CN107025547B - Payment channel detection method and device and terminal - Google Patents
Payment channel detection method and device and terminal Download PDFInfo
- Publication number
- CN107025547B CN107025547B CN201610833859.XA CN201610833859A CN107025547B CN 107025547 B CN107025547 B CN 107025547B CN 201610833859 A CN201610833859 A CN 201610833859A CN 107025547 B CN107025547 B CN 107025547B
- Authority
- CN
- China
- Prior art keywords
- payment channel
- code
- abnormal
- trained
- svm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/08—Payment architectures
- G06Q20/16—Payments settled via telecommunication systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Accounting & Taxation (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a payment channel detection method, a device and a terminal, wherein the method comprises the following steps: acquiring code operating parameters of systems forming a payment channel, wherein each system at least comprises a section of code; performing abnormal operation parameter identification on the acquired code operation parameters through a trained SVM multi-classification model, wherein the trained SVM multi-classification model comprises a plurality of SVM classifiers, and one SVM classifier is used for identifying one abnormal operation parameter in the acquired code operation parameters; and determining the abnormal operation part of the payment channel according to the recognition result. By implementing the method and the device, when the payment channel is abnormal, the abnormal operation part can be quickly and accurately positioned without manual intervention, and the operation can be simplified while the positioning speed is improved.
Description
Technical Field
The present application relates to the field of network communication technologies, and in particular, to a payment channel detection method, an apparatus, and a terminal.
Background
With the increase of payment providers and payment agents, more and more payment platforms are built, and more payment channels for payment based on the payment platforms are built. In the actual payment process, because the number of payment providers, payment agents or payment networks involved in the payment channel is large, the abnormal situation of the payment channel is inevitable, such as: the transaction success rate per unit time is significantly reduced.
Because the stability of a single payment channel has certain uncertainty, the current payment channel maintenance system is difficult to automatically locate a problem channel position. Therefore, when the payment channel is abnormal, the problem can be checked only through manual intervention, and the problem checking is performed manually, so that a large amount of checking experience is needed, certain blindness exists, the checking time is long, and the efficiency is low. In addition, the long troubleshooting time may also affect the payment experience of the user.
Disclosure of Invention
The application provides a payment channel detection method, a payment channel detection device and a payment channel detection terminal, which are used for solving the problems of long time and low efficiency of the conventional payment channel maintenance system in channel abnormity troubleshooting.
According to a first aspect of embodiments of the present application, there is provided a payment channel detection method, including the following steps:
acquiring code operating parameters of systems forming a payment channel, wherein each system at least comprises a section of code;
performing abnormal operation parameter identification on the acquired code operation parameters through a trained SVM multi-classification model, wherein the trained SVM multi-classification model comprises a plurality of SVM classifiers, and one SVM classifier is used for identifying one abnormal operation parameter in the acquired code operation parameters;
and determining the abnormal operation part of the payment channel according to the recognition result.
In an embodiment, the performing, by using the trained SVM multi-classification model, abnormal operation parameter identification on the obtained code operation parameters includes:
inputting the obtained code operation parameters into the trained feature extraction model to obtain the features of the payment channel;
and inputting the obtained features into the trained SVM multi-classification model for abnormal operation parameter recognition to obtain a recognition result.
In one embodiment, the generating of the trained feature extraction model comprises:
when each payment channel normally runs, taking the payment channel as a dimension, acquiring code running parameters of each system forming each payment channel, and forming a normal data set of each payment channel, wherein when the payment channel is taken as the dimension, the code running parameters of each system forming the same payment channel are divided into data of the same dimension;
when the mth section of code of each payment channel runs abnormally, taking the payment channel as a dimension, acquiring code running parameters of each system forming the payment channel, and forming an abnormal data set M of the payment channel, wherein M is an integer which is more than or equal to 1 and less than or equal to M, and M is the sum of the sections of the section of code of each system forming the payment channel;
merging the normal data set and the abnormal data set m of each payment channel into a training data set of each payment channel;
respectively inputting the training data set of each payment channel into a plurality of preset feature extraction models for training, and obtaining normal features and abnormal features m of each payment channel output by each preset feature extraction model, wherein the model parameters of each preset feature extraction model are different;
inputting the normal features and the abnormal features m of each payment channel, which are output by each preset feature extraction model, into a plurality of SVM classifiers to obtain a corresponding recognition result of each preset feature extraction model;
calculating the accuracy of each identification result;
and determining the preset feature extraction model corresponding to the recognition result with the highest accuracy as the trained feature extraction model.
In one embodiment, the trained SVM multi-classification model generating step includes:
combining every two of the M +1 features of the normal feature and the abnormal feature M of the same payment channel output by the trained feature extraction model to generate each group of combined features;
and respectively inputting the groups of combination characteristics into an SVM classifier for training to obtain the SVM classifier respectively corresponding to the groups of combination characteristics, and forming a trained SVM multi-classification model.
In one embodiment, the inputting the obtained features into the trained SVM multi-classification model to perform abnormal operation parameter recognition to obtain a recognition result, including:
and respectively inputting the obtained characteristics into each SVM classifier of the trained SVN model to obtain abnormal operation parameters identified by each SVM classifier.
In one embodiment, the determining the abnormal operation position of the payment channel according to the recognition result includes:
calculating the number of each abnormal operation parameter in the identification result;
determining a section of code corresponding to the most abnormal operation parameter;
and positioning the determined section of code as an abnormal operation part of the payment channel.
