A method and apparatus for detecting transaction data

Abstract

The application discloses a kind of methods and devices for detecting transaction data, and it relates to mobile internet technology field.The specific embodiment of the method includes: it can be according to the transaction service type of transaction data to be detected and the transaction interface and transaction channel to which it belongs, from multiple preset global detection models and / or unit detection models, find one or more target detection models matched, and utilize the target detection model found to detect transaction data in real time.The embodiments of the application construct global detection model for transaction service type, and determine to construct unit detection model for the combination of different transaction interfaces and different transaction channels, to detect transaction data using multiple detection models, improve the accuracy, real-time performance and detection refinement degree of detecting abnormal data in transaction data, improve the efficiency of detecting data.

Description

Technical Field

[0001] This invention relates to the field of financial technology, and in particular to a method and apparatus for detecting transaction data. Background Technology

[0002] In the digital economy era, more and more business platforms offer a variety of transaction types and usually provide these transactions through multiple channels. For business platforms with dispersed business scenarios and large business volumes, the platform providers need to monitor transaction data associated with these transactions in real time to monitor changes and risks in the business platform's transactions.

[0003] Existing methods for detecting transaction data typically involve aggregating data from all transaction types and channels into a database, and then performing statistical analysis and detection based on the database. However, these methods suffer from poor real-time performance and limited flexibility and scalability in the detected data. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method and apparatus for detecting transaction data. This method can search for one or more matching target detection models from a variety of preset global detection models and / or unit detection models based on the transaction business type, the transaction interface, and the transaction channel to which the transaction data to be detected belongs. The embodiments of the present invention improve the accuracy and refinement of detecting abnormal data in transaction data and increase the efficiency of data detection by constructing global detection models for transaction business types and unit detection models for combinations of different transaction interfaces and transaction channels, thereby utilizing different detection models to detect transaction data.

[0005] To achieve the above objectives, according to one aspect of the present invention, a method for detecting transaction data is provided, characterized in that it includes: in response to real-time detection of received transaction data, acquiring the transaction data and the transaction business type associated with the transaction data; determining the transaction interface and transaction channel to which the transaction data belongs; searching for one or more target detection models from a plurality of preset detection models that match the combination of the transaction business type and the transaction interface and the transaction channel; wherein the preset detection model includes at least one of a global detection model constructed for each of the transaction business types and a unit detection model constructed for the combination of multiple transaction interfaces and multiple transaction channels of the transaction business type; detecting the transaction data using the target detection model, and sending the detection result to a transaction management terminal.

[0006] Optionally, the method for detecting transaction data, after finding one or more target detection models that match the combination of the business type, the transaction interface, and the transaction channel, further includes: loading one or more of the target detection models into memory; the step of using the target detection models to detect the transaction data includes: using the target detection models loaded into memory to detect the transaction data.

[0007] Optionally, the method for detecting transaction data further includes: after the detection is completed and if it is determined that the target detection model meets the unloading conditions, unloading the target detection model from memory.

[0008] Optionally, the method for detecting transaction data further includes: each unit detection model includes a preset statistical model and unit sample data input to the preset statistical model; the step of constructing unit detection models for the various transaction interfaces and various transaction channels of the transaction business type included in the preset detection model includes: determining multiple detection indicators for the transaction interface required to calculate the unit sample data; for each combination of transaction interface and various transaction channels, obtaining historical transaction data corresponding to each detection indicator in the combination of transaction interface and various transaction channels; and calculating unit sample data of the combination of transaction interface and various transaction channels based on the historical transaction data corresponding to each detection indicator.

[0009] Optionally, the method for detecting transaction data further includes: the global detection model includes a preset statistical model and global sample data input to the preset statistical model; the preset detection model includes constructing a global detection model for each of the transaction business types, including: determining multiple detection indicators for the transaction interface required to calculate the global sample data; for each transaction business type, obtaining historical transaction data corresponding to each detection indicator in the combination of each of the transaction interfaces and multiple transaction channels included in the transaction business type; and calculating the global sample data corresponding to the transaction business type based on the historical transaction data corresponding to each detection indicator.

[0010] Optionally, for each combination of trading interfaces and multiple trading channels, calculating the unit sample data corresponding to the combination based on historical trading data of the combination includes: determining multiple detection indicators for the trading interfaces required to calculate the unit sample data; for each detection indicator of the trading interfaces in the combination, performing the following steps: obtaining sample data of the detection indicator under each trading channel; calculating a first statistical value of the detection indicator based on the sample data of each detection indicator, and calculating the information entropy of the detection indicator based on the probability between the first statistical value and the second statistical value of the detection indicator under each trading channel; calculating the entropy weight of the detection indicator among all detection indicators based on the information entropy of the detection indicator; and combining each detection indicator and the entropy weight corresponding to the detection indicator to obtain the unit sample data corresponding to the combination of the trading interface and multiple trading channels.

[0011] Optionally, the step of calculating the global sample data corresponding to each transaction business type based on the historical transaction data of the transaction business type includes: determining each transaction interface included in the transaction business type and the multiple transaction channels corresponding to each transaction interface; for each combination of transaction interface and multiple transaction channels corresponding to each transaction interface, performing the step of calculating the unit sample data corresponding to the combination; and combining the unit sample data corresponding to each combination to determine the global sample data of the transaction business type.

[0012] Optionally, the method for detecting transaction data further includes: the detection indicators include statistical transaction indicators and time-series transaction indicators; determining the multiple detection indicators for the transaction interface required to calculate the unit sample data includes: determining one or more of the statistical transaction indicators based on the statistical data of the historical transaction data corresponding to the detection indicators; extracting one or more time-series features from the historical transaction data using a time series model, including periodic features, stationary features, and drift features; and determining one or more of the time-series transaction indicators based on the statistical data of the historical transaction data corresponding to the detection indicators and one or more of the time-series features.

