Method and device for monitoring reduction, electronic device and computer-readable storage medium

Abstract

The embodiments of the present invention provide a reduction monitoring method and device, an electronic device, and a computer-readable storage medium, wherein the method comprises: performing feature extraction processing on the business data related to the reduction of the target object to obtain the feature data of the target object, wherein the feature data of the target object includes the ratio of the reduction data to the income data; performing classification processing on the feature data of the target object through a classification model obtained based on clustering processing to obtain the type of the target object; comparing the obtained type of the target object with the stored type of the target object, and determining that the reduction of the target object is abnormal if the two are inconsistent. The embodiments of the present invention can promptly discover abnormal problems in the reduction data, ensure the consistency and accuracy of the reduction data, and avoid the occurrence of settlement losses and complaint escalation events.

Description

Technical Field

[0001] The present invention relates to the field of Internet technology, and in particular to a reduction monitoring method and a reduction monitoring device, an electronic device and a computer-readable storage medium. Background Art

[0002] Settlement reduction is an extremely important part of comprehensive settlement. Settlement data is usually receivable data. In order to avoid settlement losses, it is necessary to use reduction documents to make up for the settlement differences in receivable settlement. Refunds from users, arrears in write-offs, anomalies in the billing process of the Business & Operation Support System (BOSS), or failure to issue a list due to business audits, etc., will all generate reduction documents.

[0003] Due to the influence of unknown factors such as system failure, program logic, background maintenance and modification, etc., there may be deviations in the generation of reduction data. For example, the reduction of DreamNet business is complicated from data access, reduction generation, reduction error detection to reduction upload. Abnormal conditions in a certain link or normal business penalties may cause changes in reduction data, resulting in deviations in reduction data.

[0004] Usually, reduction documents are generated later than income and settlement reports, and it is not easy to discover abnormalities in the reduction data in advance or in time. If abnormalities in the reduction data cannot be discovered in time before the reduction documents are uploaded, the consistency and accuracy of the settlement cannot be guaranteed, resulting in settlement losses and even triggering escalated complaints from partners. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide a reduction monitoring method and a reduction monitoring device, an electronic device and a computer-readable storage medium, which can promptly discover abnormal problems in the reduction data, ensure the consistency and accuracy of the reduction data, and avoid settlement losses and complaint escalation events.

[0006] In order to solve the above technical problems, the embodiment of the present invention is implemented as follows:

[0007] In a first aspect, an embodiment of the present invention provides a method for monitoring reduction, characterized by comprising:

[0008] Performing feature extraction processing on the business data related to the reduction of the target object to obtain feature data of the target object, wherein the feature data of the target object includes a ratio of the reduction data to the income data;

[0009] Classify the characteristic data of the target object by using a classification model obtained based on clustering processing to obtain the type of the target object;

[0010] The obtained type of the target object is compared with the stored type of the target object, and when the two are inconsistent, it is determined that there is an abnormality in the reduction of the target object.

[0011] In a second aspect, an embodiment of the present invention provides a reduction monitoring device, characterized in that it includes:

[0012] A processing module, used to perform feature extraction processing on the business data related to the reduction of the target object to obtain feature data of the target object, wherein the feature data of the target object includes a ratio of the reduction data to the income data;

[0013] A classification module, used to classify the characteristic data of the target object by using a classification model obtained based on clustering processing to obtain the type of the target object;

[0014] The monitoring module is used to compare the obtained type of the target object with the stored type of the target object, and determine that there is an abnormality in the reduction of the target object when the two are inconsistent.

[0015] In a third aspect, an embodiment of the present invention provides an electronic device, characterized in that it includes a processor, a communication interface, a memory and a communication bus; wherein the processor, the communication interface and the memory communicate with each other through the bus; the memory is used to store computer programs; and the processor is used to execute the programs stored in the memory to implement the reduction monitoring method as described in the first aspect.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the reduction monitoring method as described in the first aspect is implemented.