According to a second aspect of embodiments of the present application, there is provided a payment channel detection apparatus, including:
the system comprises a code parameter acquisition module, a code parameter acquisition module and a payment processing module, wherein the code parameter acquisition module is used for acquiring code operating parameters of each system forming a payment channel, and each system at least comprises a section of code;
the abnormal parameter identification module is used for identifying abnormal operation parameters of the acquired code operation parameters through a trained SVM multi-classification model, wherein the trained SVM multi-classification model comprises a plurality of SVM classifiers, and one SVM classifier is used for identifying one abnormal operation parameter in the acquired code operation parameters;
and the abnormal positioning module is used for determining the abnormal operation part of the payment channel according to the identification result.
In one embodiment, the anomaly parameter identification module comprises:
the characteristic acquisition module is used for inputting the acquired code operation parameters into the trained characteristic extraction model to acquire the characteristics of the payment channel;
and the recognition result acquisition module is used for inputting the acquired features into the trained SVM multi-classification model to perform abnormal operation parameter recognition to acquire a recognition result.
In one embodiment, the apparatus further comprises:
the system comprises a normal data set acquisition module, a data processing module and a data processing module, wherein the normal data set acquisition module is used for acquiring code operation parameters of each system forming each payment channel by taking the payment channel as a dimension when each payment channel operates normally to form a normal data set of each payment channel, and the code operation parameters of each system forming the same payment channel are divided into data of the same dimension when the payment channel is taken as the dimension;
the system comprises an mth data set acquisition module, a data processing module and a data processing module, wherein the mth data set acquisition module is used for acquiring code operation parameters of each system forming the payment channel by taking the payment channel as a dimension when the mth section of code of each payment channel abnormally operates, and forming an abnormal data set M of the payment channel, wherein M is an integer which is more than or equal to 1 and less than or equal to M, and M is the sum of the sections of the section of code of each system forming the payment channel;
the training data set acquisition module is used for combining the normal data set and the abnormal data set m of each payment channel into a training data set of each payment channel;
the first model training module is used for inputting the training data set of each payment channel into a plurality of preset feature extraction models respectively for training to obtain the normal feature and the abnormal feature m of each payment channel output by each preset feature extraction model, wherein the model parameters of each preset feature extraction model are different;
the characteristic input module is used for inputting the normal characteristic and the abnormal characteristic m of each payment channel, which are output by each preset characteristic extraction model, into the plurality of SVM classifiers to obtain a corresponding recognition result of each preset characteristic extraction model;
the accuracy calculation module is used for calculating the accuracy of each identification result;
and the characteristic extraction model determining module is used for determining the preset characteristic extraction model corresponding to the recognition result with the highest accuracy as the trained characteristic extraction model.
In one embodiment, the apparatus further comprises:
the characteristic combination module is used for combining the M +1 characteristics of the normal characteristic and the abnormal characteristic M of the same payment channel output by the trained characteristic extraction model in pairs to generate each group of combination characteristics;
and the second model training module is used for inputting each group of combination characteristics into the SVM classifier respectively for training to obtain the SVM classifier respectively corresponding to each group of combination characteristics, and the SVM multi-classification model after training is formed.
In one embodiment, the recognition result obtaining module includes:
and the recognition result acquisition submodule is used for respectively inputting the acquired features into each SVM classifier of the trained SVN model to obtain the abnormal operation parameters recognized by each SVM classifier.
In one embodiment, the anomaly locating module comprises:
the number calculation module is used for calculating the number of each abnormal operation parameter in the identification result;
the code determining module is used for determining a section of code corresponding to the maximum number of abnormal operation parameters;
and the abnormity positioning submodule is used for positioning the determined section of code as an abnormal operation part of the payment channel.
According to a third aspect of embodiments of the present application, there is provided a terminal, including:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to:
acquiring code operating parameters of systems forming a payment channel, wherein each system at least comprises a section of code;
performing abnormal operation parameter identification on the acquired code operation parameters through a trained SVM multi-classification model, wherein the trained SVM multi-classification model comprises a plurality of SVM classifiers, and one SVM classifier is used for identifying one abnormal operation parameter in the acquired code operation parameters;
and determining the abnormal operation part of the payment channel according to the recognition result.
By applying the method and the device, the abnormal operation parameters of the code operation parameters of each system forming the payment channel are identified through the trained SVM multi-classification model, the abnormal operation part of the payment channel can be determined according to the identification result, when the payment channel is abnormal, the abnormal operation part can be quickly and accurately positioned without manual intervention, and the operation can be simplified while the positioning speed is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
FIG. 1 is a schematic structural diagram of one embodiment of a payment channel of the present application;
FIG. 2 is a flow chart of one embodiment of a payment channel detection method of the present application;
FIG. 3 is a flow chart of another embodiment of a payment channel detection method of the present application;
FIG. 4 is a hardware configuration diagram of a terminal where the payment channel detection apparatus of the present application is located;
FIG. 5 is a block diagram of one embodiment of a payment channel detection arrangement of the present application;
fig. 6 is a block diagram of another embodiment of a payment channel detection apparatus of the present application.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present application. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.
Referring to fig. 1, fig. 1 is a schematic structural diagram of an embodiment of a payment channel of the present application:
in fig. 1, the payment channel 10 may be formed by a business system 11, a financial system 12, and a gateway system connected to each other, and is interfaced with a banking system 20, a code of the business system 11 may be divided into j +1 segments of codes by a code node 1, a code node 2. And printing a code running log at each code node, and analyzing the code running log to obtain code running parameters of the corresponding system, namely the running parameters of each section of code in the corresponding system, wherein the code running parameters are parameters capable of expressing the running characteristics of each section of code in the corresponding system, such as code running times, code running time and the like. In actual application, a unique identifier can be allocated to the code operating parameter of each section of code, so that the code operating parameters of different sections of codes can be distinguished conveniently.