[0013] Optionally, for each combination of transaction interface and multiple transaction channels, after obtaining the historical transaction data corresponding to each detection indicator in the combination of the transaction interface and multiple transaction channels, the method further includes: performing preprocessing operations on the historical transaction data, wherein the preprocessing operations include one or more of the following: dispersion analysis, missing value imputation, data smoothing, and removal of abnormal data.

[0014] Optionally, detecting the transaction data using the target detection model includes: when the target detection model is a unit detection model, calculating a first difference between the transaction data and the unit sample data based on a preset statistical model included in the unit detection model and unit sample data input to the preset statistical model, and detecting the transaction data based on the first difference; when the target detection model is a global detection model, calculating a second difference between the transaction data and the global sample data based on a preset statistical model included in the global detection model and global sample data input to the preset statistical model, and detecting the transaction data based on the second difference.

[0015] Optionally, the method for detecting transaction data further includes: collecting transaction data for multiple transaction business types using a distributed message model; wherein the transaction data for each transaction business type is generated by calling one or more transaction interfaces of that transaction business type.

[0016] To achieve the above objectives, according to a second aspect of the present invention, an apparatus for detecting transaction data is provided, characterized in that it comprises: a data receiving module, a model determining module, and a data detecting module; wherein,

[0017] The data receiving module is used to respond to real-time detection of transaction data received from the client, obtain the transaction data and the transaction business type associated with the transaction data; and determine the transaction interface and transaction channel to which the transaction data belongs.

[0018] The model determination module is used to search for one or more target detection models from a variety of preset detection models that match the combination of the transaction business type, the transaction interface, and the transaction channel; wherein, the preset detection models include at least one of a global detection model constructed for each of the transaction business types and a unit detection model constructed for the combination of multiple transaction interfaces and multiple transaction channels of the transaction business type.

[0019] The detection data module is used to detect the transaction data using the target detection model and send the detection results to the transaction management terminal.

[0020] Optionally, the apparatus for detecting transaction data, after finding one or more target detection models that match the combination of the business type, the transaction interface, and the transaction channel, further includes: loading one or more of the target detection models into memory; the step of detecting the transaction data using the target detection models includes: detecting the transaction data using the target detection models loaded into memory.

[0021] Optionally, the device for detecting transaction data is further configured to unload the target detection model from memory after the detection is completed and if it is determined that the target detection model meets the unloading conditions.

[0022] Optionally, the device for detecting transaction data is further configured to include a preset statistical model and unit sample data input to the preset statistical model for each unit detection model; the step of constructing unit detection models for the various transaction interfaces and various transaction channels of the transaction business type included in the preset detection model includes: determining multiple detection indicators for the transaction interface required to calculate the unit sample data; for each combination of transaction interface and various transaction channels, obtaining historical transaction data corresponding to each detection indicator in the combination of transaction interface and various transaction channels; and calculating unit sample data of the combination of transaction interface and various transaction channels based on the historical transaction data corresponding to each detection indicator.

[0023] Optionally, the device for detecting transaction data includes a global detection model comprising a preset statistical model and global sample data input to the preset statistical model; the preset detection model includes constructing a global detection model for each of the transaction business types, comprising: determining multiple detection indicators for the transaction interface required to calculate the global sample data; for each transaction business type, acquiring historical transaction data corresponding to each detection indicator in the combination of each of the transaction interfaces and multiple transaction channels included in the transaction business type; and calculating global sample data corresponding to the transaction business type based on the historical transaction data corresponding to each detection indicator.

[0024] Optionally, the device for detecting transaction data is configured to, for each combination of transaction interface and multiple transaction channels, calculate unit sample data corresponding to the combination based on historical transaction data of the combination of transaction interface and multiple transaction channels, including: determining multiple detection indicators for the transaction interface required to calculate the unit sample data; for each detection indicator of the transaction interface in the combination, performing the following steps: acquiring sample data of the detection indicator under each of the transaction channels; calculating a first statistical value of the detection indicator based on the sample data of each detection indicator, and calculating the information entropy of the detection indicator based on the probability between the first statistical value and the second statistical value of the detection indicator under each transaction channel; calculating the entropy weight of the detection indicator among all detection indicators based on the information entropy of the detection indicator; and combining each detection indicator and the entropy weight corresponding to the detection indicator to obtain the unit sample data corresponding to the combination of the transaction interface and multiple transaction channels.

[0025] Optionally, the device for detecting transaction data is configured to calculate global sample data corresponding to each transaction business type based on historical transaction data of that transaction business type, including: determining each transaction interface included in the transaction business type and multiple transaction channels corresponding to each transaction interface; for each combination of transaction interface and multiple transaction channels corresponding to each transaction interface, performing the step of calculating unit sample data corresponding to the combination; and combining the unit sample data corresponding to each combination to determine the global sample data of the transaction business type.

[0026] Optionally, the device for detecting transaction data includes the detection indicators comprising statistical transaction indicators and time-series transaction indicators; determining the multiple detection indicators for the transaction interface required to calculate the unit sample data includes: determining one or more of the statistical transaction indicators based on the statistical data of the historical transaction data corresponding to the detection indicators; extracting one or more time-series features from the historical transaction data using a time series model, including periodic features, stationary features, and drift features; and determining one or more of the time-series transaction indicators based on the statistical data of the historical transaction data corresponding to the detection indicators and one or more of the time-series features.

[0027] Optionally, the device for detecting transaction data is configured to, for each combination of transaction interface and multiple transaction channels, after acquiring historical transaction data corresponding to each detection indicator in the combination of transaction interface and multiple transaction channels, further include: performing preprocessing operations on the historical transaction data, wherein the preprocessing operations include one or more of the following: dispersion analysis, missing value imputation, data smoothing, and removal of abnormal data.

