participation, and improve the accuracy and reliability of the processing strategy.

[0048] The method for determining a data processing strategy provided by the present application aims to solve the above technical problems in the prior art.

[0049] The following will specifically describe the technical solution of the present application and how the technical solution of the present application solves the above technical problems with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0050] Figure 1 FIG. is a schematic diagram of an application scenario of the method for determining a data processing strategy provided by the present application. As Figure 1 shown, this method can be applied to a scenario including a terminal device 11 and a server 12. The terminal device 11 and the server 12 are communicatively connected.

[0051] The user generates corresponding data by using the terminal device 11 to make calls, send and receive text messages, etc.

[0052] The server 12 is used to construct a first data set according to the obtained data. The first data set includes at least one item of data related to the problem source, or includes at least one item of data related to the problem source and at least one item of data related to abnormal manifestations. The data related to the problem source includes one or more of the following data: user bill data, service data subscribed by the user. The data related to abnormal manifestations includes one or more of the following data: communication anomaly data, tariff anomaly data, abnormal change of user profile; respectively determine the similarity between the elements in a plurality of pre-stored second data sets in the first database and the elements in the first data set. Any second data set includes at least one item of data related to the problem source and at least one item of data related to abnormal manifestations; according to the corresponding relationship between each pre-stored second data set and the processing strategy, as well as the similarity, determine the target processing strategy corresponding to the first data set.

[0053] Figure 2 FIG. is a schematic flowchart of the method for determining a data processing strategy provided by the present application. This method is applied to the server. AsFigure 2 As shown in Figure 2 , the method includes:

[0054] S201. Construct a first data set according to the acquired data.

[0055] The first data set includes at least one piece of data related to the problem source, or includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal performance.

[0056] The data related to the problem source includes one or more of the following data: user bill data, service data subscribed by the user. Exemplarily, the user bill data includes the tariff situation of a certain service subscribed by the user, the user's monthly bill, the user's traffic balance, etc.; the service data subscribed by the user includes data such as subscribing to a family affection tariff service, the service tables mainly involved in the operation, etc. It should be noted that this is only an example here, not an exhaustive list, and any data related to the problem source belongs to the content disclosed in this application.

[0057] The data related to the abnormal performance includes one or more of the following data: communication abnormal data, tariff abnormal data, abnormal change of user information. Exemplarily, the communication abnormal data includes being unable to make a call, unable to access a call, the user being unable to use the SMS function, etc.; the tariff abnormal data includes the user being charged for calling a family affection number, etc. It should be noted that this is only an example here, not an exhaustive list, and any data related to the abnormal performance belongs to the content disclosed in this application.

[0058] The data acquired by the server can be the data input by the first-level operation and maintenance. The first-level operation and maintenance makes a first-level judgment on the user problem, and writes the reasons for the escalated problem according to the verification process, which can specifically include "problem overview", "verification process", "problem source discrimination", and "final abnormal performance". Among them, "problem overview" and "verification process" can be used to assist manual verification. After the server acquires the data input by the first-level operation and maintenance, it analyzes and identifies the data related to the problem source and / or the data related to the abnormal performance in the data, and constructs a first data set according to the data related to the problem source and / or the data related to the abnormal performance. That is, the first data set may include the data related to the problem source and the data related to the abnormal performance, or only include one of the data, and the missing other data can be marked as empty.

[0059] Optionally, the data input by the first-level operation and maintenance can be a text description, or can be data in the form of coding identifiers such as the data table name, data bit name, special number, etc. corresponding to the text content. For example. For example, assuming that being unable to access a call corresponds to ERROR_NUM_159, the first-level operation and maintenance can input ERROR_NUM_159, or can input being unable to access a call.

[0060] The data obtained by the server can also be data obtained through active monitoring. For example, the server can regularly scan the values of corresponding data items of users according to the data items stored in its database, and determine whether there is an abnormality in the data by analyzing the data values. If there is an abnormality, a first data set is constructed based on the data items with abnormalities. For continuous data (such as a user's monthly bill, user traffic balance), since it generally conforms to a normal distribution, single Gaussian attribute outlier detection can be used to identify abnormalities. For example, for a certain data item of a user, the corresponding values in a recent period are obtained for fitting to generate a mean value and a standard deviation σ. If the values of this data item in a recent period all fall within the range of 2σ, it is considered that this data item has no abnormality; otherwise, it is considered that this data item has an abnormality. It can also be considered to set a value range. If the values in a recent period all fall within the set value range, it is considered that this data item has no abnormality; otherwise, it is considered that this data item has an abnormality. For discrete data, the abnormality of the data item is judged by comparing the actual value of the user with the preset value in the database. When it is detected that the user has an abnormality, the server will determine the category to which the abnormal data item belongs, and determine whether it is data related to the problem source or data related to the abnormal manifestation, and construct a first data set according to the analysis result.