[0017] The reduction monitoring method and reduction monitoring device, electronic device and computer-readable storage medium provided by the embodiments of the present invention perform feature extraction processing on the business data related to the reduction of the target object to obtain the feature data of the target object, wherein the feature data of the target object includes the ratio of the reduction data to the income data, classifies the feature data of the target object through the classification model obtained based on clustering processing to obtain the type of the target object, compares the obtained type of the target object with the stored type of the target object, and determines that the reduction of the target object is abnormal when the two are inconsistent; the reduction amount of the target object has a relatively stable characteristic within a certain period of time, combined with the classification model established based on clustering processing, classifies the target object according to the business data related to the reduction of the target object, and analyzes the rationality of the reduction of the target object according to whether the type of the target object changes, so as to timely discover the reduction abnormality caused by unknown factors such as system failure, program logic, and background maintenance modification, ensure the consistency and accuracy of the reduction data, ensure that the reduction data is uploaded accurately and on time, provide reliable guarantee for accurate settlement, improve the efficiency of settlement maintenance to a certain extent, reduce the settlement maintenance cost, avoid post-adjustment, and reduce the risk of complaints from partners. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0019] Figure 1 A flowchart of an implementation method of the reduction monitoring method according to an embodiment of the present invention;

[0020] Figure 2 A schematic diagram of a flow chart of an implementation method for classifying target objects according to an embodiment of the present invention;

[0021] Figure 3 A schematic diagram of a process for implementing a method of obtaining a classification model through clustering processing according to an embodiment of the present invention;

[0022] Figure 4 A schematic diagram of an implementation method of obtaining a classification model by dividing a cluster tree through classification lines;

[0023] Figure 5 A flowchart of an implementation method for determining a reduction exception according to the type of a target object according to an embodiment of the present invention;

[0024] Figure 6A schematic diagram of the structure of a reduction monitoring device according to an embodiment of the present invention;

[0025] Figure 7 It is another schematic diagram of the composition structure of the reduction monitoring device according to an embodiment of the present invention;

[0026] Figure 8 The present invention is a schematic diagram of the structure of an electronic device. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0028] Figure 1 1 is a flow chart of an implementation method of the reduction monitoring method according to an embodiment of the present invention. Figure 1 The method in the embodiment can be executed by a reduction monitoring device as an execution subject. The reduction monitoring device can be applied to an operator's server, for example, a server of the Next Generation Business Operation Support System (NGBOSS). The server can be a physical server including an independent host, or a virtual server carried by a host cluster, or a cloud server. The embodiment of the present invention does not limit this. Figure 1 As shown, the method at least includes:

[0029] S102, performing feature extraction processing on the business data related to the reduction of the target object to obtain feature data of the target object, wherein the feature data of the target object includes the ratio of the reduction data to the income data.

[0030] In an embodiment of the present invention, the target object may be a service provider (Service Provider, referred to as SP) of a mobile network information service business. For example, the target object may be represented by an SP enterprise code. The target object may provide one of the following types of services to mobile phone users through a mobile communication network: video, animation, games, reading, etc. The embodiment of the present invention does not limit the type of service provided by the target object. Reduction-related business data may include reduction-related business data directly obtained from original data, such as reduction data and revenue data, etc., and may also include reduction-related business data calculated from reduction-related business data directly obtained from original data, such as the ratio of reduction data to revenue data obtained by calculating reduction data and revenue data, etc. The embodiment of the present invention does not limit the type of reduction-related business data.

[0031] Optionally, the original business data related to the reduction of the target object can be cleaned, aggregated, and reduced in dimension to obtain standardized business data related to the reduction of the target object, and then feature extraction processing can be performed on the standardized business data related to the reduction of the target object, and business data including the ratio of the reduction data to the income data can be selected as the feature data of the target object. For example, the feature data of the target object can be represented in the form of a feature vector. The embodiment of the present invention does not limit the type of feature data of the target object and the implementation method of obtaining the feature data of the target object. Optionally, the target object can have only one type of reduction data, and the feature data of the target object can include the ratio of the reduction data of this type to the income data; or, the target object can also have more than two types of reduction data, and the feature data of the target object can include the ratio of each type of reduction data to the income data. For example, the type of reduction data can include refund reduction data, arrears reduction data, etc., and the feature data of the target object can include the ratio of refund reduction data to income data, the ratio of arrears reduction data to income data, etc.

[0032] S104, classifying the feature data of the target object by using the classification model obtained based on the clustering process to obtain the type of the target object.