[0028] Optionally, the apparatus for detecting transaction data is used to detect the transaction data using the target detection model, comprising: when the target detection model is a unit detection model, calculating a first difference between the transaction data and the unit sample data based on a preset statistical model included in the unit detection model and unit sample data input to the preset statistical model, and detecting the transaction data based on the first difference; when the target detection model is a global detection model, calculating a second difference between the transaction data and the global sample data based on a preset statistical model included in the global detection model and global sample data input to the preset statistical model, and detecting the transaction data based on the second difference.

[0029] Optionally, the device for detecting transaction data is further configured to collect transaction data for multiple transaction business types using a distributed message model; wherein the transaction data for each transaction business type is generated by calling one or more transaction interfaces of that transaction business type.

[0030] To achieve the above objectives, according to a third aspect of the present invention, an electronic device for detecting transaction data is provided, characterized in that it includes: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement any of the methods described above for detecting transaction data.

[0031] To achieve the above objectives, according to a fourth aspect of the present invention, a computer-readable medium is provided having a computer program stored thereon, characterized in that, when the program is executed by a processor, it implements any of the methods described above for detecting transaction data.

[0032] To achieve the above objectives, according to a fifth aspect of the present invention, a computer program product is provided, comprising a computer program, characterized in that, when the program is executed by a processor, it implements any of the methods described above for detecting transaction data.

[0033] One embodiment of the above invention has the following advantages or beneficial effects: by determining the corresponding global detection model for the transaction business type and determining the unit detection model for the combination of different transaction interfaces and different transaction channels, the transaction data of the transaction business type can be detected in real time using different detection models, thereby improving the accuracy, real-time performance and fineness of abnormal data detection in the transaction data and improving the efficiency of data detection.

[0034] The further effects of the aforementioned unconventional alternative methods will be explained below in conjunction with specific implementation methods. Attached Figure Description

[0035] The accompanying drawings are provided to better understand the invention and are not intended to unduly limit the scope of the invention. Wherein:

[0036] Figure 1 This is a flowchart illustrating a method for detecting transaction data according to an embodiment of the present invention;

[0037] Figure 2 This is a schematic diagram of the structure of a platform for detecting transaction data provided in one embodiment of the present invention;

[0038] Figure 3 This is a schematic diagram of the structure of a device for detecting transaction data according to an embodiment of the present invention;

[0039] Figure 4 This is an exemplary system architecture diagram in which embodiments of the present invention can be applied;

[0040] Figure 5This is a schematic diagram of the structure of a computer system suitable for implementing terminal devices or servers of the present invention. Detailed Implementation

[0041] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of the present invention, including various details to aid understanding. These details should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description. The acquisition, storage, use, and processing of data in the technical solutions of this application all comply with relevant national laws and regulations.

[0042] like Figure 1 As shown, this embodiment of the invention provides a method for detecting transaction data, which may include the following steps:

[0043] Step S101: In response to the real-time detection of received transaction data, obtain the transaction data and the transaction business type associated with the transaction data; determine the transaction interface and transaction channel to which the transaction data belongs.

[0044] Specifically, transaction data can be received by being pushed by the application system, collected from data sources of multiple application systems, or obtained through real-time message interaction via a distributed message center. Transaction data refers to data generated by the application system under various transaction business types through multiple transaction interfaces and multiple transaction channels. For example, transaction data can be the transaction volume generated through one or more transaction interfaces, or the transaction amount generated through transaction interfaces. Transaction interfaces can be, for example, APIs (Application Programming Interfaces) used in the code that processes transaction business.

[0045] Furthermore, the transaction business type is set by the application scenario. For example, the transaction business type can be payment business, collection business, buying and selling financial products, etc. Multiple transaction interfaces can be applied to multiple pages. For example, taking payment business as an example, payment business includes payment interface 1, payment interface 2, and payment interface 3. Payment interface 1, payment interface 2, and payment interface 3 are transaction interfaces, and different transaction interfaces can be used for different payment scenarios. For example, payment scenarios on product details pages, order pages, or shopping cart pages, etc. Furthermore, multiple transaction channels are transaction methods associated with the transaction business type. Again, taking payment business as an example, multiple transaction channels include, for example, payment through bank applications, payment through third-party applications, or payment through bank mini-programs, etc. This invention does not limit the specific content and form of the transaction interface and transaction channel to which the transaction data belongs. That is, in response to a certain transaction business type, the transaction data and the transaction business type associated with the transaction data are obtained; the transaction interface and transaction channel to which the transaction data belongs are determined.

[0046] Further preferably, in embodiments of the present invention, a distributed message model is used to collect transaction data for multiple transaction business types; wherein the transaction data for each transaction business type is generated by one or more transaction interfaces running that transaction business type. For example, for different transaction interfaces, a distributed message collection module bound to the transaction interface is used to collect real-time transaction data generated for that transaction interface, and a distributed message center in cluster mode is used to obtain messages collected from each distributed message collection module, thereby obtaining the transaction data to be detected through the transaction data information carried in the messages. The transaction data for each transaction business type is generated by one or more transaction interfaces running that transaction business type; taking payment business as an example, for example, the different pages corresponding to the various transaction interfaces include product details pages, order pages, shopping cart pages, etc.; wherein the page management end (e.g., the front end) can obtain transaction data, transaction data-related logs, and related data that can reflect the transaction data (e.g., transaction volume, transaction amount, etc.) through event tracking according to the application scenario.

[0047] Step S102: From a variety of preset detection models, find one or more target detection models that match the combination of the transaction business type, the transaction interface, and the transaction channel; wherein, the preset detection models include at least one of a global detection model constructed for each of the transaction business types and a unit detection model constructed for the combination of multiple transaction interfaces and multiple transaction channels of the transaction business type.