[0061] Optionally, the format of the binary data set {data related to the problem source, data related to the abnormal manifestation} constructed from the data related to the problem source and the data related to the abnormal manifestation can be further standardized. For example, it can be unified as {data related to the problem source, data related to the abnormal manifestation, problem overview, verification process}, such as {{operate to order the 219 service of the family affection tariff, order the call reminder tariff}, {the user is deducted for calling a family affection number, the user information is abnormally changed, the user cannot use the SMS function}}.

[0062] S202. Respectively determine the similarity between the elements in a plurality of second data sets pre-stored in the first database and the elements in the first data set.

[0063] Among them, any second data set includes at least one item of data related to the problem source and at least one item of data related to the abnormal manifestation.

[0064] Respectively determine the similarity between the elements in a plurality of second data sets pre-stored in the first database and the elements in the first data set, that is, match each second data set with the first data set respectively.

[0065] Optionally, the similarity can be the Jaccard similarity coefficient.

[0066] Exemplarily, the three second data sets pre-stored in the first database are {A1, B1}, {A2, B2}, and {A3, B3} respectively; the first data set is {A, B}, and then the Jaccard similarity coefficients between {A, B} and {A1, B1}, {A2, B2}, and {A3, B3} are calculated respectively.

[0067] S203. Determine the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored second data set and the processing strategy, and the similarity.

[0068] The higher the similarity, the closer the first data set is to the second data set, and correspondingly, the processing strategy corresponding to the second data set is more applicable to the first data set.

[0069] It is also possible to compare the similarity with a preset threshold. If the similarity between the first data set and any second data set is less than the preset threshold, it is considered that no appropriate processing strategy is matched, and then it can be transferred to manual processing; if there is at least one similarity between the first data set and any second data set that is greater than or equal to the preset threshold, the processing strategy corresponding to the first data set is determined from the processing strategies corresponding to the second data sets whose similarity is greater than or equal to the preset threshold.

[0070] The first data set is actually used to represent that there is a problem with the user data, and the target processing strategy corresponding to the first data set is the method and strategy for solving the problem corresponding to the first data set.

[0071] The method for determining the data processing strategy provided by this application constructs a first data set including at least one item of data related to the problem source, or including at least one item of data related to the problem source and at least one item of data related to the abnormal manifestation according to the acquired data; respectively determine the similarity between the elements in the multiple second data sets pre-stored in the first database and the elements in the first data set, and any second data set includes at least one item of data related to the problem source and at least one item of data related to the abnormal manifestation; determine the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored second data set and the processing strategy, and the similarity, which can reduce manual participation, improve the determination efficiency of the processing strategy, and improve the reliability and accuracy of the determined processing strategy.

[0072] Figure 3 This is another flow schematic diagram of the method for determining the data processing strategy provided by this application. This method is applied to a server, as Figure 3 shown, and this method includes:

[0073] S301. Construct a first data set according to the acquired data.

[0074] Among them, the first data set includes at least one piece of data related to the problem source, or includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal performance.

[0075] The data related to the problem source includes one or more of the following data: user bill data, service data subscribed by the user. The data related to the abnormal performance includes one or more of the following data: communication abnormal data, tariff abnormal data, abnormal change of user information.

[0076] In a possible implementation manner, according to the obtained data, a first data set is constructed, including: constructing the first data set according to the data in the obtained user input data that is consistent with the preset element type.

[0077] For example, the preset element types are data related to the problem source and data related to the abnormal performance. The user input data includes data related to the problem source, data related to the abnormal performance, data related to the problem overview, and data related to the verification process. Then, by analyzing the data related to the problem source and the data related to the abnormal performance in the user input data, the first data set is constructed.

[0078] For example, a first-level operation and maintenance personnel found that the user operated service 67 (changing the user's product and tariff), resulting in abnormal billing for the user, but did not know how to fix it and chose to escalate the problem. The data related to the problem source submitted by the first-level operation and maintenance personnel is: A_U_DT (user tariff data table), TRADE_TYPE_67 (subscribing to the VRD platform tariff service for No. 67); the data related to the abnormal performance is: A_ALOG (user monthly bill data table), MONEY (amount).