[0033] In an embodiment of the present invention, the classification model obtained based on clustering processing may be a classification model obtained by clustering samples in advance. The samples may be divided into different types by clustering the samples according to preset rules, so that the similarity of samples belonging to the same type is as large as possible, and the difference of samples belonging to different types is as large as possible, that is, the samples belonging to the same type are clustered together as much as possible, and the samples belonging to different types are separated as much as possible through clustering processing. For example, the preset rule may be the distance or similarity between the characteristic data of the samples. The embodiment of the present invention does not limit the implementation method of the classification model obtained based on clustering processing. Among them, the sample may be a service provider of mobile network information service business. For example, the sample may be represented by an SP enterprise code. The sample may provide mobile phone users with services of the same or different business types through the mobile communication network. For example, the business types may include video, animation, games, reading, etc. The embodiment of the present invention does not limit the business types of the services provided by the sample. The characteristic data of the sample may include the proportion of reduction data and income data.

[0034] Optionally, the clustering process of obtaining the classification model may adopt at least one of the following clustering algorithms: hierarchical clustering algorithm, partitioning clustering algorithm, density-based clustering algorithm, grid-based clustering algorithm and model-based clustering algorithm, etc. The embodiment of the present invention does not limit the type of clustering algorithm used for clustering process. Optionally, the number of types of classification models obtained based on clustering process can be reasonably set according to demand, and the embodiment of the present invention does not limit this. For example, when the number of types of classification models is large, the accuracy of target object reduction monitoring can be improved. Optionally, when determining the type of the target object, the similarity between the target object and each type of classification model can be determined according to the feature data of the target object and the centroid data of each type of classification model according to preset rules, and the type of the classification model with the greatest similarity to the target object is used as the type of the target object. The embodiment of the present invention does not limit the implementation method of determining the type of the target object through the classification model obtained based on clustering process.

[0035] S106, comparing the obtained type of the target object with the stored type of the target object, and determining that there is an abnormality in the reduction of the target object when the two are inconsistent.

[0036] Since the reduction amount can indirectly reflect the user's satisfaction with the service business provided by a target object, it is usually relatively stable within a certain period of time. Therefore, under normal circumstances, the reduction situation of the same target object remains stable within a certain range within a certain period of time, and there will be no large fluctuations. In an embodiment of the present invention, the type of target object obtained by classifying the business data related to the reduction of the target object will not change. Based on this characteristic of the reduction of the target object, it is possible to determine whether there is an abnormality in the reduction of the target object by judging whether the type of the target object obtained by classifying the business data related to the reduction of the target object has changed.

[0037] In an embodiment of the present invention, the type of the target object currently obtained can be compared with the type of the target object previously stored to determine whether the type of the target object has changed. If the type of the target object currently obtained is inconsistent with the type of the target object previously stored, that is, the type of the target object has changed, then it is indicated that the current reduction of the target object may be abnormal. If the type of the target object currently obtained is consistent with the type of the target object previously stored, that is, the type of the target object has not changed, then it is indicated that the current reduction of the target object is normal. Among them, the type of the target object previously stored can be the type of the target object obtained by the classification model last time the classification model classified the target object, or it can also be the type of the target object obtained by the classification model at a previous time when the classification model classified the target object. The embodiment of the present invention does not limit this. For example, the classification model can classify the target object according to the periodicity of the settlement account period.

[0038] The reduction monitoring method of the embodiment of the present invention performs feature extraction processing on the business data related to the reduction of the target object to obtain the feature data of the target object, wherein the feature data of the target object includes the ratio of the reduction data to the income data, and classifies the feature data of the target object through the classification model obtained based on clustering processing to obtain the type of the target object, and compares the obtained type of the target object with the stored type of the target object, and determines that the reduction of the target object is abnormal when the two are inconsistent; using the characteristic that the reduction amount of the target object is relatively stable within a certain period of time, combined with the classification model established based on clustering processing, the target object is classified according to the business data related to the reduction of the target object, and the rationality of the reduction of the target object is analyzed according to whether the type of the target object changes, so as to timely discover the reduction abnormality caused by unknown factors such as system failure, program logic, and background maintenance modification, ensure the consistency and accuracy of the reduction data, ensure that the reduction data is uploaded accurately and on time, provide reliable guarantee for accurate settlement, improve the efficiency of settlement maintenance to a certain extent, reduce the settlement maintenance cost, avoid post-adjustment, and reduce the risk of complaints from partners.

[0039] The following will be combined Figures 2 to 5 The embodiment of the present invention is described in detail the reduction monitoring method.