In the embodiment of the present application, the service system 11 can provide various service applications to the user side to meet service requirements of different user sides, for example: customizing airline tickets for users, providing financial products to users, and the like.
The financial system 12 is an entity formed by a series of financial brokerages and financial markets that connect surplus and shortage funds parties, through which funds may flow from surplus to shortage funds through the financial system 12, such as other third party payment platforms such as a paypal system.
The gateway system 13 is a transit system between the banking system and the financial system, and may include a protocol converter, a content converter, and the like for the interaction between the banking system and the financial system.
The bank system 20 may be a one-card payment system, a credit card payment system, an online banking payment system, etc. of a certain bank, and different payment channels 10 interface different bank systems 20 when payment is made.
In other embodiments of the present application, the payment channel is composed of a financial system 12 and a gateway system 13, for example, the payment channel is connected to a one-card payment system of a bank through the gateway system, that is, the payment channel is composed of the payment system and the gateway system; the payment channel may be formed by other types of systems, and the non-limiting embodiments are not limited thereto.
In practical application, if payment transaction is carried out through the payment channel and the payment transaction fails due to the fact that the payment channel is abnormal, code operation parameters of all systems forming the payment channel can be automatically acquired, abnormal operation parameter recognition is carried out on the code operation parameters of all the systems forming the payment channel through the trained SVM multi-classification model, the abnormal operation part of the payment channel is rapidly determined according to the recognition result, and then the influence caused by the transaction failure is rapidly made up.
The embodiments of the present application will be described in detail below with reference to fig. 1.
Referring to fig. 2, it is a flowchart of an embodiment of the payment channel detection method of the present application, and the implemented method may include the following steps 201 and 203:
step 201: code operating parameters of systems forming the payment channel are acquired, wherein each system at least comprises a section of code.
In the embodiment of the present application, the payment channel may be composed of a business system 11, a financial system 12, and a gateway system as shown in fig. 1, and the code operation parameters of each system may include a code operation parameter of each piece of code of each system, and the code operation parameter may be a code operation number, a code operation time, and the like.
In practical application, the optional implementation manner of obtaining the code operating parameters of each system forming the payment channel may include: and reading code running logs printed at code nodes of each system, and analyzing code running parameters of each section of code from the read logs. The code running log can be read once every preset time interval, and the preset time interval can be set according to actual needs, such as: the preset time period may be 5 seconds in order to increase the detection frequency, and 1 minute in order to relieve the data processing pressure.
Step 202: and carrying out abnormal operation parameter identification on the acquired code operation parameters through a trained SVM multi-classification model, wherein the trained SVM multi-classification model comprises a plurality of SVM classifiers, and one SVM classifier is used for identifying one abnormal operation parameter in the acquired code operation parameters.
In the embodiment of the application, the abnormal operation parameter identification is performed on the obtained code operation parameters, that is, the abnormal operation parameters in the obtained code operation parameters are identified, where the abnormal operation parameters may be defined as parameters that do not conform to the parameter characteristics of the code in normal operation. For example: when the code operation parameter refers to the operation times of a section of code, in the whole payment process, the operation times of the code after the normal operation is finished is 100, and if the operation times of a certain section of code is 50, the operation parameter of the section of code is an abnormal operation parameter.
Generally, an SVM (Support Vector Machine) classifier performs pattern recognition, classification, and regression analysis based on a Support Vector Machine method, which is a two-classification method for converting a low-dimensional input space linear indivisible sample into a high-dimensional feature space by using a nonlinear mapping algorithm so that the linear indivisible sample can be divided. The SVM multi-class classification can decompose the multi-class problem into a series of two classes of problems which can be directly solved by the SVM, and a final judgment result is obtained based on the series of SVM solving results.
In practical application, the multi-class problem can be decomposed into two classes of problems which can be directly solved by a series of SVM in the following ways:
the first method is as follows: and constructing k two classes of classifiers (k classes are set), wherein the ith classifier divides the ith class from the rest classes, and during training, the ith classifier takes the ith class in the training set as a positive class and the rest classes as negative classes for training. During judgment, the input signals respectively pass through k classifiers to obtain k output values fi (x) ═ sgn (gi (x)), and if only one +1 (positive class) appears, the corresponding class is the input signal class; if the output is more than one +1 (more than one class claims it to belong to itself), or none of the outputs is +1 (i.e. none of the classes claims it to belong to itself), then the g (x) output values are compared and the largest corresponds to the class for which the input is the class.
The second method comprises the following steps: finding out pairwise combinations of all different classes in a training set T (k different classes), wherein the total number is P ═ k (k-1)/2, respectively using the two class sample points to form two classes of problem training sets T (i, f), and then respectively using an SVM for solving the two classes of problems to obtain P discriminant functions f (i, f) (x) ═ sgn (gi, f (x)). When the judgment is carried out, the input signal X is respectively sent to P judgment functions f (i, f) (X), if f (i, f) (X) is +1, the X is judged to be i type, the i type obtains a ticket, otherwise, the F type obtains a ticket. And respectively counting the number of votes obtained by the k categories in the P discrimination function results, wherein the category with the largest number of votes obtained is the final judgment category.
In some scenes, the number of segments of codes of each system forming the payment channel is large, and in order to reduce the number of parameters input into the trained SVM multi-classification model and the number of SVM classifiers, when the SVM multi-classification model is trained and the payment channel is actually detected, the obtained code operation parameters can be firstly subjected to feature extraction, and then the extracted features are input into the SVM multi-classification model.