[0079] Another example, a first-level operation and maintenance personnel found that the user operated service 159 (changing the user's product), resulting in a problem with another product of the user, but did not know how to fix it and chose to escalate the problem. The data related to the problem source submitted by the first-level operation and maintenance personnel is: A_U_P (user product data table), TRADE_TYPE_159 (subscribing to the product service that is refreshed monthly for No. 159), P_ID_9305 (product 9305). The data related to the abnormal performance is: A_U_P (user product data table), P_ID_9142 (product 9142).

[0080] In another possible implementation, a first data set is constructed based on the acquired data, including: obtaining at least one third data corresponding to any second data pre-stored in a second database within a first preset time period, where the second data is data related to the problem source or data related to abnormal manifestations; determining whether there is abnormal data among at least one third data according to the values of each third data, and if so, constructing the first data set with the abnormal data as elements.

[0081] Optionally, determining whether there is abnormal data among at least one third data according to the values of each third data includes: for any third data, calculating the standard deviation corresponding to the third data respectively according to the value of the third data; determining the value range according to the standard deviation; if there is at least one value among the multiple values corresponding to the third data that does not belong to the value range, it is determined that the third data is abnormal.

[0082] Exemplarily, the second database can be an exception tracking pool for storing data items with abnormal values. Suppose the second database stores "monthly phone bill of the user", then obtain the monthly phone bill situation of the user in the recent 10 months, determine the standard deviation σ corresponding to the user's monthly phone bill in 10 months by analyzing the data, and determine 2σ as the value range. Suppose there is a monthly phone bill of the user in the recent 10 months that does not fall within the value range of 2σ, then it is considered that the monthly phone bill of the user is abnormal.

[0083] Through this method, it is possible to actively monitor user data, implement dynamic auditing of users, capture abnormal data in a timely manner, determine the processing strategy in a timely manner, improve the auditing efficiency and the processing efficiency of abnormal data, better solve the problems of telecom users, and enhance the user experience.

[0084] For example, the system automatically monitors that the minimum consumption data of a user has become abnormal recently, and audits that the user has performed 110 service operations to change the number status recently. At this time, the system initiates a problem, and the data related to the problem source is: TRADE_TYPE_110 (110th user applies for disconnection service), A_U_S (user number data table). The data related to the abnormal manifestation is: A_U_LCS (user minimum consumption data table).

[0085] Optionally, after determining that the third data is abnormal, it further includes: sending an alarm message.

[0086] S302. Determine the first similarity between the data related to the problem source in each pre-stored second data set and the data related to the problem source in the first data set.

[0087] S303. Determine the second similarity between the data related to the abnormal manifestation in each pre-stored second data set and the data related to the abnormal manifestation in the first data set.

[0088] The first similarity and the second similarity can be Jaccard similarity coefficients.

[0089] Optionally, before calculating the similarity, meaningless items can be filtered first, that is, data items that have little impact on determining the processing strategy. The meaningless items are data items with a high occurrence frequency.

[0090] Through this method, the similarity between the data related to the problem source in the first data set and the data related to the problem source in any second data set, and the similarity between the data related to the abnormal manifestation in the first data set and the data related to the abnormal manifestation in any second data set can be determined respectively, so as to more accurately determine the similarity between the first data set and any second data set.

[0091] S304. Determine the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored second data set and the processing strategy, and each first similarity and each second similarity.

[0092] In a possible implementation manner, determining the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored second data set and the processing strategy, and each first similarity and each second similarity includes: comparing each first similarity and each second similarity with a first preset threshold respectively, and determining the second data set corresponding to when the first similarity is greater than or equal to the first preset threshold, or the second data set corresponding to when the second similarity is greater than or equal to the first preset threshold as the first preselected data set; counting the number of the first preselected data sets; if the number of the first preselected data sets is greater than or equal to a second preset threshold, determining the target data set from the first preselected data sets; and determining the target processing strategy corresponding to the first data set according to the corresponding relationship between each target data set and the processing strategy pre-stored.