[0040] Figure 2 FIG. 1 is a flow chart of an implementation method of classifying target objects according to an embodiment of the present invention. Figure 2 As shown, the characteristic data of the target object is classified by the classification model obtained based on clustering processing to obtain the type of the target object, which at least includes:

[0041] S202, determining the distance between the feature data of the target object and the centroid data of each type of the classification model obtained based on clustering processing.

[0042] In an embodiment of the present invention, the distance between the target object and each type of classification model can be determined based on the feature data of the target object and the centroid data of each type of classification model, for example, the Euclidean distance between the feature data of the target object and the centroid data of each type of classification model is determined. The embodiment of the present invention does not limit the implementation method of determining the distance between the target object and each type of classification model. The centroid data of each type of classification model can be determined based on the feature data of all or part of the samples contained in each type of classification model. For example, when the feature data of the sample is a feature vector, the centroid data of each type of classification model is a feature vector of the same dimension. Optionally, the centroid data of each type of classification model can be obtained by calculating the average value of the feature data of all or part of the samples contained in each type of classification model. The embodiment of the present invention does not limit the implementation method of determining the centroid data of each type of classification model.

[0043] S204: The type of the classification model corresponding to the centroid data having the smallest distance from the feature data of the target object is used as the type of the target object.

[0044] In an embodiment of the present invention, the distance between the characteristic data of the target object and the centroid data of each type of the classification model can be compared, and the type of the classification model corresponding to the centroid data with the smallest distance between the characteristic data of the target object can be determined as the type of the target object. For example, when the distance between the characteristic data of the target object and the centroid data of each type of the classification model is the Euclidean distance, the Euclidean distance between the characteristic data of the target object and the centroid data of each type of the classification model can be compared, and the type of the classification model corresponding to the centroid data with the smallest Euclidean distance between the characteristic data of the target object can be determined as the type of the target object.

[0045] This embodiment uses the characteristics of the classification model obtained through clustering processing to determine the type of the target object by determining the distance between the feature data of the target object and the centroid data of each type of the classification model. This can simplify the classification processing process of the classification model, improve the efficiency of monitoring, ensure timely discovery of abnormal problems in the reduction, and ensure that the reduction data is uploaded accurately and on time.

[0046] Figure 3 FIG. 1 is a flow chart of an implementation method of obtaining a classification model through clustering processing according to an embodiment of the present invention. Figure 3 As shown, the classification model obtained by clustering processing includes at least:

[0047] S302, clustering the characteristic data of the sample using a hierarchical clustering algorithm to obtain a clustering tree of the sample, wherein the characteristic data of the sample includes the ratio of the reduction data to the income data.

[0048] In some optional examples, the characteristic data of the sample can be obtained by performing feature extraction processing on the business data related to the sample reduction. The business data related to the sample reduction is shown in Table 1.

[0049] Table 1

[0050]

[0051] The original business data related to sample reduction can be cleaned, aggregated, and reduced in dimension to obtain standardized business data related to sample reduction. Feature extraction is then performed on the standardized business data related to sample reduction to select the ratio of reduction data to revenue data that best reflects user satisfaction as the feature data of the sample. The feature data of the sample can be represented in the form of feature vectors, for example, v1 = (b1, c1, ...), v2 = (b2, c2, ...).

[0052] The samples can be clustered according to the Euclidean distance between the characteristic data of the samples. The process is as follows:

[0053] Step 1: Take the feature vector of each sample as a class, calculate the Euclidean distance between any two classes, and determine the minimum Euclidean distance d min .

[0054] Step 2: Merge the two classes with the smallest Euclidean distance into a new class, whose feature vector is the average of the feature vectors of the two merged classes.

[0055] Step 3: Continue to calculate the Euclidean distance between the new class and any two classes in all unmerged classes, and determine the minimum Euclidean distance d min .

[0056] Step 4: Repeat steps 2 and 3 until all classes are finally merged into one class, and the clustering tree of the sample is obtained, such as Figure 4 As shown. Among them, the formula of the distance function that selects the minimum Euclidean distance as the measure of the degree of proximity is as follows:

[0057]

[0058] S304, dividing the cluster tree by classification lines to obtain sample groups, and using the sample groups as classifications of the classification model.