In an example, the step of performing abnormal operation parameter identification on the acquired code operation parameters through the trained SVM multi-classification model may include: inputting the obtained code operation parameters into the trained feature extraction model to obtain the features of the payment channel; and inputting the obtained features into the trained SVM multi-classification model for abnormal operation parameter recognition to obtain a recognition result.
In this example, the feature extraction model may be a PCA (Principal Component Analysis) model, an ICA (Independent Component Analysis) model, a KPCA (kernel Principal Component Analysis) model, or the like, wherein the PCA model may perform a dimensionality reduction process on the high-dimensional data by a Principal Component Analysis method, remove invalid components in the data, maximally retain original features in the data, and have a variance cumulative contribution rate > of 85%.
Before payment channel detection is carried out, a feature extraction model and an SVM multi-classification model can be trained in advance, and the generation steps of the trained feature extraction model comprise:
and when the payment channels are used as the dimensionalities, the code operation parameters of the systems forming the payment channels are obtained to form a normal data set of the payment channels, wherein when the payment channels are used as the dimensionalities, the code operation parameters of the systems forming the same payment channel are divided into data with the same dimensionality.
And when the mth section of code of each payment channel abnormally runs, acquiring code running parameters of each system forming the payment channel by taking the payment channel as a dimension to form an abnormal data set M of the payment channel, wherein M is an integer which is more than or equal to 1 and less than or equal to M, and M is the sum of the sections of the section of code of each system forming the payment channel.
And combining the normal data set and the abnormal data set m of each payment channel into a training data set of each payment channel. Wherein the training data set for each payment channel comprises M anomaly data sets for each payment channel.
And respectively inputting the training data set of each payment channel into a plurality of preset feature extraction models for training, and obtaining the normal feature and the abnormal feature m of each payment channel output by each preset feature extraction model, wherein the model parameters of each preset feature extraction model are different.
And inputting the normal features and the abnormal features m of each payment channel, which are output by each preset feature extraction model, into a plurality of SVM classifiers to obtain a corresponding recognition result of each preset feature extraction model. Wherein M anomaly characteristics for each payment channel are entered.
And calculating the accuracy of each identification result.
And determining the preset feature extraction model corresponding to the recognition result with the highest accuracy as the trained feature extraction model.
Before training the feature extraction model, all section codes of all systems forming the payment channel can be numbered uniformly in advance, the number is M, M is 1, 2 and 3.
In practical application, when the training data sets are combined, the normal data set can be used as a row or a column of data of the training data set, and each abnormal data set can be used as a row or a column of data of the training data set.
When the normal features and the abnormal features m of each payment channel output by each preset feature extraction model are input into a plurality of SVM classifiers, the input can be performed in two modes of decomposing the multi-class problems into a series of two classes of problems which can be directly solved by SVM.
In addition, in some examples, in order to reduce the amount of data acquired, the code operating parameters of each system constituting the payment channel may be acquired with the payment channel as a dimension only when some section codes of the payment channel are abnormally operated, and some section codes may be codes which are executed every payment transaction.
In another example, the trained SVM multi-classification model generating step includes:
and combining the normal features and the M +1 features of the abnormal feature M of the same payment channel output by the trained feature extraction model in pairs to generate each group of combined features.
And respectively inputting the groups of combination characteristics into an SVM classifier for training to obtain the SVM classifier respectively corresponding to the groups of combination characteristics, and forming a trained SVM multi-classification model. Wherein a group of combined features corresponds to an SVM classifier.
In this example, when training the SVM classifier, the normal feature and the M +1 features of the abnormal feature M of the same payment channel are combined pairwise, that is, the normal feature, the abnormal feature 1 and the abnormal feature 2 … of the same payment channel are combined pairwise to generate each group of combined features.
In addition, when training the SVM classifier, the kernel function is selected as a Gaussian kernel function, and the parameters are generated by optimizing by using a grid search method, wherein the grid search method comprises the following steps: assuming that the value range of the parameter a is [0, 1] and the value range of the parameter b is [1, 10], equally dividing the value range of the parameter a into 10 parts and equally dividing the value range of the parameter b into 10 parts, so as to form 100 combinations of (a, b): (0, 1), (0, 2), …, (0.5, 1), (0.5, 2), …, (1, 9), (1, 10), respectively, the 100 parameter combinations are substituted into the SVM classifier, the classification result is checked, and the parameter combination with the best classification effect is selected as the value of a and b.
When the payment channel is actually detected, the obtained features are input into the trained SVM multi-classification model for identifying abnormal operation parameters, and the optional implementation mode for obtaining the identification result comprises the following steps:
and respectively inputting the obtained characteristics into each SVM classifier of the trained SVN model to obtain abnormal operation parameters identified by each SVM classifier. The obtained features need to be input into each SVM classifier.
In other embodiments, other ways may also be adopted to generate the trained SVM multi-classification model, such as a way two corresponding to the above-mentioned decomposition of the multi-class problem into a series of two classes of problems that the SVM can directly solve.
Step 203: and determining the abnormal operation part of the payment channel according to the recognition result.
In the embodiment of the application, when all the abnormal operation parameters output by the trained multi-classification model are the same, namely only one abnormal operation parameter is output, the abnormal operation parameter is searched for as the code operation parameter of the section of code, and after the section of code is searched for, the channel position where the section of code is located is determined as the abnormal operation part of the detected payment channel; when the abnormal operation parameters output by the trained multi-classification model are different, the number of the same abnormal operation parameters can be respectively counted, the code operation parameter of which code section is the most abnormal operation parameter is searched, and after the code section is searched, the channel position of the code section is determined to be the abnormal operation part of the detected payment channel.