[0048] Specifically, for the transaction business type of the received transaction data, one or more target detection models are searched that match the combination of the transaction business type, the transaction interface, and the transaction channel. In the embodiments of the present invention, the preset detection model includes at least one of a global detection model constructed for each of the transaction business types and a unit detection model constructed for the combination of multiple transaction interfaces and multiple transaction channels of the transaction business type. It can be seen that the detection model of the present invention can provide different detection models from different dimensions and different detection needs, which improves the precision of transaction data detection, thereby improving the accuracy, real-time performance, and efficiency of transaction data detection.

[0049] Each of the multiple unit detection models is associated with one of the combinations of multiple transaction interfaces and multiple transaction channels. For the same transaction business type, multiple unit detection models and / or global detection models can be set. Taking the payment business type as an example, unit detection models can be built for transaction interface A on Bank A's APP, Bank B's mini-program, and a third-party APP respectively. A corresponding unit detection model can be built for transaction interface B on Bank A's APP, Bank B's mini-program, and a third-party APP respectively. The global detection model for the payment business is built for the payment business type, such as a detection model that includes multiple transaction interfaces and multiple transaction channels. That is, the preset detection model includes at least one of the global detection model built for each of the transaction business types and the unit detection models built for the combinations of multiple transaction interfaces and multiple transaction channels of the transaction business type.

[0050] Furthermore, after finding one or more target detection models (including unit detection models and / or global detection models) corresponding to the transaction business type, the target detection model is loaded into memory. After the transaction data is detected using the target detection model, if the target detection model meets the unloading conditions, it is unloaded from memory. The unloading conditions for the target detection model include, for example, the absence of transaction data for the current transaction business type to be detected, the completion of all transaction data detection, the reaching of a set unloading time point, or the receipt of a message triggering the unloading of the target detection model. Therefore, loading the target detection model into memory improves the speed of transaction data detection; unloading the target detection model reduces memory resource consumption. In addition to unloading the target detection model during the model unloading phase, configuration information loaded along with the target detection model (such as model rules, threshold values, and keyword matrices) can also be unloaded simultaneously. That is, after finding one or more target detection models that match the combination of the business type, the transaction interface, and the transaction channel, the process further includes: loading one or more of the target detection models into memory; and detecting the transaction data using the target detection models, which includes: detecting the transaction data using the target detection models loaded into memory. Further, after detection is completed and if it is determined that the target detection model meets the unloading conditions, the target detection model is unloaded from memory.

[0051] Furthermore, embodiments of the present invention provide methods for constructing unit detection models and global detection models.

[0052] Specifically, the method for constructing a unit detection model includes constructing a preset statistical model in the unit detection model and calculating the unit sample data input to the preset statistical model. That is, each unit detection model includes a preset statistical model and unit sample data input to the preset statistical model;

[0053] Furthermore, the preset statistical model can be a preset statistical model built based on the median absolute deviation (MAD) algorithm of the data. The MAD model improves the detection efficiency of business types in application scenarios with large data fluctuations (such as securities and finance). The preset statistical model can also utilize various methods such as scatter plots, box plots, distance measurement, and clustering models of statistical data. It can be understood that the calculation method or model parameters of the unit detection model can be updated according to the distribution changes of historical transaction data, which further improves the accuracy and efficiency of detecting transaction data.

[0054] In an embodiment of the present invention, MAD is used as an example to illustrate the preset statistical model. The calculation formula of MAD is shown in formula (1):

[0055] MAD = C × median|S i -median(S)|(1) where the confidence interval is S i ∈(median-k×MAD,median+k×MAD), where C and k are parameters of the MAD model. C and k can be adjusted according to the actual application scenario to adjust the confidence interval, thereby meeting the accuracy requirements of the preset statistical model; S is the unit sample data (or global sample data) input to the preset statistical model; S i The data to be detected is the transaction data (e.g., transaction volume).

[0056] First, the method for calculating the unit sample data input to the preset statistical model for the unit detection model is explained:

[0057] The preset detection model includes constructing unit detection models for the various transaction interfaces and combinations of transaction channels for the transaction business types, including: determining multiple detection indicators for the transaction interfaces required to calculate the unit sample data; for each combination of transaction interface and multiple transaction channels, obtaining historical transaction data corresponding to each detection indicator in the combination of transaction interface and multiple transaction channels; and calculating the unit sample data of the combination of transaction interface and multiple transaction channels based on the historical transaction data corresponding to each detection indicator.

[0058] In one embodiment of the present invention, multiple detection indicators for the transaction interface required to calculate the unit sample data are determined. Specifically, for a certain transaction interface, taking transaction volume as an example, the seven detection indicators determined for a certain time range are: average transaction volume of the transaction interface, peak transaction volume of the transaction interface, month-on-month change of transaction volume of the transaction interface within a previous set time range, year-on-year change of transaction volume of the transaction interface in the same order of the previous day, year-on-year change of transaction volume of the transaction interface in the same order of the previous week, year-on-year change of transaction volume of the transaction interface in the same order of the previous month, and year-on-year change of transaction volume of the transaction interface in the same order of the previous month.

[0059] Furthermore, based on the historical transaction data corresponding to each of the detection indicators, unit sample data of the combination of the transaction interface and the various transaction channels are calculated.