[0093] Continuing to refer to the above example, the similarity between A and A1 is 0.5; the similarity between A and A2 is 0.25; the similarity between A and A3 is 1; the similarity between B and B1 is 1; the similarity between B and B2 is 0.33; the similarity between B and B3 is 1. Assuming that the first preset threshold is 0.3, then the similarities between A and A1, A and A3, B and B1, and B and B3 are greater than the first preset threshold, and the similarities between A and A2 and between B and B2 are less than the first preset threshold. Then it will be determined that the first preselected data set includes {A1, B1} and {A3, B3}. Assuming that the second preset threshold is 1, then {A1, B1} and {A3, B3} are determined as the target data set, and according to the processing strategies corresponding to {A1, B1} and {A3, B3} stored in advance, the target processing strategy corresponding to the first data set is determined.

[0094] It should be noted that both the first preset threshold and the second preset threshold can be set according to the actual situation, and this application does not make any restrictions.

[0095] Optionally, the second data set with a similarity less than the first preset threshold can also be determined as an irrelevant item.

[0096] Optionally, if each first similarity and each second similarity are both less than the first preset threshold, or if the number of the first preselected data sets is less than the second preset threshold, the process will be transferred to manual processing.

[0097] Through this method, the determination range of the target processing strategy corresponding to the first data set can be narrowed, and the processing speed can be improved.

[0098] Optionally, determining the target data set from the first preselected data set includes: determining the first preselected data set with the first similarity greater than or equal to the third preset threshold, or the second similarity greater than or equal to the third preset threshold, as the second preselected data set; counting the number of the second preselected data sets; if the number of the first preselected data sets is greater than or equal to the fourth preset threshold, then determining the second preselected data set as the target data set.

[0099] Continuing to refer to the above example, the similarity between A and A1 is 0.5; the similarity between A and A3 is 1; the similarity between B and B1 is 1; the similarity between B and B3 is 1. Assuming that the third preset threshold is 0.7, then among the two similarities corresponding to the second data set 1 {A1, B1}, one is greater than 0.7 and one is less than 0.7; the two similarities corresponding to the second data set 3 {A3, B3} are both greater than 0.7. Therefore, it is determined that the second preselected data sets include {A1, B1} and {A3, B3}. Assuming that the fourth preset threshold is 1, then {A1, B1} and {A3, B3} are determined as the target data sets.

[0100] Optionally, a second data set with at least one similarity greater than or equal to the first preset threshold and less than the third preset threshold can be determined as a general match item; a second data set with at least one similarity greater than the third preset threshold can be determined as a complete match item.

[0101] Optionally, if the number of second preselected data sets is less than the fourth preset threshold, the process is transferred to manual processing, and the second preselected data sets and the first data set are sent to the corresponding operation and maintenance personnel.

[0102] Optionally, if the number of second preselected data sets is less than the fourth preset threshold, the first data set is marked as a weak reference; if the number of second preselected data sets is greater than or equal to the fourth preset threshold, the first data set is marked as a strong reference.

[0103] Optionally, if there is no second preselected data set, the first data set is transferred to the problem to-do pool; if the number of second preselected data sets is less than the fourth preset threshold and the number of second preselected data sets is greater than or equal to 1, according to the order of the frequencies of occurrence of each processing strategy in the multiple processing strategies corresponding to each second preselected data set from high to low, the processing process is sequentially transferred to the corresponding operation and maintenance personnel until there is an operation and maintenance personnel to accept the processing process, then the second preselected data sets and the first data set are sent to the corresponding operation and maintenance personnel; if there is no operation and maintenance personnel to accept the processing process, the first data set is transferred to the problem to-do pool.

[0104] Optionally, according to the corresponding relationship between each target data set and the processing strategy stored in advance, determining the target processing strategy corresponding to the first data set includes: determining the frequency of occurrence of each processing strategy according to the corresponding relationship between each target data set and the processing strategy stored in advance; and determining the processing strategy with the highest frequency of occurrence as the target processing strategy corresponding to the first data set.

[0105] Exemplarily, the processing strategy with the highest frequency of occurrence is direct processing, and the system can automatically call the interface to process data without going through manual operation.

[0106] Optionally, after the processing is completed, the system can directly reply to the operation and maintenance personnel according to the processing situation.

[0107] Optionally, if the requirements of the direct processing solution are not met, the successfully matched processing strategy is sent to the operation and maintenance personnel for reference.

[0108] Optionally, it further includes:

[0109] S305. Obtain at least one third data set.

[0110] Wherein, the third data set is the first data set for which the corresponding target processing strategy has not changed within the second preset time period.