Referring to fig. 3, a flow chart of another embodiment of the payment channel detection method of the present application is shown, where the method of this embodiment may include the following steps 301-305:
step 301: code operating parameters of systems forming the payment channel are acquired, wherein each system at least comprises a section of code.
Step 302: and carrying out abnormal operation parameter identification on the acquired code operation parameters through a trained SVM multi-classification model, wherein the trained SVM multi-classification model comprises a plurality of SVM classifiers, and one SVM classifier is used for identifying one abnormal operation parameter in the acquired code operation parameters.
Step 303: and calculating the number of each abnormal operation parameter in the identification result.
In the embodiment of the application, the number of each abnormal operation parameter can be calculated according to the pre-allocated identifier of each abnormal operation parameter, the abnormal operation parameters with the same identifier are divided into the similar parameters, and then the total number of the similar parameters is calculated.
Step 304: and determining a section of code corresponding to the abnormal operation parameter with the maximum number.
In the embodiment of the application, the identifier of the abnormal operation parameter with the largest number is the same as the identifier of the code operation parameter of a certain section of code, and the abnormal operation parameter with the largest number can be determined to correspond to the section of code.
Step 305: and positioning the determined section of code as an abnormal operation part of the payment channel.
In the embodiment of the application, the determined section of code is positioned as the abnormal operation part of the payment channel, the system abnormality of the section of code can be determined, the system maintenance can be rapidly performed, and the normal operation of the payment channel can be rapidly recovered.
If the payment channel is as shown in fig. 1, the node position of the code node dividing the determined section of code can be positioned as the abnormal operation part of the payment channel.
According to the embodiment, the abnormal operation parameter identification is carried out on the code operation parameters of each system forming the payment channel through the trained SVM multi-classification model, the abnormal operation part of the payment channel can be determined according to the identification result, when the payment channel is abnormal, the abnormal operation part can be quickly and accurately positioned without manual intervention, the positioning rate is improved, and meanwhile, the operation can be simplified.
Corresponding to the embodiment of the payment channel detection method, the application also provides an embodiment of a payment channel detection device.
The embodiment of the payment channel detection device can be applied to the terminal. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. The software implementation is taken as an example, and as a logical device, the device is formed by reading corresponding computer program instructions in the nonvolatile memory into the memory for operation through the processor of the terminal where the device is located. From a hardware aspect, as shown in fig. 4, which is a hardware structure diagram of a server where an information processing apparatus of the present application is located, except for the processor 410, the memory 420, the network interface 430, and the nonvolatile memory 440 shown in fig. 4, a terminal where the apparatus is located in the embodiment may also include other hardware according to an actual function of the terminal, which is not described again.
The processor 410 may be configured to: acquiring code operating parameters of systems forming a payment channel, wherein each system at least comprises a section of code; performing abnormal operation parameter identification on the acquired code operation parameters through a trained SVM multi-classification model, wherein the trained SVM multi-classification model comprises a plurality of SVM classifiers, and one SVM classifier is used for identifying one abnormal operation parameter in the acquired code operation parameters; and determining the abnormal operation part of the payment channel according to the recognition result.
Referring to fig. 5, a block diagram of an embodiment of a payment channel detection apparatus according to the present application may include: a code parameter acquisition module 510, an exception parameter identification module 520, and an exception location module 530.
The code parameter acquiring module 510 is configured to acquire code operating parameters of systems that form a payment channel, where each system includes at least one code segment.
An abnormal parameter recognition module 520, configured to perform abnormal operation parameter recognition on the obtained code operation parameters through a trained SVM multi-classification model, where the trained SVM multi-classification model includes multiple SVM classifiers, and one SVM classifier is used to recognize one abnormal operation parameter in the obtained code operation parameters.
And an abnormal positioning module 530, configured to determine an abnormal operation portion of the payment channel according to the identification result.
In an alternative implementation, the anomaly parameter identification module 520 may include (not shown in fig. 5):
the characteristic acquisition module is used for inputting the acquired code operation parameters into the trained characteristic extraction model to acquire the characteristics of the payment channel;
and the recognition result acquisition module is used for inputting the acquired features into the trained SVM multi-classification model to perform abnormal operation parameter recognition to acquire a recognition result.
In another optional implementation, the payment channel detection apparatus may further include (not shown in fig. 5):
the system comprises a normal data set acquisition module, a data processing module and a data processing module, wherein the normal data set acquisition module is used for acquiring code operation parameters of each system forming each payment channel by taking the payment channel as a dimension when each payment channel operates normally to form a normal data set of each payment channel, and the code operation parameters of each system forming the same payment channel are divided into data of the same dimension when the payment channel is taken as the dimension;
the system comprises an mth data set acquisition module, a data processing module and a data processing module, wherein the mth data set acquisition module is used for acquiring code operation parameters of each system forming the payment channel by taking the payment channel as a dimension when the mth section of code of each payment channel abnormally operates, and forming an abnormal data set M of the payment channel, wherein M is an integer which is more than or equal to 1 and less than or equal to M, and M is the sum of the sections of the section of code of each system forming the payment channel;
the training data set acquisition module is used for combining the normal data set and the abnormal data set m of each payment channel into a training data set of each payment channel;
the first model training module is used for inputting the training data set of each payment channel into a plurality of preset feature extraction models respectively for training to obtain the normal feature and the abnormal feature m of each payment channel output by each preset feature extraction model, wherein the model parameters of each preset feature extraction model are different;
the characteristic input module is used for inputting the normal characteristic and the abnormal characteristic m of each payment channel, which are output by each preset characteristic extraction model, into the plurality of SVM classifiers to obtain a corresponding recognition result of each preset characteristic extraction model;
the accuracy calculation module is used for calculating the accuracy of each identification result;
and the characteristic extraction model determining module is used for determining the preset characteristic extraction model corresponding to the recognition result with the highest accuracy as the trained characteristic extraction model.