[0059] Continuing with the above optional example, after obtaining the cluster tree of the samples through the hierarchical clustering algorithm, the cluster tree can be divided by classification lines to obtain the grouping of the samples, such as Figure 4 As shown in Figure 2, the samples are divided into three groups by dividing the cluster tree through classification lines. These three groups are used as three categories reflecting the quality of user satisfaction, namely high, medium and low. A classification model reflecting user satisfaction is obtained. The samples contained in each category are shown in Table 2.

[0060] Table 2

[0061] Satisfaction with use sample high quality v1, v2, v6, ... Medium quality v3, v4, v5, ... Low quality ……、vn

[0062] S306, determining the centroid data of each type of the classification model according to the characteristic data of each group of samples.

[0063] According to the above optional example, after the classification model is obtained based on the hierarchical clustering algorithm, the centroid data of the high, medium and low quality categories of the classification model can be obtained by calculating the arithmetic mean of the feature data of all samples contained in the high, medium and low quality categories of the classification model respectively. H , o M , o L When classifying the target object according to its feature data v0, the formula for determining the type of the target object and the minimum distance between the target object and the centroid of the three categories of the classification model is as follows:

[0064] The type of the target object = min(d H (o H ,v0),d M (o M ,v0),d L (o L ,v0)) (Formula 2)

[0065] This embodiment adopts a hierarchical clustering algorithm to obtain a classification model. The clustering method is simple and easy to implement. Since the number of clusters is not pre-set, the classification of the classification model can be flexibly determined or adjusted according to needs, so that the obtained classification model has strong applicability. At the same time, the classification model obtained by the hierarchical clustering algorithm can also obtain the hierarchical relationship of each type of classification model, so that the classification model can meet the needs of different application scenarios.

[0066] Figure 5 FIG. 1 is a flow chart of an implementation method for determining a subtraction exception according to the type of a target object according to an embodiment of the present invention. Figure 5 As shown, the obtained type of the target object is compared with the stored type of the target object, and when the two are inconsistent, it is determined that there is an abnormality in the reduction of the target object, which at least includes:

[0067] S502: Compare the obtained type of the target object of the current billing period with the stored type of the target object of the previous billing period.

[0068] S504: When the type of the target object of the current account period is inconsistent with the type of the target object of the previous account period, it is determined that there is an abnormality in the reduction of the target object of the current account period.

[0069] Continuing with the above optional example, after the reduction is generated, by extracting the characteristic data of the target object with SP enterprise code SP6 in the current account period, and classifying the target object through the above classification model, it is obtained that the type of the target object SP6 of the current account period is high quality, and by comparing the type of the target object SP6 of the current account period with the type of the target object SP6 of the previous account period stored, the types of the target object SP6 in the current account period and in the previous account period are shown in Table 3.

[0070] Table 3

[0071] Satisfaction with use Monitoring object (current account period) Monitoring object (previous account period) high quality SP1, SP2, SP6, ... SP1, SP2, ... Medium quality SP3, SP4, SP5, ... SP3, SP4, SP5, SP6,… Low quality ……、SPn ……、SPn

[0072] It can be seen from Table 3 that the type of the target object SP6 of the previous account period is medium quality, that is, the type of the target object SP6 in the current account period and the previous account period has changed. The type of the target object SP6 of the current account period is inconsistent with the type of the target object SP6 of the previous account period. It can be determined that there may be an abnormality in the reduction of the target object SP6 of the current account period.

[0073] Optionally, after determining that there is an abnormality in the reduction of the target object in the current account period, the cause of the abnormality may be further determined based on detailed data of business data related to the reduction of the target object.

[0074] Based on the above optional example, the type of the target object SP6 changes from medium quality in the previous account period to high quality in the current account period, indicating that there may be an abnormality in the reduction of the target object SP6 in the current account period. It is necessary to further determine the cause of the abnormality based on the detailed data of the business data related to the reduction of the target object. After positioning and analysis, when the target object SP6 on-demand service is charged online, the business code in the online billing message sent by the platform to the DCC interface on the BOSS side is filled with the content code, resulting in the business code of the Convergence Billing Engine (CBE) output list being replaced with the content code, causing the final reduction data to be suspended because the corresponding data cannot be found during settlement, resulting in settlement losses. Subsequently, the business department can urge the platform to fix the problem of incorrect business code filling, and emergency intervention through dispatching orders is required before the problem is fixed.