[0111] As above, the first data set is actually used to characterize that there are problems with the user data, and the target processing strategy corresponding to the first data set is the method and strategy for solving the problems corresponding to the first data set. When

[0112] When the processing strategy determined by the system fails to solve the existing problems or is not accepted by the operation and maintenance personnel, the user will initiate a task again for this problem and re-determine a more appropriate processing strategy.

[0113] If the target processing strategy corresponding to the first data set has not changed within the second preset time period, it can be considered that this processing strategy matches the first data set and can solve the problems corresponding to the first data set.

[0114] S306. Determine the support degree and confidence degree corresponding to any third data set according to the elements in each third data set.

[0115] S307. Determine the third data sets with the support degree greater than or equal to the fifth preset threshold and the confidence degree greater than or equal to the sixth preset threshold as the fourth data set.

[0116] Continuing to refer to the above example, assuming that the fifth preset threshold corresponding to the support degree is 0.25 and the sixth preset threshold corresponding to the confidence degree is 0.85, the data sets that meet the conditions in the above example are {A1, B1}.

[0117] In rule deduction, it is common to calculate and delete all items to obtain the high-frequency item set and then calculate. The interpretability of the relationship between the problems before and after operation and maintenance verification itself ensures the expected usability. Using this method not only improves the efficiency but also reduces the number of hidden rules.

[0118] S308. Update the second database according to the elements in the fourth data set.

[0119] In a possible implementation, updating the second database according to the elements in the fourth data set includes: for any fourth data set, determining the third similarity between the data related to the problem source in each pre-stored second data set and the data related to the problem source in each fourth data set; determining the fourth similarity between the data related to the abnormal performance in each pre-stored second data set and the data related to the abnormal performance in each fourth data set; determining the fourth data set that only has the third similarity greater than or equal to the seventh preset threshold, or the fourth data set that only has the fourth similarity greater than or equal to the seventh preset threshold as the risk data set; and updating the second database according to the elements in the risk data set.

[0120] Optionally, label the category of the fourth data set with the third similarity greater than or equal to the seventh preset threshold and the fourth similarity greater than or equal to the seventh preset threshold as 11; label the category of the fourth data set with the third similarity less than the seventh preset threshold and the fourth similarity greater than or equal to the seventh preset threshold as 01; label the category of the fourth data set with the third similarity greater than or equal to the seventh preset threshold and the fourth similarity less than the seventh preset threshold as 10; label the category of the fourth data set with the third similarity less than the seventh preset threshold and the fourth similarity less than the seventh preset threshold as 00. 00 indicates a matching failure, discard this type of fourth data set without processing. 01 and 10 indicate partial matching. 01 indicates that the data related to the abnormal performance in this data set has a relatively high occurrence frequency recently, and 10 indicates that the data related to the problem source in this data set has a relatively high occurrence frequency recently. This type of fourth data set is determined as the risk data set and needs to be tracked and monitored, and then the second database is updated according to the elements in the risk data set.

[0121] Through this method, the second database can be dynamically updated, the detected data items can be updated, and the system monitoring ability can be improved, enabling the system to follow and respond to business changes.

[0122] Optionally, updating the second database according to the elements in the risk data set includes: obtaining all the first data sets constructed within a preset time period; comparing the first element in the risk data set with the elements in each first data set, and if there is at least one first data set that includes the first element, supplement it into the second database.

[0123] Exemplarily, taking 3 days as a cycle, at the end of each cycle, the risk data set will be compared with all the first data sets constructed within the cycle. If the first element in the risk data set does not appear in the first data set for 3 consecutive cycles, it is considered a sudden problem and the first element is deleted; if there is at least one first data set that includes the first element, the first element is retained and supplemented into the second database.

[0124] Through this method, it is possible to identify whether the elements in the risk data set are related to frequently occurring problems or to sudden problems, and it is possible to avoid over - attention to sudden problems, reduce redundant data, optimize data processing, and relieve the burden on operation and maintenance personnel.

[0125] Based on the above - mentioned embodiments, the method for determining the data - processing strategy provided by this application further includes: determining the first similarity between the data related to the problem source in each pre - stored second data set and the data related to the problem source in the first subset; determining the second similarity between the data related to the abnormal performance in each pre - stored second data set and the data related to the abnormal performance in the second subset; determining the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre - stored second data set and the processing strategy, as well as each first similarity and each second similarity, which can improve the accuracy of the determined target processing strategy and the matching degree with the first data set. Further, by obtaining at least one third data set, where the third data set is the first data set for which the corresponding target processing strategy has not changed within the second preset time period; determining the support degree and confidence degree corresponding to any third data set according to the elements in each third data set; determining the third data sets with the support degree greater than or equal to the fifth preset threshold and the confidence degree greater than or equal to the sixth preset threshold as the fourth data set; and updating the second database according to the elements in the fourth data set, it is also possible to dynamically update the system, enabling the system to audit user data in a timely manner following business changes, and improving the data - auditing effect and the abnormal - data processing effect.