In another optional implementation, the payment channel detection apparatus may further include (not shown in fig. 5):
and the characteristic combination module is used for combining the M +1 characteristics of the normal characteristic and the abnormal characteristic M of the same payment channel output by the trained characteristic extraction model in pairs to generate each group of combination characteristics.
And the second model training module is used for inputting each group of combination characteristics into the SVM classifier respectively for training to obtain the SVM classifier respectively corresponding to each group of combination characteristics, and the SVM multi-classification model after training is formed.
In another alternative implementation, the recognition result obtaining module may include (not shown in fig. 5):
and the recognition result acquisition submodule is used for respectively inputting the acquired features into each SVM classifier of the trained SVN model to obtain the abnormal operation parameters recognized by each SVM classifier.
Referring to fig. 6, a block diagram of another embodiment of a payment channel detection apparatus according to the present application is shown, where the apparatus may include: the system comprises a code parameter acquisition module 610, an abnormal parameter identification module 620 and an abnormal positioning module 630, wherein the abnormal positioning module 630 further comprises a number calculation module 631, a code determination module 632 and an abnormal positioning sub-module 633.
The code parameter acquiring module 610 is configured to acquire code operating parameters of systems that form a payment channel, where each system at least includes a segment of code.
The abnormal parameter recognition module 620 is configured to perform abnormal operation parameter recognition on the acquired code operation parameters through a trained SVM multi-classification model, where the trained SVM multi-classification model includes a plurality of SVM classifiers, and one SVM classifier is configured to recognize one abnormal operation parameter of the acquired code operation parameters.
The number calculating module 631 is configured to calculate the number of each abnormal operating parameter in the identification result.
The code determining module 632 is configured to determine a section of code corresponding to the maximum number of one abnormal operation parameter.
And the abnormal positioning sub-module 633 is used for positioning the determined code as the abnormal operation part of the payment channel.
The implementation process of the functions and actions of each module in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, wherein the modules described as separate parts may or may not be physically separate, and the parts displayed as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the application. One of ordinary skill in the art can understand and implement it without inventive effort.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.
Claims (11)
1. A payment channel detection method is characterized by comprising the following steps:
acquiring code operating parameters of systems forming a payment channel, wherein each system at least comprises a section of code;
performing abnormal operation parameter identification on the acquired code operation parameters through a trained SVM multi-classification model, wherein the trained SVM multi-classification model comprises a plurality of SVM classifiers, and one SVM classifier is used for identifying one abnormal operation parameter in the acquired code operation parameters;
determining an abnormal operation part of the payment channel according to the recognition result;
the abnormal operation parameter identification of the acquired code operation parameters through the trained SVM multi-classification model comprises the following steps:
inputting the obtained code operation parameters into the trained feature extraction model to obtain the features of the payment channel; wherein, the data set adopted in the training process of the feature extraction model comprises: normal data sets of each payment channel in normal operation and abnormal data sets of each section of codes of each payment channel in abnormal operation; the obtained features comprise normal features and abnormal features;
and inputting the obtained features into the trained SVM multi-classification model for abnormal operation parameter recognition to obtain a recognition result.
2. The method of claim 1, wherein the step of generating the trained feature extraction model comprises:
when each payment channel normally runs, taking the payment channel as a dimension, acquiring code running parameters of each system forming each payment channel, and forming a normal data set of each payment channel, wherein when the payment channel is taken as the dimension, the code running parameters of each system forming the same payment channel are divided into data of the same dimension;
when the M-th section of code of each payment channel runs abnormally, taking the payment channel as a dimension, acquiring code running parameters of each system forming the payment channel, and forming an abnormal data set with the number of M of the payment channel, wherein M is an integer which is more than or equal to 1 and less than or equal to M, and M is the sum of the sections of the section of codes of each system forming the payment channel;
merging the normal data set and M abnormal data sets of each payment channel into a training data set of each payment channel;
respectively inputting the training data set of each payment channel into a plurality of preset feature extraction models for training, and obtaining normal features and M abnormal features of each payment channel output by each preset feature extraction model, wherein the model parameters of each preset feature extraction model are different;
inputting the normal features and the M abnormal features of each payment channel output by each preset feature extraction model into a plurality of SVM classifiers to obtain a corresponding recognition result of each preset feature extraction model;
calculating the accuracy of each identification result;
and determining the preset feature extraction model corresponding to the recognition result with the highest accuracy as the trained feature extraction model.
3. The method of claim 2, wherein the trained SVM multi-classification model generating step comprises:
combining every two of the normal features and M +1 of the abnormal features of the same payment channel output by the trained feature extraction model to generate each group of combined features;
and respectively inputting the groups of combination characteristics into an SVM classifier for training to obtain the SVM classifier respectively corresponding to the groups of combination characteristics, and forming a trained SVM multi-classification model.