[0075] This embodiment selects the type of the target object of the previous account period adjacent to the current account period as a comparison basis to determine whether there is an abnormality in the reduction of the target object in the current account period. Since the probability that the types of the target objects of two adjacent account periods are consistent is relatively higher, the accuracy of the judgment on the abnormality of the reduction of the target object is effectively guaranteed.

[0076] Corresponding to the method described above, based on the same technical concept, an embodiment of the present invention further provides a reduction monitoring device, Figure 6 FIG. 1 is a schematic diagram of a composition structure of a reduction monitoring device according to an embodiment of the present invention. The reduction monitoring device can be applied to a server of an operator and can be used to execute Figure 1 Describe the reduction monitoring methods, such as Figure 6 As shown, the reduction monitoring device at least includes: a processing module 610, a classification module 620 and a monitoring module 630, wherein the processing module 610, the classification module 620 and the monitoring module 630 are connected in sequence.

[0077] The processing module 610 is used to perform feature extraction processing on the business data related to the reduction of the target object to obtain feature data of the target object, wherein the feature data of the target object includes the ratio of the reduction data to the income data.

[0078] In this embodiment, the description of the processing module 610 can be found in Figure 1 The description of S102 is given in the specification, so it will not be repeated here.

[0079] A classification module 620 is used to classify the characteristic data of the target object by using a classification model obtained based on clustering processing to obtain the type of the target object;

[0080] In this embodiment, the description of the classification module 620 can be found in Figure 1The description of S104 is given in , so it will not be repeated here.

[0081] The monitoring module 630 is used to compare the obtained type of the target object with the stored type of the target object, and determine that there is an abnormality in the reduction of the target object when the two are inconsistent.

[0082] In this embodiment, the description of the monitoring module 630 can be found in Figure 1 The description of S106 is given in , so it will not be repeated here.

[0083] The reduction monitoring device of the embodiment of the present invention obtains feature data of the target object by performing feature extraction processing on the business data related to the reduction of the target object, wherein the feature data of the target object includes the ratio of the reduction data to the income data, and classifies the feature data of the target object through a classification model obtained based on clustering processing to obtain the type of the target object, and compares the obtained type of the target object with the stored type of the target object, and determines that there is an abnormality in the reduction of the target object when the two are inconsistent; utilizing the relatively stable characteristic of the reduction amount of the target object within a certain period of time, combined with the classification model established based on clustering processing, classifies the target object according to the business data related to the reduction of the target object, and analyzes the rationality of the reduction of the target object according to whether the type of the target object changes, so as to timely discover the reduction abnormality caused by unknown factors such as system failure, program logic, and background maintenance modification, ensure the consistency and accuracy of the reduction data, ensure that the reduction data is uploaded accurately and on time, provide reliable guarantee for accurate settlement, improve the efficiency of settlement maintenance to a certain extent, reduce the settlement maintenance cost, avoid post-adjustment, and reduce the risk of complaints from partners.

[0084] Optionally, the classification module 620 is used to determine the distance between the characteristic data of the target object and the centroid data of each type of classification model obtained based on clustering processing; the type of the classification model corresponding to the centroid data with the smallest distance from the characteristic data of the target object is used as the type of the target object.

[0085] Optionally, the clustering process adopts at least one of the following clustering algorithms: a hierarchical clustering algorithm, a partitioning clustering algorithm, a density-based clustering algorithm, a grid-based clustering algorithm, and a model-based clustering algorithm.

[0086] Alternatively, if Figure 7 As shown, Figure 7This is another schematic diagram of the composition structure of the reduction monitoring device of an embodiment of the present invention. In addition to the processing module 610, the classification module 620 and the monitoring module 630, the reduction monitoring device also includes a clustering module 640, which are respectively connected to the processing module 610 and the classification module 620, and is used to use a hierarchical clustering algorithm to cluster the characteristic data of the samples to obtain a clustering tree of the samples, wherein the characteristic data of the samples include the proportion of the reduction data and the income data; the clustering tree is divided by classification lines to obtain grouping of the samples, and the grouping of the samples is used as the classification of the classification model; according to the characteristic data of each group of samples, the centroid data of each type of the classification model is determined.