[0126] Figure 4 As shown in the following figure, it is a structural schematic diagram of a device for determining the data - processing strategy provided by this application. Figure 4 As shown, the device includes:

[0127] An obtaining module 41, configured to construct a first data set according to the obtained data. The first data set includes at least one item of data related to the problem source, or includes at least one item of data related to the problem source and at least one item of data related to the abnormal performance. The data related to the problem source includes one or more of the following types of data: user bill data, service data subscribed by the user. The data related to the abnormal performance includes one or more of the following types of data: communication - anomaly data, tariff - anomaly data, abnormal change of user information.

[0128] A processing module 42 is configured to respectively determine the similarity between the elements in a plurality of second data sets pre-stored in the first database and the elements in the first data set. Any second data set includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal performance.

[0129] The processing module 42 is further configured to determine the target processing strategy corresponding to the first data set according to the corresponding relationship between each pre-stored second data set and the processing strategy and the similarity.

[0130] Optionally, the obtaining module 41 is specifically configured to obtain at least one third data corresponding to any second data pre-stored in the second database within a first preset time period. The second data is data related to the problem source or data related to the abnormal performance. According to the values of each third data, it is determined whether there is abnormal data in at least one third data. If so, the abnormal data is used as an element to construct the first data set.

[0131] Optionally, the processing module 42 is specifically configured to determine the first similarity between the data related to the problem source in each pre-stored second data set and the data related to the problem source in the first data set; determine the second similarity between the data related to the abnormal performance in each pre-stored second data set and the data related to the abnormal performance in the first data set; according to the corresponding relationship between each pre-stored second data set and the processing strategy, and each first similarity and each second similarity, determine the target processing strategy corresponding to the first data set.

[0132] Optionally, the processing module 42 is specifically configured to compare each first similarity and each second similarity with a first preset threshold respectively, and determine the second data set corresponding to the case where the first similarity is greater than or equal to the first preset threshold, or the second data set corresponding to the case where the second similarity is greater than or equal to the first preset threshold as the first preselected data set; count the number of the first preselected data sets; if the number of the first preselected data sets is greater than or equal to a second preset threshold, determine the target data set from the first preselected data sets; according to the corresponding relationship between each pre-stored target data set and the processing strategy, determine the target processing strategy corresponding to the first data set.

[0133] Optionally, the processing module 42 is specifically configured to determine the first preselected data set with the first similarity greater than or equal to a third preset threshold, or the second similarity greater than or equal to the third preset threshold as the second preselected data set; count the number of the second preselected data sets; if the number of the first preselected data sets is greater than or equal to a fourth preset threshold, determine the second preselected data set as the target data set.

[0134] Optionally, the processing module 42 is specifically configured to determine the frequency of occurrence of each processing policy according to the corresponding relationship between each pre-stored target data set and the processing policy; and determine the target processing policy corresponding to the first data set as the processing policy with the highest frequency of occurrence.

[0135] Optionally, the processing module 42 is further configured to obtain at least one third data set, where the third data set is the first data set for which the corresponding target processing policy has not changed within the second preset duration; determine the support degree and confidence degree corresponding to any one of the third data sets according to the elements in each third data set; determine the third data set with the support degree greater than or equal to the fifth preset threshold and the confidence degree greater than or equal to the sixth preset threshold as the fourth data set; and update the second database according to the elements in the fourth data set.

[0136] Optionally, the processing module 42 is specifically configured to, for any one of the fourth data sets: determine the third similarity between the third subset included in the fourth data set and each second data set according to the elements in the fourth data set and the elements in each second data set; determine the fourth similarity between the fourth subset included in the fourth data set and each second data set according to the elements in the fourth data set and the elements in each second data set; determine the data set with only the third similarity greater than or equal to the seventh preset threshold, or the data set with only the fourth similarity greater than or equal to the seventh preset threshold as the risk data set; and update the second database according to the elements in the risk data set.