4. The method of claim 3, wherein the inputting the obtained features into the trained SVM multi-classification model for abnormal operation parameter recognition to obtain a recognition result comprises:
and respectively inputting the obtained characteristics into each SVM classifier of the trained SVN model to obtain abnormal operation parameters identified by each SVM classifier.
5. The method according to any one of claims 1 to 4, wherein the determining the abnormal operation part of the payment channel according to the identification result comprises:
calculating the number of each abnormal operation parameter in the identification result;
determining a section of code corresponding to the most abnormal operation parameter;
and positioning the determined section of code as an abnormal operation part of the payment channel.
6. A payment channel detection apparatus, comprising:
the system comprises a code parameter acquisition module, a code parameter acquisition module and a payment processing module, wherein the code parameter acquisition module is used for acquiring code operating parameters of each system forming a payment channel, and each system at least comprises a section of code;
the abnormal parameter identification module is used for identifying abnormal operation parameters of the acquired code operation parameters through a trained SVM multi-classification model, wherein the trained SVM multi-classification model comprises a plurality of SVM classifiers, and one SVM classifier is used for identifying one abnormal operation parameter in the acquired code operation parameters;
the abnormal positioning module is used for determining an abnormal operation part of the payment channel according to the identification result;
the abnormal parameter identification module comprises:
the characteristic acquisition module is used for inputting the acquired code operation parameters into the trained characteristic extraction model to acquire the characteristics of the payment channel; wherein, the data set adopted in the training process of the feature extraction model comprises: normal data sets of each payment channel in normal operation and abnormal data sets of each section of codes of each payment channel in abnormal operation; the obtained features comprise normal features and abnormal features;
and the recognition result acquisition module is used for inputting the acquired features into the trained SVM multi-classification model to perform abnormal operation parameter recognition to acquire a recognition result.
7. The apparatus of claim 6, further comprising:
the system comprises a normal data set acquisition module, a data processing module and a data processing module, wherein the normal data set acquisition module is used for acquiring code operation parameters of each system forming each payment channel by taking the payment channel as a dimension when each payment channel operates normally to form a normal data set of each payment channel, and the code operation parameters of each system forming the same payment channel are divided into data of the same dimension when the payment channel is taken as the dimension;
the system comprises an mth data set acquisition module, a data processing module and a data processing module, wherein the mth data set acquisition module is used for acquiring code operation parameters of each system forming the payment channel by taking the payment channel as a dimension when the mth section of code of each payment channel abnormally operates, and forming an abnormal data set M of the payment channel, wherein M is an integer which is more than or equal to 1 and less than or equal to M, and M is the sum of the sections of the section of code of each system forming the payment channel;
the training data set acquisition module is used for combining the normal data set and the abnormal data set m of each payment channel into a training data set of each payment channel;
the first model training module is used for inputting the training data set of each payment channel into a plurality of preset feature extraction models respectively for training to obtain the normal feature and the abnormal feature m of each payment channel output by each preset feature extraction model, wherein the model parameters of each preset feature extraction model are different;
the characteristic input module is used for inputting the normal characteristic and the abnormal characteristic m of each payment channel, which are output by each preset characteristic extraction model, into the plurality of SVM classifiers to obtain a corresponding recognition result of each preset characteristic extraction model;
the accuracy calculation module is used for calculating the accuracy of each identification result;
and the characteristic extraction model determining module is used for determining the preset characteristic extraction model corresponding to the recognition result with the highest accuracy as the trained characteristic extraction model.
8. The apparatus of claim 7, further comprising:
the characteristic combination module is used for combining the M +1 characteristics of the normal characteristic and the abnormal characteristic M of the same payment channel output by the trained characteristic extraction model in pairs to generate each group of combination characteristics;
and the second model training module is used for inputting each group of combination characteristics into the SVM classifier respectively for training to obtain the SVM classifier respectively corresponding to each group of combination characteristics, and the SVM multi-classification model after training is formed.
9. The apparatus of claim 8, wherein the recognition result obtaining module comprises:
and the recognition result acquisition submodule is used for respectively inputting the acquired features into each SVM classifier of the trained SVN model to obtain the abnormal operation parameters recognized by each SVM classifier.
10. The apparatus of any one of claims 6 to 9, wherein the anomaly locating module comprises:
the number calculation module is used for calculating the number of each abnormal operation parameter in the identification result;
the code determining module is used for determining a section of code corresponding to the maximum number of abnormal operation parameters;
and the abnormity positioning submodule is used for positioning the determined section of code as an abnormal operation part of the payment channel.