[0060] The detection indicators include statistical transaction indicators and time-series transaction indicators. Statistical transaction indicators include, for example, the average transaction volume of the transaction interface and the peak transaction volume of the transaction interface. Time-series transaction indicators include, for example, the month-on-month change in transaction volume over a previous set time range, the year-on-year change in transaction volume on the previous day, the year-on-year change in transaction volume on the previous week, the year-on-year change in transaction volume on the same day of the previous month, and the year-on-year change in transaction volume on the same day of the previous month. The method for extracting time-series transaction indicators can utilize time series models to extract one or more time-series features from the historical transaction data, including periodicity, stability, and drift. Specifically, extracting one or more time-series features from the historical transaction data is crucial because transaction data is directly related to time; therefore, further extraction of one or more time-series features from the historical transaction data is necessary. Features, such as daily transaction volume, monthly transaction volume, month-on-month transaction volume, and year-on-year transaction volume at a specific point in time, are considered periodic features. Stability features include methods such as using historical transaction data of the unit detection index to calculate maximum / minimum differences, statistical methods, and percentage measurements to determine data stability. Drift features include detecting significant drift in the distribution of historical transaction data as it changes over time, and predicting deviations after historical transaction data aggregation and time-series feature statistics. In one embodiment of the invention, the Holt-Winters time series algorithm can be used to predict deviations after historical transaction data aggregation and to perform time-series feature statistics on historical transaction data to obtain one or more time-series features. Therefore, by obtaining time-related time-series features from historical transaction data, the accuracy and scalability of the unit detection model are further improved. Furthermore, one or more time-series transaction indicators are determined based on the statistical data of the historical transaction data corresponding to the detection indicators and one or more of the aforementioned time-series features. That is, determining the multiple detection indicators for the transaction interface required to calculate the unit sample data includes: determining one or more statistical transaction indicators based on the statistical data of the historical transaction data corresponding to the detection indicators; extracting one or more time-series features from the historical transaction data using a time series model, including periodic features, stationary features, and drift features; and determining one or more time-series transaction indicators based on the statistical data of the historical transaction data corresponding to the detection indicators and one or more of the time-series features.

[0061] Preferably, the historical transaction data used to construct the unit detection model or the global detection model is preprocessed. By improving the effectiveness of the historical transaction data, the accuracy of the unit detection model is improved, and the computational resources required to calculate the unit detection model or the global detection model are reduced. Specifically, the data preprocessing operations include one or more of the following: dispersion analysis, missing value imputation, data smoothing, and outlier removal. The preprocessed data is then used to determine the unit sample data or global template data for the detection indicators. That is, for each combination of transaction interfaces and multiple transaction channels, after obtaining the historical transaction data corresponding to each detection indicator in the combination of the transaction interface and multiple transaction channels, the process further includes: performing preprocessing operations on the historical transaction data, wherein the preprocessing operations include one or more of the following: dispersion analysis, missing value imputation, data smoothing, and outlier removal.

[0062] Furthermore, for each combination of a transaction interface and multiple transaction channels, based on historical transaction data of the combination, unit sample data corresponding to the combination is calculated, including: determining multiple detection indicators for the transaction interface required to calculate the unit sample data; for example: using C i The seven detection indicators are as follows:

[0063] C i ={C1,C2,C3,C4,C5,C6,C7},i={1,2,3,4,5,6,7}

[0064] Suppose there are n transaction channels, denoted as j = {1, 2, 3, ..., n};

[0065] Then, for each detection indicator of the transaction interface in the combination, perform the following steps:

[0066] 1) Obtain sample data of the detection indicator under each of the transaction channels; calculate the first statistical value of the detection indicator based on the sample data of each of the detection indicators;

[0067] Specifically, for example, for each detection indicator, the sample data for that indicator is normalized and standardized based on historical transaction data. The calculation of normalization and standardization is as follows: for C i The sample data set for each channel is represented as C. ij Then, perform normalized and standardized calculations (i.e., calculate the first statistical value of the detection index), and the calculation formula is as follows:

[0068] X ij =(C ij -min(C i )) / (max(C i)-min(C i ))

[0069] Where min represents the minimum value, max represents the maximum value, and X ij Represents the first statistical value;

[0070] 2) And calculate the information entropy of the detection indicator based on the probability between the first statistical value and the second statistical value of the detection indicator under each transaction channel; wherein, the second statistical value is X. ij The sum of these values ​​across n transaction channels represents the second statistical value of the detection indicator across each transaction channel; according to the following formula, P... ij The numerator represents the first statistical value, and the denominator represents the second statistical value of the detection index under various transaction channels.

[0071]

[0072] Further, based on the probability between the first statistical value and the second statistical value of the detection indicator under various transaction channels, the information entropy of the detection indicator is calculated; the information entropy is calculated.

[0073]

[0074] The method for calculating entropy is shown in the following formula:

[0075] Among them, E i The information entropy represents the detection index.

[0076] For the seven detection indicators, the information entropy of each indicator is expressed as: E i ={E1,E2,E3,E4,E5,E6,E7},i={1,2,3,4,5,6,7}

[0077] 3) Based on the information entropy of the detection index, calculate the entropy weight of the detection index among all detection indices;

[0078] Specifically, the entropy weight of the detection index among all detection indexes is calculated using the following formula; where Wi represents the entropy weight, and in the embodiment of the present invention, k = 7;

[0079]

[0080] Furthermore, by iterating through steps 1)-3), the entropy weight of each detection indicator can be calculated.

[0081] 4) By combining each of the detection indicators and the entropy weights corresponding to the detection indicators, the unit sample data corresponding to the combination of the transaction interface and multiple transaction channels is obtained.

[0082] Specifically, the unit sample data corresponding to the combination of the transaction interface and multiple transaction channels is obtained using the following formula:

[0083]

[0084] Where b represents the number of trading interfaces, and BW represents the weight array corresponding to different trading interfaces;

[0085] When b=1 and BW=1, S represents the unit sample data of the unit detection model; the calculated S represents the unit sample data of the unit detection model.