[0137] Optionally, the processing module 42 is specifically configured to obtain a new third data set generated within the third preset duration; determine whether the first element exists in the new third data set for the first element in the risk data set; and if the first element exists, supplement the first element into the second database.

[0138] Optionally, the obtaining module 41 is specifically configured to construct a first data set according to the data consistent with the preset element type in the obtained user input data.

[0139] Optionally, the obtaining module 41 is specifically configured to, for any one of the third data, calculate the standard deviation corresponding to the third data according to the value of the third data; determine the value range according to the standard deviation; and if at least one of the multiple values corresponding to the third data does not belong to the value range, determine that the third data is abnormal.

[0140] The apparatus for determining the data processing policy can execute the above method for determining the data processing policy, and the content and effects thereof can be found in the method embodiment part, which will not be elaborated herein.

[0141] Figure 5A schematic structural diagram of the electronic device provided by this application is as follows Figure 5 As shown, the electronic device includes: a processor 51 and a memory 52; the processor 51 is communicatively connected to the memory 52. The memory 52 is used to store computer programs. The processor 51 is used to call the computer programs stored in the memory 52 to implement the methods in the above method embodiments.

[0142] Optionally, the electronic device further includes: a transceiver 53, which is used to communicate with other devices.

[0143] The electronic device can execute the method for determining the above data processing strategy, and its content and effects can be referred to the method embodiment part, which will not be elaborated here.

[0144] This application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the method for determining the above data processing strategy.

[0145] When the computer-executable instructions stored in the computer-readable storage medium are executed by a processor, they can implement the method for determining the above data processing strategy, and its content and effects can be referred to the method embodiment part, which will not be elaborated here.

[0146] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily think of other embodiments of this application. This application is intended to cover any variations, uses, or adaptations of this application, which follow the general principles of this application and include the common general knowledge or conventional technical means in the technical field not disclosed in this application. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of this application are pointed out by the following claims.

[0147] It should be understood that this application is not limited to the exact structure already described and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is only limited by the appended claims.

Claims

1. A method for determining a data processing strategy, characterized in that Including: Construct a first data set according to the obtained data. The first data set includes at least one piece of data related to the problem source, or includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal performance. The data related to the problem source includes one or more of the following data: user bill data, service data ordered by the user. The data related to the abnormal performance includes one or more of the following data: communication abnormal data, tariff abnormal data, abnormal change of user profile; Respectively determine the similarity between the elements in a plurality of second data sets pre-stored in the first database and the elements in the first data set. Any one of the second data sets includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal performance; According to the corresponding relationship between each of the pre-stored second data sets and the processing strategy, and the similarity, determine the target processing strategy corresponding to the first data set; The constructing a first data set according to the obtained data includes: Obtain at least one third data corresponding to any one of the second data pre-stored in the second database within a first preset time period. The second data is the data related to the problem source or the data related to the abnormal performance; According to the values of each of the third data, determine whether there is abnormal data in the at least one third data. If so, construct the first data set with the abnormal data as elements; Obtain at least one third data set. The third data set is the first data set whose corresponding target processing strategy has not changed within a second preset time period; According to the elements in each of the third data sets, determine the support degree and confidence degree corresponding to any one of the third data sets; Determine the third data set with the support degree greater than or equal to the fifth preset threshold and the confidence degree greater than or equal to the sixth preset threshold as the fourth data set; Update the second database according to the elements in the fourth data set.

2. The method according to claim 1, wherein The respectively determining the similarity between the elements in a plurality of second data sets pre-stored in the first database and the elements in the first data set includes: Determine the first similarity between the data related to the problem source in each of the pre-stored second data sets and the data related to the problem source in the first data set; Determine the second similarity between the data related to the abnormal performance in each of the pre-stored second data sets and the data related to the abnormal performance in the first data set; Correspondingly, the determining the target processing strategy corresponding to the first data set according to the corresponding relationship between each of the pre-stored second data sets and the processing strategy, and the similarity includes: According to the corresponding relationship between each of the pre-stored second data sets and the processing strategy, and each of the first similarities and each of the second similarities, determine the target processing strategy corresponding to the first data set.