[0087] Optionally, the monitoring module 630 is used to compare the type of the target object of the current account period with the type of the target object of the previous account period stored; when the type of the target object of the current account period is inconsistent with the type of the target object of the previous account period, it is determined that there is an abnormality in the reduction of the target object of the current account period.

[0088] Optionally, the processing module 610 is further configured to determine the reason why the reduction of the target object is abnormal based on the detailed data of the business data related to the reduction of the target object.

[0089] Optionally, the processing module 610 is further used to clean, aggregate and reduce the dimension of the original business data related to the reduction of the target object to obtain the business data related to the reduction of the target object.

[0090] Corresponding to the method described above, based on the same technical concept, an embodiment of the present invention further provides an electronic device, Figure 8 A schematic diagram of the structure of an electronic device for implementing an embodiment of the present invention is shown in FIG. Figure 8 As shown, the electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors 801 and memory 802, and the memory 802 may store one or more storage applications or data. Among them, the memory 802 may be a short-term storage or a persistent storage. The application stored in the memory 802 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the computer device. Furthermore, the processor 801 may be configured to communicate with the memory 802 and execute a series of computer executable instructions in the memory 802 on the computer device. The computer device may also include one or more power supplies 803, one or more wired or wireless network interfaces 804, one or more input and output interfaces 805, and one or more keyboards 806.

[0091] In this embodiment, the electronic device includes a processor, a communication interface, a memory and a communication bus; wherein the processor, the communication interface and the memory communicate with each other through the bus; the memory is used to store computer programs; the processor is used to execute the programs stored in the memory to implement the following method:

[0092] Perform feature extraction processing on the business data related to the reduction of the target object to obtain feature data of the target object, wherein the feature data of the target object includes the ratio of the reduction data to the income data;

[0093] The characteristic data of the target object is classified by using the classification model obtained based on clustering processing to obtain the type of the target object;

[0094] The obtained type of the target object is compared with the stored type of the target object, and when the two are inconsistent, it is determined that there is an abnormality in the reduction of the target object.

[0095] The electronic device of the embodiment of the present invention obtains the characteristic data of the target object by performing feature extraction processing on the business data related to the reduction of the target object, wherein the characteristic data of the target object includes the ratio of the reduction data to the income data, and classifies the characteristic data of the target object by using the classification model obtained based on clustering processing to obtain the type of the target object, and compares the obtained type of the target object with the stored type of the target object, and determines that the reduction of the target object is abnormal when the two are inconsistent; using the characteristic that the reduction amount of the target object is relatively stable within a certain period of time, combined with the classification model established based on clustering processing, the target object is classified according to the business data related to the reduction of the target object, and the rationality of the reduction of the target object is analyzed according to whether the type of the target object changes, so as to timely discover the reduction abnormality caused by unknown factors such as system failure, program logic, and background maintenance modification, ensure the consistency and accuracy of the reduction data, ensure that the reduction data is uploaded accurately and on time, provide reliable guarantee for accurate settlement, improve the efficiency of settlement maintenance to a certain extent, reduce the settlement maintenance cost, avoid post-adjustment, and reduce the risk of complaints from partners.

[0096] Corresponding to the method described above, based on the same technical concept, an embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the following method is implemented:

[0097] Perform feature extraction processing on the business data related to the reduction of the target object to obtain feature data of the target object, wherein the feature data of the target object includes the ratio of the reduction data to the income data;

[0098] The characteristic data of the target object is classified by using the classification model obtained based on clustering processing to obtain the type of the target object;

[0099] The obtained type of the target object is compared with the stored type of the target object, and when the two are inconsistent, it is determined that there is an abnormality in the reduction of the target object.

[0100] The computer-readable storage medium of the embodiment of the present invention obtains feature data of the target object by performing feature extraction processing on business data related to the reduction of the target object, wherein the feature data of the target object includes the ratio of the reduction data to the income data, and classifies the feature data of the target object by using a classification model obtained based on clustering processing to obtain the type of the target object, and compares the obtained type of the target object with the stored type of the target object, and determines that there is an abnormality in the reduction of the target object when the two are inconsistent; utilizing the relatively stable characteristic of the reduction amount of the target object within a certain period of time, combined with the classification model established based on clustering processing, classifies the target object according to the business data related to the reduction of the target object, and analyzes the rationality of the reduction of the target object according to whether the type of the target object changes, so as to timely discover the reduction abnormality caused by unknown factors such as system failure, program logic, and background maintenance modification, ensure the consistency and accuracy of the reduction data, ensure that the reduction data is uploaded accurately and on time, provide reliable guarantee for accurate settlement, improve the efficiency of settlement maintenance to a certain extent, reduce the settlement maintenance cost, avoid post-adjustment, and reduce the risk of complaints from partners.