[0086] Furthermore, the method for calculating the global sample data of the preset statistical model as input to the global detection model is explained as follows: It can be understood that a global detection model can correspond to a transaction business type, a transaction business type can contain multiple transaction interfaces, and each transaction interface corresponds to multiple transaction channels. Therefore, the detection indicators contained in the global sample data of the global detection model can be consistent with those of the unit detection model. Thus, for each transaction interface in the global sample data, the calculation method can be consistent with the steps for calculating the unit sample data, and will not be elaborated here. That is, for each transaction business type, calculating the global sample data corresponding to the transaction business type based on the historical transaction data of the transaction business type includes: determining each transaction interface contained in the transaction business type and the multiple transaction channels corresponding to each transaction interface; for each combination of transaction interfaces and the multiple transaction channels corresponding to each transaction interface, performing the step of calculating the unit sample data corresponding to the combination; and combining the unit sample data corresponding to each combination to determine the global sample data of the transaction business type.

[0087] Furthermore, let's take the following formula as an example to illustrate the global sample data;

[0088]

[0089] Where b represents the number of transaction interfaces, and BW represents the weight array corresponding to different transaction interfaces; when b>1, S represents the global sample data of the global detection model; that is, the global sample data corresponding to the transaction business type is calculated based on the historical transaction data corresponding to each detection indicator. Specifically, the global detection model includes a preset statistical model and global sample data input to the preset statistical model; the preset detection model includes constructing a global detection model for each transaction business type, including: determining multiple detection indicators for the transaction interfaces required to calculate the global sample data; for each transaction business type, obtaining historical transaction data corresponding to each detection indicator in the combination of each transaction interface and multiple transaction channels included in the transaction business type; and calculating the global sample data corresponding to the transaction business type based on the historical transaction data corresponding to each detection indicator.

[0090] Step S103: Use the target detection model to detect the transaction data and send the detection results to the transaction management terminal.

[0091] Specifically, the transaction data is detected using one or more identified target detection models, and the detection results are sent to the transaction management terminal. The detection results may include the presence of abnormal data, detailed information about the abnormal data, various visualization styles of the abnormal data, and abnormal data analysis, allowing the transaction management terminal to perform targeted further processing based on the received real-time detection results. The transaction management terminal can be a client, server, or similar entity.

[0092] Furthermore, the step of detecting the transaction data using the target detection model includes:

[0093] 1) When the target detection model is a unit detection model, based on the preset statistical model included in the unit detection model and the unit sample data input to the preset statistical model, calculate the first difference between the transaction data and the unit sample data, and detect the transaction data according to the first difference; Specifically, according to the preset statistical model described by formula (1), when the target detection model is a unit detection model, S is the unit sample data, then calculate the first difference between the transaction data and the unit sample data according to formula (1), and determine whether the first difference is within the preset transaction data threshold range, for example: whether it is within the preset transaction volume range. If it is, then determine that the transaction data is not abnormal, that is, detect the transaction data according to the first difference;

[0094] 2) When the target detection model is a global detection model, based on the preset statistical model included in the global detection model and the global sample data input to the preset statistical model, calculate the second difference between the transaction data and the global sample data, and detect the transaction data according to the second difference.

[0095] Specifically, according to the preset statistical model described by formula (1), when the target detection model is a unit detection model, S is the global sample data. Then, according to formula (1), the second difference between the transaction data and the global sample data is calculated, and it is determined whether the second difference is within the preset transaction data threshold range, for example, whether it is within the preset transaction volume range. If it is, it is determined that the transaction data is not abnormal, that is, the transaction data is detected according to the second difference.

[0096] It is understandable that different types of transaction management terminals may have different detection business scopes, and the unit detection model and / or global detection model used to detect transaction data can be determined according to the different detection business scopes.

[0097] like Figure 2 As shown in the diagram, this embodiment of the invention provides a structural schematic of a platform for detecting transaction data; it includes: a business platform 201 and an anomaly monitoring platform 202;

[0098] In embodiments of the present invention, the automatic detection of transaction data can be completed through multiple components in the business platform 201 and the anomaly detection platform 202, as well as the interaction between the components. For example, the business platform 201 uses multiple distributed message collection components combined with data acquisition operations to collect various transaction data from the corresponding business scenario system (e.g., a business scenario using multiple transaction interfaces) and the transaction channels corresponding to those transaction interfaces (e.g., multiple applications). The business platform 201 and the distributed message center included in the anomaly detection platform 202 transmit message and transaction data.

[0099] The anomaly detection platform 202 includes components such as a distributed message center, an anomaly detection system, and an operations module. The distributed message center (e.g., in cluster mode) within the anomaly detection platform 202 can receive messages and maintain the effectiveness and efficiency of transaction data interaction with the business platform 201 through a transaction data receiving module, a transaction data storage module, and a fault recovery module (used in cluster mode to retrieve data from the transaction data source for recovery in the event of a system failure, ensuring platform and data stability). Simultaneously, the distributed message center can also receive detection alerts from the anomaly detection system and send them to the transaction management terminal in real time. The anomaly detection system includes, for example, a... The anomaly detection platform 202 includes modules for loading and monitoring the detection model, detecting transaction data, and obtaining detection results. Each module executes the process of the method described in this embodiment. The operation module can perform various operations such as sending anomaly views, monitoring model updates, obtaining model configurations, and maintaining detection results. It receives configuration information related to the model from the transaction management terminal (e.g., transaction business type, one or more transaction interfaces, one or more transaction channels), and can also record, store, analyze, and maintain detection results, and send anomaly results (e.g., anomaly views) to the transaction management terminal. By using different components, the coupling in the anomaly detection process is reduced, thereby improving the reliability of the detection data. Preferably, the anomaly detection platform 202 also includes dynamic model training and computation of various detection models (global detection model, unit detection model). It is understood that transaction data has a high probability of various fluctuations; therefore, by dynamically updating various detection models, the real-time performance of the detection models is maintained, thereby improving the accuracy of detecting real-time transaction data.