3. The method according to claim 2, wherein Determining the target processing policy corresponding to the first data set according to the pre-stored corresponding relationships between the second data sets and the processing policies, as well as the first similarities and the second similarities, includes: Comparing each of the first similarities and each of the second similarities with a first preset threshold respectively, and determining, as the first preselected data set, the second data set corresponding to when the first similarity is greater than or equal to the first preset threshold, or the second data set corresponding to when the second similarity is greater than or equal to the first preset threshold; Counting the number of the first preselected data sets; If the number of the first preselected data sets is greater than or equal to a second preset threshold, determining a target data set from the first preselected data sets; Determining the target processing policy corresponding to the first data set according to the pre-stored corresponding relationships between the target data sets and the processing policies; 4. The method according to claim 3, wherein The determining a target data set from the first preselected data sets includes: Determining, as the second preselected data set, the first preselected data set for which the first similarity is greater than or equal to a third preset threshold, or the second similarity is greater than or equal to the third preset threshold; Counting the number of the second preselected data sets; If the number of the first preselected data sets is greater than or equal to a fourth preset threshold, determining the second preselected data set as the target data set; 5. The method according to claim 3 or 4, characterized in that, The determining the target processing policy corresponding to the first data set according to the pre-stored corresponding relationships between the target data sets and the processing policies includes: Determining the occurrence frequency of each of the processing policies according to the pre-stored corresponding relationships between the target data sets and the processing policies; Determining the processing policy with the highest occurrence frequency as the target processing policy corresponding to the first data set; 6. The method according to claim 1, wherein The updating the second database according to the elements in the fourth data set includes: For any one of the fourth data sets: Determining the third similarity between the third subset included in the fourth data set and each of the second data sets according to the elements in the fourth data set and the elements in each of the second data sets; Determining the fourth similarity between the fourth subset included in the fourth data set and each of the second data sets according to the elements in the fourth data set and the elements in each of the second data sets; Determining, as the risk data set, the fourth data set in which only the third similarity is greater than or equal to a seventh preset threshold, or the fourth data set in which only the fourth similarity is greater than or equal to the seventh preset threshold; Updating the second database according to the elements in the risk data set; 7. The method according to claim 6, wherein The updating the second database according to the elements in the risk data set includes: Obtaining a new third data set generated within a third preset duration; Determining whether the new third data set has the first element for the first element in the risk data set; If the first element exists, supplementing the first element into the second database; 8. The method according to claim 1, wherein Construct a first data set according to the acquired data, including: Construct the first data set according to the data in the acquired user input that is consistent with the preset element type.

9. The method according to any one of claims 2-4, characterized in that, Determine whether there is abnormal data in the at least one third data according to the values of the respective third data, including: For any one of the third data, calculate the standard deviation corresponding to the third data according to the value of the third data. Determine the value range according to the standard deviation. If there is at least one value among the multiple values corresponding to the third data that does not belong to the value range, determine that the third data is abnormal.

10. An apparatus for determining a data processing strategy, characterized in that, Include: An acquisition module, configured to construct a first data set according to the acquired data. The first data set includes at least one piece of data related to the problem source, or includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal manifestation. The data related to the problem source includes one or more of the following types of data: user bill data, service data subscribed by the user. The data related to the abnormal manifestation includes one or more of the following types of data: communication abnormal data, tariff abnormal data, abnormal change of user information. A processing module, configured to respectively determine the similarity between the elements in a plurality of second data sets prestored in the first database and the elements in the first data set. Any one of the second data sets includes at least one piece of data related to the problem source and at least one piece of data related to the abnormal manifestation. The processing module is configured to determine the target processing strategy corresponding to the first data set according to the corresponding relationship between each of the prestored second data sets and the processing strategy and the similarity. The acquisition module is specifically configured to acquire at least one third data corresponding to any one of the second data prestored in the second database within a first preset time period. The second data is the data related to the problem source or the data related to the abnormal manifestation. Determine whether there is abnormal data in the at least one third data according to the values of the respective third data. If so, construct the first data set with the abnormal data as an element. Acquire at least one third data set, where the third data set is the first data set for which the corresponding target processing strategy has not changed within a second preset time period. Determine the support degree and confidence degree corresponding to any one of the third data sets according to the elements in each of the third data sets. Determine the third data set with the support degree greater than or equal to a fifth preset threshold and the confidence degree greater than or equal to a sixth preset threshold as the fourth data set. Update the second database according to the elements in the fourth data set.

11. A control device, characterized in that, Include: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor. The instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, and when the processor executes the computer-executable instructions, the method described in any one of claims 1-9 is implemented.

13. A computer program product, comprising a computer program which, when executed by a processor, implements the method described in any one of claims 1-9.

Images