[0101] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0102] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0103] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0104] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0105] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0106] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0107] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0108] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0109] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0110] The above are only embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A method for monitoring reduction, characterized in that: include: Performing feature extraction processing on the business data related to the reduction of the target object to obtain feature data of the target object, wherein the feature data of the target object includes a ratio of the reduction data to the income data; The characteristic data of the target object is classified by a classification model obtained based on clustering processing to obtain the type of the target object; the type of the target object is divided into high quality, medium quality and low quality according to the usage satisfaction; Compare the obtained type of the target object with the stored type of the target object, and determine that there is an abnormality in the reduction of the target object if the two are inconsistent; The step of comparing the obtained type of the target object with the stored type of the target object, and determining that there is an abnormality in the reduction of the target object when the two types are inconsistent, includes: Compare the obtained type of the target object of the current account period with the stored type of the target object of the previous account period; The current account period and the previous account period are adjacent account periods; When the type of the target object in the current account period is inconsistent with the type of the target object in the previous account period, it is determined that there is an abnormality in the reduction of the target object in the current account period.

2. The method according to claim 1, characterized in that The classification model obtained based on clustering processing is used to classify the characteristic data of the target object to obtain the type of the target object, including: Determine the distance between the characteristic data of the target object and the centroid data of each type of the classification model obtained based on clustering processing; The type of the classification model corresponding to the centroid data having the smallest distance from the feature data of the target object is used as the type of the target object.

3. The method according to claim 2, characterized in that The clustering process adopts at least one of the following clustering algorithms: a hierarchical clustering algorithm, a partitioning clustering algorithm, a density-based clustering algorithm, a grid-based clustering algorithm, and a model-based clustering algorithm.

4. The method according to claim 3, characterized in that The clustering process comprises: A hierarchical clustering algorithm is used to perform clustering processing on the characteristic data of the sample to obtain a clustering tree of the sample, wherein the characteristic data of the sample includes the ratio of the reduction data to the income data; Dividing the cluster tree by classification lines to obtain groups of the samples, and using the groups of the samples as classifications of the classification model; According to the characteristic data of each group of samples, the centroid data of each type of the classification model is determined.

5. The method according to any one of claims 1 to 4, characterized in that: After determining that the reduction of the target object in the current account period is abnormal, the method further includes: The reason why the reduction of the target object is abnormal is determined based on the detailed data of the business data related to the reduction of the target object.

6. The method according to any one of claims 1 to 4, characterized in that: Before extracting features from the target object's business data related to reduction to obtain feature data of the target object, the method further includes: The original business data related to the reduction of the target object is cleaned, aggregated and dimensionally reduced to obtain the business data related to the reduction of the target object.

7. A reduction monitoring device, characterized in that: include: A processing module, used to perform feature extraction processing on the business data related to the reduction of the target object to obtain feature data of the target object, wherein the feature data of the target object includes a ratio of the reduction data to the income data; A classification module, used to classify the characteristic data of the target object through a classification model obtained based on clustering processing to obtain the type of the target object; the type of the target object is divided into high quality, medium quality and low quality according to usage satisfaction; A monitoring module, used for comparing the obtained type of the target object with the stored type of the target object, and determining that there is an abnormality in the reduction of the target object when the two types are inconsistent; In the monitoring module, the obtained type of the target object is compared with the stored type of the target object, and when the two types are inconsistent, it is determined that the reduction of the target object is abnormal, including: Compare the obtained type of the target object of the current account period with the stored type of the target object of the previous account period; The current account period and the previous account period are adjacent account periods; When the type of the target object in the current account period is inconsistent with the type of the target object in the previous account period, it is determined that there is an abnormality in the reduction of the target object in the current account period.

8. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus; wherein the processor, the communication interface and the memory communicate with each other through the bus; the memory is used to store computer programs; the processor is used to execute the programs stored in the memory to implement the reduction monitoring method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the reduction monitoring method according to any one of claims 1 to 6 is implemented.

Images