[0100] like Figure 3 As shown, this embodiment of the invention provides a device 300 for detecting transaction data, including: a data receiving module 301, a model determining module 302, and a data detecting module 303; wherein,

[0101] The data receiving module 301 is used to respond to real-time detection of transaction data received from the client, obtain the transaction data and the transaction business type associated with the transaction data; and determine the transaction interface and transaction channel to which the transaction data belongs.

[0102] The model determination module 302 is used to search for one or more target detection models from a variety of preset detection models that match the combination of the transaction business type, the transaction interface, and the transaction channel; wherein, the preset detection models include at least one of a global detection model constructed for each of the transaction business types and a unit detection model constructed for the combination of multiple transaction interfaces and multiple transaction channels of the transaction business type.

[0103] The detection data module 303 is used to detect the transaction data using the target detection model and send the detection results to the transaction management terminal.

[0104] This invention also provides an electronic device for detecting transaction data, comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method provided in any of the above embodiments.

[0105] This invention also provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the method provided in any of the above embodiments.

[0106] This invention also provides a computer program product, including a computer program, characterized in that the program, when executed by a processor, implements the method provided in any of the above embodiments.

[0107] Figure 4 An exemplary system architecture 400 for detecting transaction data, which can be applied to embodiments of the present invention, is shown.

[0108] like Figure 4 As shown, system architecture 400 may include terminal devices 401, 402, and 403, a network 404, and a server 405. Network 404 serves as the medium for providing communication links between terminal devices 401, 402, and 403 and server 405. Network 404 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0109] Users can use terminal devices 401, 402, and 403 to interact with server 405 via network 404 to receive or send messages, etc. Various client applications can be installed on terminal devices 401, 402, and 403, such as e-commerce client applications, web browser applications, search applications, instant messaging tools, and email clients.

[0110] Terminal devices 401, 402, and 403 can be various electronic devices with displays and supporting various client applications, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0111] Server 405 can be a server that provides various services, such as a backend management server that supports the client application systems used by users through terminal devices 401, 402, and 403. The backend management server can detect received real-time transaction data and feed the detection results back to the terminal devices.

[0112] It should be noted that the method for detecting transaction data provided in this embodiment of the invention is generally executed by server 405, and correspondingly, the device for detecting transaction data is generally located in server 405.

[0113] It should be understood that Figure 4 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0114] The following is for reference. Figure 5 It shows a schematic diagram of the structure of a computer system 500 suitable for implementing a terminal device of the present invention. Figure 5 The terminal device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0115] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 502 or programs loaded from storage section 508 into random access memory (RAM) 503. The RAM 503 also stores various programs and data required for the operation of the system 500. The CPU 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0116] The following components are connected to I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to I / O interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 510 as needed so that computer programs read from it can be installed into storage section 508 as needed.

[0117] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by central processing unit (CPU) 501, it performs the functions defined above in the system of this invention.

[0118] It should be noted that the computer-readable medium shown in this invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0119] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0120] The modules and / or units described in the embodiments of the present invention can be implemented in software or hardware. The described modules and / or units can also be housed in a processor; for example, a processor can be described as including a data receiving module, a model determining module, and a data detection module. The names of these modules do not necessarily limit the module itself; for example, the data detection module can also be described as "a module that detects the transaction data using a target detection model."

[0121] In another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable medium carries one or more programs, which, when executed by the device, cause the device to include: in response to detecting received transaction data, acquiring the transaction data and the transaction business type associated with the transaction data; determining the transaction interface and transaction channel to which the transaction data belongs; searching from a plurality of preset detection models for one or more target detection models that match the combination of the transaction business type and the transaction interface and the transaction channel; wherein the preset detection model includes at least one of a global detection model constructed for each of the transaction business types and a unit detection model constructed for the combination of multiple transaction interfaces and multiple transaction channels of the transaction business type; detecting the transaction data using the target detection model and sending the detection result to a transaction management terminal.

[0122] The embodiments of the present invention can, based on the transaction business type, transaction interface, and transaction channel to which the transaction data to be detected belongs, search for one or more matching target detection models from a variety of preset global detection models and / or unit detection models, and use the found target detection models to detect the transaction data. The embodiments of the present invention, by determining a global detection model for the transaction business type and determining unit detection models for different combinations of transaction interfaces and transaction channels, utilize different detection models to detect transaction data, thereby improving the accuracy and refinement of detecting abnormal data in transaction data and increasing the efficiency of data detection.

[0123] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for detecting transaction data, characterized in that, include: In response to the detected transaction data, the system obtains the transaction data and the transaction business type associated with the transaction data. Determine the transaction interface and transaction channel to which the transaction data belongs; From a variety of preset detection models, one or more target detection models are searched to match the combination of the transaction business type, the transaction interface, and the transaction channel; wherein, the preset detection models include at least one of a global detection model constructed for each of the transaction business types and a unit detection model constructed for the combination of multiple transaction interfaces and multiple transaction channels of the transaction business type; The target detection model is used to detect the transaction data, and the detection results are sent to the transaction management terminal. The method further includes: each unit detection model includes a preset statistical model and unit sample data input to the preset statistical model; The preset detection model includes constructing unit detection models for the various transaction interfaces and combinations of transaction channels for the transaction business types, including: determining multiple detection indicators for the transaction interfaces required to calculate the unit sample data; for each combination of transaction interfaces and multiple transaction channels, obtaining historical transaction data corresponding to each detection indicator in the combination of transaction interfaces and multiple transaction channels; and calculating unit sample data for the combination of transaction interfaces and multiple transaction channels based on the historical transaction data corresponding to each detection indicator. The global detection model includes a preset statistical model and global sample data input to the preset statistical model; The preset detection model includes constructing a global detection model for each of the transaction business types, including: determining multiple detection indicators for the transaction interface required to calculate the global sample data; for each transaction business type, obtaining historical transaction data corresponding to each detection indicator in the combination of each transaction interface and multiple transaction channels included in the transaction business type; and calculating the global sample data corresponding to the transaction business type based on the historical transaction data corresponding to each detection indicator.

2. The method according to claim 1, characterized in that, After finding one or more target detection models that match the combination of the business type, the transaction interface, and the transaction channel, the method further includes: Load one or more of the target detection models described above into memory; The process of detecting the transaction data using the target detection model includes: The transaction data is detected using the target detection model loaded into memory.

3. The method according to claim 2, characterized in that, Also includes: After the detection is completed and it is determined that the target detection model meets the unloading conditions, the target detection model is unloaded from memory.

4. The method according to claim 1, characterized in that, For each combination of a transaction interface and multiple transaction channels, based on the historical transaction data of the combination, the corresponding unit sample data is calculated, including: Determine the multiple detection metrics for the transaction interface required to calculate the unit sample data; For each of the detection metrics of the transaction interface in the combination, perform: Obtain sample data of the detection indicators for each of the aforementioned transaction channels; Based on the sample data of each detection indicator, a first statistical value of the detection indicator is calculated, and the information entropy of the detection indicator is calculated based on the probability between the first statistical value and the second statistical value of the detection indicator under each transaction channel. Based on the information entropy of the detection index, the entropy weight of the detection index among all detection indexes is calculated. By combining each of the aforementioned detection indicators and the corresponding entropy weights, unit sample data corresponding to the combination of the transaction interface and multiple transaction channels is obtained.

5. The method according to claim 1 or 4, characterized in that, For each type of transaction business, the global sample data corresponding to that transaction business type is calculated based on the historical transaction data of that transaction business type, including: Identify each transaction interface included in the transaction business type and the multiple transaction channels corresponding to each transaction interface; For each transaction interface and the combination of multiple transaction channels corresponding to each transaction interface, the step of calculating the unit sample data corresponding to the combination is performed; By combining the unit sample data corresponding to each combination, the global sample data for the transaction business type is determined.

6. The method according to claim 4, characterized in that, Further includes: The detection indicators include statistical transaction indicators and time-series transaction indicators; The determination of the multiple detection metrics for the transaction interface required to calculate the unit sample data includes: One or more of the statistical transaction indicators are determined based on the statistical data of the historical transaction data corresponding to the detection indicators; Using a time series model, one or more time-series features, including periodic features, stationary features, and drift features, are extracted from the historical transaction data. One or more time-series trading indicators are determined based on statistical data of historical trading data corresponding to the detection indicators and one or more of the time-series features.

7. The method according to claim 1, characterized in that: For each combination of a transaction interface and multiple transaction channels, after obtaining the historical transaction data corresponding to each detection indicator in the combination of the transaction interface and multiple transaction channels, the process further includes: The historical transaction data is preprocessed, and the preprocessing operations include one or more of the following: dispersion analysis, missing value imputation, data smoothing, and removal of outlier data.

8. The method according to claim 1, characterized in that, The process of detecting the transaction data using the target detection model includes: When the target detection model is a unit detection model, based on the preset statistical model included in the unit detection model and the unit sample data input to the preset statistical model, a first difference between the transaction data and the unit sample data is calculated, and the transaction data is detected according to the first difference; When the target detection model is a global detection model, based on the preset statistical model included in the global detection model and the global sample data input to the preset statistical model, a second difference between the transaction data and the global sample data is calculated, and the transaction data is detected based on the second difference.

9. The method according to claim 1, characterized in that, Also includes: The system utilizes a distributed messaging model to collect transaction data for multiple transaction business types; wherein the transaction data for each transaction business type is generated by calling one or more transaction interfaces of that transaction business type.

10. A device for detecting transaction data, characterized in that, include: The system comprises a data receiving module, a model determination module, and a detection data module; among which, The data receiving module is used to, in response to detecting received transaction data, obtain the transaction data, the transaction business type associated with the transaction data, and determine the transaction interface and transaction channel to which the transaction data belongs; The model determination module is used to search for one or more target detection models from a variety of preset detection models that match the combination of the transaction business type, the transaction interface, and the transaction channel; wherein, the preset detection models include at least one of a global detection model constructed for each of the transaction business types and a unit detection model constructed for the combination of multiple transaction interfaces and multiple transaction channels of the transaction business type. The detection data module is used to detect the transaction data using the target detection model and send the detection results to the transaction management terminal. The device further includes: each of the unit detection models includes a preset statistical model and unit sample data input to the preset statistical model; The preset detection model includes constructing unit detection models for the various transaction interfaces and combinations of transaction channels for the transaction business types, including: determining multiple detection indicators for the transaction interfaces required to calculate the unit sample data; for each combination of transaction interfaces and multiple transaction channels, obtaining historical transaction data corresponding to each detection indicator in the combination of transaction interfaces and multiple transaction channels; and calculating unit sample data for the combination of transaction interfaces and multiple transaction channels based on the historical transaction data corresponding to each detection indicator. The global detection model includes a preset statistical model and global sample data input to the preset statistical model; The preset detection model includes constructing a global detection model for each of the transaction business types, including: determining multiple detection indicators for the transaction interface required to calculate the global sample data; for each transaction business type, obtaining historical transaction data corresponding to each detection indicator in the combination of each transaction interface and multiple transaction channels included in the transaction business type; and calculating the global sample data corresponding to the transaction business type based on the historical transaction data corresponding to each detection indicator.

11. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-9.

12. A computer-readable medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-9.

13. A computer program product, comprising a computer program, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-9.

Images