11. A terminal, comprising:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to:
acquiring code operating parameters of systems forming a payment channel, wherein each system at least comprises a section of code;
performing abnormal operation parameter identification on the acquired code operation parameters through a trained SVM multi-classification model, wherein the trained SVM multi-classification model comprises a plurality of SVM classifiers, and one SVM classifier is used for identifying one abnormal operation parameter in the acquired code operation parameters;
determining an abnormal operation part of the payment channel according to the recognition result;
the abnormal operation parameter identification of the acquired code operation parameters through the trained SVM multi-classification model comprises the following steps:
inputting the obtained code operation parameters into the trained feature extraction model to obtain the features of the payment channel; wherein, the data set adopted in the training process of the feature extraction model comprises: normal data sets of each payment channel in normal operation and abnormal data sets of each section of codes of each payment channel in abnormal operation; the obtained features comprise normal features and abnormal features;
and inputting the obtained features into the trained SVM multi-classification model for abnormal operation parameter recognition to obtain a recognition result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610833859.XA CN107025547B (en) | 2016-09-19 | 2016-09-19 | Payment channel detection method and device and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610833859.XA CN107025547B (en) | 2016-09-19 | 2016-09-19 | Payment channel detection method and device and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107025547A CN107025547A (en) | 2017-08-08 |
| CN107025547B true CN107025547B (en) | 2020-10-16 |
Family
ID=59525218
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610833859.XA Active CN107025547B (en) | 2016-09-19 | 2016-09-19 | Payment channel detection method and device and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107025547B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109344848A (en) * | 2018-07-13 | 2019-02-15 | 电子科技大学 | Mobile intelligent terminal security level classification method based on Adaboost |
| CN110166462B (en) * | 2019-05-25 | 2022-02-25 | 深圳市元征科技股份有限公司 | Access control method, system, electronic device and computer storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004280712A (en) * | 2003-03-18 | 2004-10-07 | Just Syst Corp | Data classification device, data classification method, and program making computer execute the method |
| CN101329638A (en) * | 2007-06-18 | 2008-12-24 | 国际商业机器公司 | Method and system for analyzing parallelism of program code |
| CN101697144A (en) * | 2009-10-28 | 2010-04-21 | 中兴通讯股份有限公司 | Software test method and software test device |
| EP2369529A1 (en) * | 2010-03-24 | 2011-09-28 | Alcatel Lucent | A method of detecting anomalies in a message exchange, corresponding computer program product, and data storage device therefor |
| CN104156562A (en) * | 2014-07-15 | 2014-11-19 | 清华大学 | Failure predication system and failure predication method for background operation and maintenance system of bank |
| CN104205111A (en) * | 2012-03-19 | 2014-12-10 | 高通股份有限公司 | Computing device to detect malware |
| CN104270275A (en) * | 2014-10-14 | 2015-01-07 | 步步高教育电子有限公司 | Auxiliary analysis method for causes of exceptions, server and intelligent equipment |
| CN105893256A (en) * | 2016-03-30 | 2016-08-24 | 西北工业大学 | Software failure positioning method based on machine learning algorithm |
-
2016
- 2016-09-19 CN CN201610833859.XA patent/CN107025547B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004280712A (en) * | 2003-03-18 | 2004-10-07 | Just Syst Corp | Data classification device, data classification method, and program making computer execute the method |
| CN101329638A (en) * | 2007-06-18 | 2008-12-24 | 国际商业机器公司 | Method and system for analyzing parallelism of program code |
| CN101697144A (en) * | 2009-10-28 | 2010-04-21 | 中兴通讯股份有限公司 | Software test method and software test device |
| EP2369529A1 (en) * | 2010-03-24 | 2011-09-28 | Alcatel Lucent | A method of detecting anomalies in a message exchange, corresponding computer program product, and data storage device therefor |
| CN104205111A (en) * | 2012-03-19 | 2014-12-10 | 高通股份有限公司 | Computing device to detect malware |
| CN104156562A (en) * | 2014-07-15 | 2014-11-19 | 清华大学 | Failure predication system and failure predication method for background operation and maintenance system of bank |
| CN104270275A (en) * | 2014-10-14 | 2015-01-07 | 步步高教育电子有限公司 | Auxiliary analysis method for causes of exceptions, server and intelligent equipment |
| CN105893256A (en) * | 2016-03-30 | 2016-08-24 | 西北工业大学 | Software failure positioning method based on machine learning algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107025547A (en) | 2017-08-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11741361B2 (en) | Machine learning-based network model building method and apparatus | |
| CN109636607B (en) | Service data processing method and device based on model deployment and computer equipment | |
| US10402163B2 (en) | Intelligent data extraction | |
| US20190087737A1 (en) | Anomaly detection and automated analysis in systems based on fully masked weighted directed | |
| US11514347B2 (en) | Identifying and remediating system anomalies through machine learning algorithms | |
| EP1883040A1 (en) | Pattern classification method | |
| KR102600745B1 (en) | Anomaly Detection Method for Sound Classification Based on Neural Network Analysis | |
| US10394631B2 (en) | Anomaly detection and automated analysis using weighted directed graphs | |
| US20230419402A1 (en) | Systems and methods of optimizing machine learning models for automated anomaly detection | |
| CN115862638B (en) | Big data safe storage method and system based on block chain | |
| CN107025547B (en) | Payment channel detection method and device and terminal | |
| CN112766427A (en) | Training method, device and equipment of image recognition model | |
| CN113159213A (en) | Service distribution method, device and equipment | |
| AU2019204930B2 (en) | Method and System for Generation of Hybrid Learning Techniques | |
| JP2021018466A (en) | Rule extracting apparatus, information processing apparatus, rule extracting method, and rule extracting program | |
| CN113468076A (en) | Application program exception testing method, device, equipment and storage medium | |
| US11954685B2 (en) | Method, apparatus and computer program for selecting a subset of training transactions from a plurality of training transactions | |
| CN113487320A (en) | Fraud transaction detection method, device, computer equipment and storage medium | |
| CN113723525A (en) | Product recommendation method, device, equipment and storage medium based on genetic algorithm | |
| CN114495224A (en) | Credit risk determination method and device | |
| CN113569879A (en) | Training method of abnormal recognition model, abnormal account recognition method and related device | |
| CN110782342B (en) | Method and device for verifying correctness of new channel feature engineering based on binary classification model | |
| US20230342402A1 (en) | Data analysis apparatus, data analysis system, and non-transitory computer-readable storage medium | |
| CN111881355B (en) | Object recommendation method and device, storage medium and processor | |
| Züfle | Dynamic Hybrid Forecasting for Self-Aware Systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200921 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20200921 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |