Payment platform transaction data processing system and method based on big data

Abstract

The invention relates to the technical field of data processing, and discloses a payment platform transaction data processing system and method based on big data. The method comprises the steps of collecting historical transaction data in a payment platform, processing the historical transaction data in a data filtering mode, analyzing the obtained filtered historical transaction data, and clustering the filtered historical transaction data according to an analysis result to obtain a clustering result; performing mutual exclusion analysis on the obtained clustered historical transaction data, detecting whether there is an inconsistent condition of the historical transaction data, and processing the inconsistent condition of the historical transaction data; and meanwhile, intrusion detection is performed on the obtained analyzed historical transaction data through the constructed payment intrusion detection module based on the analyzed historical transaction data, and classification processing is performed according to the detection result, so that the pertinence of payment platform transaction data processing is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and specifically to a payment platform transaction data processing system and method based on big data. Background Art

[0002] With the continuous development of the electronic, digital, and networked processes in the commodity trading process, a large amount of transaction payment information and network attack information have been generated on the payment transaction platform, imposing a huge burden and interference on the security of payment platform transaction data.

[0003] The existing publicly applied patent CN108469967A determines the transaction code corresponding to the transaction data according to the received transaction data; determines the memory address corresponding to the transaction code from the transaction array; and calls the dynamic library file pointed to by the memory address corresponding to the transaction code to process the transaction data according to the memory address corresponding to the transaction code. It can be seen that after each transaction data is received, the transaction code can be determined according to the transaction data, then the corresponding memory address can be determined by the transaction code, and finally the dynamic library file pointed to by the memory address corresponding to the transaction code can be called to process the transaction data. Summary of the Invention

[0004] (I) Technical Problems to be Solved Aiming at the deficiencies of the prior art, the present invention provides a payment platform transaction data processing system and method based on big data, which has the advantages of security, accuracy, and efficiency, and solves the problem of heavy security burden of payment platform transaction data.

[0005] (II) Technical Solutions To solve the above technical problem of heavy security burden of payment platform transaction data, the present invention provides the following technical solutions: The present invention discloses a payment platform transaction data processing method based on big data, which specifically includes the following steps: S1. Collect historical transaction data in the payment platform, and process the historical transaction data through a data filtering method to obtain filtered historical transaction data; S2. Based on the analysis of the obtained filtered historical transaction data, cluster the filtered historical transaction data according to the analysis results and output the clustering results. The clustering results include: historical transaction data types, historical transaction data, and historical transaction data centers; S3. Conduct mutual exclusion analysis on the clustered historical transaction data, detect whether there is an inconsistency in the historical transaction data, process the inconsistency in the historical transaction data, and output the analyzed historical transaction data; S4. Construct a payment intrusion detection module, and perform intrusion detection on the analyzed historical transaction data obtained through the constructed payment intrusion detection module to obtain the detected historical transaction data; S5. Set a data processing method to process the detected historical transaction data obtained.

[0006] The present invention collects historical transaction data in a payment platform, processes the historical transaction data through a data filtering method, and at the same time, based on the analysis of the filtered historical transaction data obtained, clusters the filtered historical transaction data according to the analysis results, and performs mutually exclusive analysis on the clustered historical transaction data to detect whether there is inconsistency in the historical transaction data and process the inconsistency in the historical transaction data; at the same time, based on the analyzed historical transaction data, perform intrusion detection on the analyzed historical transaction data through the constructed payment intrusion detection module, and classify and process according to the detection results, improving the pertinence of the processing of transaction data in the payment platform.

[0007] Preferably, the step of collecting historical transaction data in a payment platform and processing the historical transaction data through a data filtering method to obtain the filtered historical transaction data includes the following steps: S11. Filter the collected historical transaction data; Traverse the collected historical transaction data, and set that the complete historical transaction data includes: the addresses of both trading parties, the order number, the transaction amount, the transaction time, and the transaction method; Identify the missing content in the historical transaction data, and delete the historical transaction data with missing content: S12. Perform secondary filtering on the filtered historical transaction data through a filtering algorithm.

[0008] Preferably, the step of performing secondary filtering on the filtered historical transaction data through a filtering algorithm includes the following steps: Establish an array with a length of , select hash functions to calculate each data in the historical transaction data, and save the calculation results in the array, where represents the number of hash functions, and represents the number of data in the collected historical transaction data; During the process of calculating the historical transaction data by the hash function, when the calculation results of two groups of data are the same, compare each data in these two groups of data; When the comparison results are the same, set the current two groups of data as the same data; When the two groups of data are the same data, compare the collection times of the two groups of data, and keep the most recent group of data; After the calculation is completed, the historical transaction data after calculation is summarized to obtain the filtered historical transaction data.

[0009] The present invention preliminarily filters the collected historical transaction data by traversing the collected historical transaction data and identifying and deleting the missing content in the historical transaction data. At the same time, through the filter algorithm, the preliminarily filtered historical transaction data is filtered again, improving the accuracy of historical transaction data processing.

[0010] Preferably, the clustering of the filtered historical transaction data based on the analysis of the obtained filtered historical transaction data includes the following steps according to the analysis results: S21. Initialize the collected historical transaction data and construct a historical transaction data set ={ , ,..., }; Among them, represents the th historical transaction data, represents the first historical transaction data; Select historical transaction data in the historical transaction data set as the initial historical transaction data center ={ , ,..., }, and set each historical transaction data center as the representative data of a historical transaction data type; S22. Cluster the collected historical transaction data; Based on the historical transaction data in the historical transaction data set, according to the distance formula, calculate the distance from the historical transaction data to historical transaction data centers, and assign it to the class corresponding to the historical transaction data center with the smallest distance; Set to add 1 each time. After all the historical transaction data clustering is completed, then use the result of this classification as the input of step S23, and then execute step S23; Among them, represents the distance from the historical transaction data to the historical transaction data center , represents the th historical transaction data center, set to have a value range of [1, ; S23. Update the historical transaction data center; S24. Repeat steps S22 - S23 until the historical transaction data in the historical transaction data center and the category no longer changes, and then output the historical transaction data type, historical transaction data, and historical transaction data center.

[0011] Preferably, the updating of the historical transaction data center includes the following steps: After all historical transaction data is classified, recalculate the historical transaction data center for each category based on the current classification result ; where represents the distance from the r-th historical transaction data in the category to the current historical transaction data center, represents the number of historical transaction data in the current category; Based on the recalculated historical transaction data center for each category, recalculate the historical transaction data in each category.

[0012] The present invention initializes the collected historical transaction data, selects the initial clustering center for clustering, and continuously updates the clustering center through iterative calculation, and updates the historical transaction data in the corresponding category until convergence, improving the rationality of transaction data processing.

[0013] Preferably, the mutually exclusive analysis of the clustered historical transaction data, detecting whether there is inconsistent historical transaction data, processing the inconsistent historical transaction data, and outputting the analyzed historical transaction data include the following steps: S31. Traverse the historical transaction data in each type of historical transaction data after clustering respectively, and determine whether there is an inconsistent situation between the transaction amount and the corresponding amount transaction quantity; S32. When it is detected that there is an inconsistent situation between the transaction amount and the corresponding amount transaction quantity in the historical transaction data of the historical transaction data type, set a transaction lock for each transaction item, and set that only a group of transaction data is allowed to access each transaction item during the transaction process; S33. Recalculate the historical transaction data based on the set transaction lock, and summarize the calculation results to obtain the analyzed historical transaction data.

[0014] The present invention traverses the historical transaction data in each type of historical transaction data after clustering, summarizes and compares the transaction amount and the corresponding amount transaction quantity, and processes and restricts the abnormal transaction amount and the corresponding amount transaction quantity, improving the security of payment transactions.

[0015] Preferably, the construction of the payment intrusion detection module, and performing intrusion detection on the obtained analyzed historical transaction data through the constructed payment intrusion detection module to obtain the detected historical transaction data includes the following steps: S41. Construct a payment intrusion detection module; S42. Based on the constructed payment intrusion detection module, perform intrusion detection on the analyzed historical transaction data and output the detected historical transaction data.

[0016] Preferably, the steps for constructing the payment intrusion detection module are as follows: S411. Collect four types of payment transaction data to construct a training set, and extract features from the four types of payment transaction data in the training set; The four types of payment transaction data include: flood attack payment transaction data, simulation attack payment transaction data, injection attack payment transaction data, and normal payment transaction data; Preprocess the four types of payment transaction data in the training set; Since some features of the four types of payment transaction data in the training set have real values and some features have nominal attributes, before extracting features from the four types of payment transaction data, set all nominal attributes to integer values, replace missing values with zeros, and replace deleted elements with redundant information; Summarize the preprocessed four types of payment transaction data, and input the preprocessed four types of payment transaction data into a convolutional neural network for feature extraction; After receiving the preprocessed four types of payment transaction data, the convolutional neural network moves the convolutional kernel in the input preprocessed four types of payment transaction data according to the set stride, and performs convolutional calculation with the eigenvalue of the corresponding area at each step, so as to realize the feature extraction of the preprocessed four types of payment transaction data; The convolutional calculation formula is as follows: ; where, represents the preprocessed four types of payment transaction data of the z-th group of inputs, represents the weight value of the corresponding convolutional kernel, b represents the bias value, the output feature of the z-th group of preprocessed four types of payment transaction data; S412. Train the features extracted by the convolutional neural network; Use the features extracted by the convolutional neural network as the input of the BP neural network; The structure of the BP neural network includes: an input layer, a hidden layer, and an output layer; The learning algorithm of the BP neural network includes: Set the input of the input layer as where represents the first feature of the input, represents the m-th feature of the input, and the output of the output layer is where Represents the feature of the first prediction of the output, Represents the feature of the s-th prediction of the output. The hidden layer contains q neurons, and v is the weight from the input layer to the hidden layer. Is the weight from the hidden layer to the output layer; The training process of the BP neural network includes two stages: forward propagation of signals and backward propagation of errors; The forward propagation of signals is that data is input from the input layer, passes through the hidden layer, and finally reaches the output layer; The calculation formula for the forward propagation of signals is: ; ; Among them, Represents the input of the h-th hidden layer neuron, Represents the weight from the i-th input of the input layer to the h-th hidden layer neuron, and m represents the number of input data in the input layer; Represents the i-th input of the input layer; Represents the bias value of the h-th hidden layer neuron; Represents the input of the j-th output layer neuron, Represents the weight from the h-th hidden layer neuron to the j-th output layer neuron, and q represents the number of hidden layer neurons; Represents the bias value of the j-th output layer neuron, Represents the output of the j-th output layer neuron; The stage of backward propagation of errors: Calculate the error between the output layer and the expected value, set the error threshold between the output layer and the expected value. When the error between the output layer and the expected value is greater than or equal to the set error threshold, weight adjustment is performed through the stage of backward propagation of errors; The error calculation formula is as follows: ; Among them, E is the error, Is the expected value of the j-th output, and n is the number of output neurons; The calculation formula for weight adjustment is as follows: ; Among them, Represents the weight adjustment value, l Represents the learning rate, E represents the error, and Y represents the output of the output layer; The weights are adjusted by updating the current weights with the weight adjustment value, and the adjusted weights are used as the new weights for iterative calculation during the forward propagation calculation of signals until a definite BP neural network model is obtained; Set the payment intrusion detection module as the combination of a BP neural network model and a convolutional neural network, where the convolutional neural network is the front end of the payment intrusion detection module and the BP neural network model is the back end of the payment intrusion detection module; S42. Based on the constructed payment intrusion detection module, perform intrusion detection on the analyzed historical transaction data and output the detected historical transaction data; Input the analyzed historical transaction data into the constructed payment intrusion detection module, extract the features of the analyzed historical transaction data, detect the features of the analyzed historical transaction data through the BP neural network model in the payment intrusion detection module, judge the type of payment transaction data to which the analyzed historical transaction data belongs, and output the detected historical transaction data.

[0017] The present invention constructs a payment intrusion detection module by collecting four types of payment transaction data and performing feature extraction and iterative training on the collected four types of payment transaction data. At the same time, based on the constructed payment intrusion detection module, perform intrusion detection on the analyzed historical transaction data, and judge the attacked status of the payment transaction data in real time, improving the security of payment transactions.

[0018] Preferably, the processing of the obtained detected historical transaction data by the set data processing method includes the following steps: Process according to the type of payment transaction data to which the historical transaction data belongs; Set the access permission of the transaction data for the flood attack payment transaction data to isolate the flood attack; Set private key encryption for the simulated attack payment transaction data and detect the source address of the payment transaction; Set the transaction user verification method for the injection attack payment transaction data to isolate the injection attack.

[0019] The present invention also discloses a payment platform transaction data processing system based on big data for implementing the payment platform transaction data processing method based on big data. The system includes: a data collection module, a data processing module, a mutual exclusion analysis module, a clustering module, a payment intrusion detection module, and a transaction data processing module; The data collection module is used to collect historical transaction data in the payment platform; The data processing module is used to process the historical transaction data by means of data filtering; The clustering module is used to cluster the processed historical transaction data by means of clustering; The mutual exclusion analysis module is used to perform mutual exclusion analysis on the clustered historical transaction data; The payment intrusion detection module is used to perform payment intrusion detection on the historical transaction data after mutual exclusion analysis; The said transaction data processing module is used for special processing according to the payment intrusion detection result.

[0020] (III) Beneficial effects: Compared with the prior art, the present invention provides a payment platform transaction data processing system and method based on big data, having the following beneficial effects: 1. The invention collects historical transaction data in the payment platform, processes the historical transaction data through data filtering, and analyzes the obtained filtered historical transaction data. According to the analysis result, the filtered historical transaction data is clustered, and mutually exclusive analysis is performed on the clustered historical transaction data to detect whether there is inconsistency in the historical transaction data and process the inconsistent situation of the historical transaction data. At the same time, based on the analyzed historical transaction data, intrusion detection is performed on the analyzed historical transaction data through the constructed payment intrusion detection module, and classification processing is performed according to the detection result, improving the pertinence of payment platform transaction data processing.

[0021] 2. The invention initially filters the collected historical transaction data by traversing the collected historical transaction data and identifying and deleting the missing content in the historical transaction data. At the same time, through the filter algorithm, the initially filtered historical transaction data is filtered again, improving the accuracy of historical transaction data processing.

[0022] 3. The invention initializes the collected historical transaction data, selects the initial clustering center for clustering, and continuously updates the clustering center through iterative calculation, and updates the historical transaction data in the corresponding category until convergence, improving the rationality of transaction data processing.

[0023] 4. The invention traverses the historical transaction data in each type of historical transaction data after clustering, summarizes and compares the transaction amount and the corresponding amount transaction quantity, and processes and restricts the abnormal transaction amount and the corresponding amount transaction quantity, improving the security of payment transactions.

[0024] 5. The invention constructs a payment intrusion detection module by collecting four types of payment transaction data, extracting features from the collected four types of payment transaction data, and performing iterative training. At the same time, based on the constructed payment intrusion detection module, intrusion detection is performed on the analyzed historical transaction data to real-time judge the attacked situation of payment transaction data, improving the security of payment transactions. Description of the drawings

[0025] Figure 1 It is a schematic structural diagram of the payment platform transaction data processing flow of the present invention. Detailed implementation manners

[0026] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0027] Embodiment 1: Please refer to Figure 1 , this embodiment discloses a method for processing transaction data of a payment platform based on big data, which specifically includes the following steps: S1. Collect historical transaction data in the payment platform, and process the historical transaction data through a data filtering method to obtain filtered historical transaction data; S2. Based on the analysis of the obtained filtered historical transaction data, cluster the filtered historical transaction data according to the analysis results, and output the clustering results, where the clustering results include: historical transaction data types, historical transaction data, and historical transaction data centers; S3. Perform mutual exclusion analysis on the clustered historical transaction data, detect whether there is inconsistent historical transaction data, process the inconsistent historical transaction data, and output the analyzed historical transaction data; S4. Build a payment intrusion detection module, and perform intrusion detection on the obtained analyzed historical transaction data through the built payment intrusion detection module to obtain detected historical transaction data; S5. Set a data processing method to process the obtained detected historical transaction data; Further, please refer to Figure 1 , collecting historical transaction data in the payment platform, and processing the historical transaction data through a data filtering method to obtain filtered historical transaction data includes the following steps: S11. Filter the collected historical transaction data; Traverse the collected historical transaction data. It is assumed that the complete historical transaction data includes: the addresses of both parties to the transaction, order numbers, transaction amounts, transaction times, and transaction methods; Identify the missing content in the historical transaction data, and delete the historical transaction data with missing content: S12. Perform secondary filtering on the filtered historical transaction data through a filtering algorithm; Establish an array with a length of , select hash functions to calculate each data in the historical transaction data, and save the calculation results in the array, where represents the number of hash functions, represents the number of data in the collected historical transaction data; During the calculation of the historical transaction data by the hash function, when the calculation results of two groups of data are the same, each data in these two groups of data is compared; When the comparison results are the same, set the current two groups of data as the same data; When the two groups of data are the same data, compare the collection times of the two groups of data and retain the most recent group of data; After the calculation is completed, summarize the calculated historical transaction data to obtain the filtered historical transaction data; Further, please refer to Figure 1 , based on the analysis of the obtained filtered historical transaction data, cluster the filtered historical transaction data according to the analysis results, and output the clustering results. The clustering results include: historical transaction data types, historical transaction data, and the historical transaction data center, including the following steps: S21. Initialize the collected historical transaction data and construct a historical transaction data set ={ , ,..., }; Among them, represents the th historical transaction data, represents the first historical transaction data; Select historical transaction data from the historical transaction data set as the initial historical transaction data center ={ , ,..., }, and set each historical transaction data center as the representative data of a historical transaction data type; S22. Cluster the collected historical transaction data; Based on the historical transaction data in the historical transaction data set, according to the distance formula, calculate the distance from the historical transaction data to the historical transaction data centers, and classify it into the class corresponding to the historical transaction data center with the smallest distance; Further, set to add 1 each time. After all the historical transaction data clustering is completed, then use the classification result of this time as the input of step S23, and then execute step S23; Among them, represents the distance from the historical transaction data to the historical transaction data center , represents the a historical transaction data center is set The value range is [1, ; S23. Update the historical transaction data center; After all historical transaction data are classified, recalculate the historical transaction data center for each category based on the current classification result ; Among them, represents the distance from the r-th historical transaction data in the category to the current historical transaction data center, represents the number of historical transaction data in the current category; Based on the recalculated historical transaction data center for each category, recalculate the historical transaction data in each category; S24. Repeat steps S22 - S23 until the historical transaction data center and the historical transaction data in the category no longer change, and then output the historical transaction data type, historical transaction data, and historical transaction data center; Furthermore, please refer to Figure 1 , perform mutual exclusion analysis on the clustered historical transaction data, detect whether there is inconsistent historical transaction data, process the inconsistent historical transaction data, and output the analyzed historical transaction data, including the following steps: S31. Traverse the historical transaction data in each historical transaction data type after clustering respectively, and determine whether there is an inconsistent situation between the transaction amount and the corresponding number of transactions with that amount; S32. When it is detected that there is an inconsistent situation between the transaction amount and the corresponding number of transactions with that amount in the historical transaction data of a historical transaction data type, set a transaction lock for each transaction item, and set that only one set of transaction data is allowed to access each transaction item during the transaction process; S33. Recalculate the historical transaction data based on the set transaction lock, and summarize the calculation results to obtain the analyzed historical transaction data; Furthermore, please refer to Figure 1 , construct a payment intrusion detection module, and perform intrusion detection on the obtained analyzed historical transaction data through the constructed payment intrusion detection module to obtain the detected historical transaction data, including the following steps: S41. Construct a payment intrusion detection module; S411. Collect four types of payment transaction data to construct a training set, and extract features from the four types of payment transaction data in the training set; The four types of payment transaction data include: flood attack payment transaction data, simulation attack payment transaction data, injection attack payment transaction data, and normal payment transaction data; Furthermore, preprocess the four types of payment transaction data in the training set; Since some features of the four types of payment transaction data in the training set have real values and some features have nominal attributes, before feature extraction of the four types of payment transaction data, all nominal attributes are set to be converted into integer values, missing values are replaced with zeros, and deleted elements are replaced with redundant information; Summarize the four types of payment transaction data after preprocessing, and input the four types of payment transaction data after preprocessing into a convolutional neural network for feature extraction; After receiving the four types of payment transaction data after preprocessing, the convolutional neural network moves through the input four types of payment transaction data after preprocessing with a convolutional kernel according to the set stride, and performs a convolutional calculation with the feature values in the corresponding area at each step, so as to achieve feature extraction of the four types of payment transaction data after preprocessing; The convolutional calculation formula is as follows: ; Among them, represents the four types of payment transaction data after preprocessing of the z-th group of inputs, represents the weights of the corresponding convolutional kernel, b represents the bias value, the output features of the four types of payment transaction data after preprocessing of the z-th group; S412. Train the features extracted by the convolutional neural network; Use the features extracted by the convolutional neural network as the input of the BP neural network; The structure of the BP neural network includes: an input layer, a hidden layer, and an output layer; The learning algorithm of the BP neural network includes: Set the input of the input layer to be , where represents the first feature of the input, represents the m-th feature of the input, and the output of the output layer is , where represents the first predicted feature of the output, represents the s-th predicted feature of the output, the hidden layer contains q neurons, v is the weight from the input layer to the hidden layer, is the weight from the hidden layer to the output layer; The training process of the BP neural network includes two stages: forward propagation of signals and backward propagation of errors; The forward propagation of signals is that data is input from the input layer, passes through the hidden layer, and finally reaches the output layer; The calculation formula for the forward propagation of signals is: ; ; Among them, represents the input of the h-th hidden layer neuron, represents the weight from the i-th input of the input layer to the h-th hidden layer neuron, and m represents the number of input data in the input layer; represents the i-th input of the input layer; represents the bias value of the h-th hidden layer neuron; represents the input of the j-th output layer neuron, represents the weight from the h-th hidden layer neuron to the j-th output layer neuron, and q represents the number of hidden layer neurons; represents the bias value of the j-th output layer neuron, represents the output of the j-th output layer neuron; Backpropagation stage of the error: Calculate the error between the output layer and the expected value, set the error threshold between the output layer and the expected value. When the error between the output layer and the expected value is greater than or equal to the set error threshold, weight adjustment is performed through the backpropagation stage of the error; The error calculation formula is as follows: ; where E is the error, is the expected value of the j-th output, and n is the number of output neurons; The calculation formula for weight adjustment is as follows: ; where, represents the weight adjustment value, l represents the learning rate, E represents the error, and Y represents the output of the output layer; Adjust the weights by updating the current weights with the weight adjustment value, and use the adjusted weights as the new weights for iterative calculation during the forward propagation of the signal until a definite BP neural network model is obtained; Furthermore, set the payment intrusion detection module as the combination of a BP neural network model and a convolutional neural network, where the convolutional neural network is at the front end of the payment intrusion detection module and the BP neural network model is at the back end of the payment intrusion detection module; S42. Perform intrusion detection on the analyzed historical transaction data based on the constructed payment intrusion detection module, and output the detected historical transaction data; Input the analyzed historical transaction data into the constructed payment intrusion detection module, extract the features of the analyzed historical transaction data, detect the features of the analyzed historical transaction data through the BP neural network model in the payment intrusion detection module, determine the type of payment transaction data to which the analyzed historical transaction data belongs, and output the detected historical transaction data; Furthermore, please refer toFigure 1 Set the data processing method to process the detected historical transaction data, including the following steps: Process according to the types of payment transaction data belonging to the historical transaction data respectively; Set the transaction data access permission for the flood attack payment transaction data to isolate the flood attack; Set the private key encryption for the simulated attack payment transaction data and detect the source address of the payment transaction; Set the transaction user authentication method for the injection attack payment transaction data to isolate the injection attack; Embodiment 2: Please refer to Figure 1 This embodiment also discloses a payment platform transaction data processing system based on big data for implementing the payment platform transaction data processing method based on big data. The system includes: a data collection module, a data processing module, a mutual exclusion analysis module, a clustering module, a payment intrusion detection module, and a transaction data processing module; The data collection module is used to collect historical transaction data in the payment platform; The data processing module is used to process the historical transaction data through data filtering; The clustering module is used to cluster the processed historical transaction data through clustering; The mutual exclusion analysis module is used to perform mutual exclusion analysis on the clustered historical transaction data; The payment intrusion detection module is used to perform payment intrusion detection on the historical transaction data after mutual exclusion analysis; The transaction data processing module is used to perform special processing according to the payment intrusion detection results.

[0028] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A payment platform transaction data processing method based on big data, characterized in that: The following steps are involved: S1. Collect historical transaction data in the payment platform, and process the historical transaction data by data filtering to obtain filtered historical transaction data; S2. Based on analyzing the filtered historical transaction data, clustering the filtered historical transaction data according to the analysis result, and outputting the clustering result, the clustering result includes: historical transaction data type, historical transaction data, and historical transaction data center; S3, performing mutual exclusion analysis on the clustered historical transaction data, detecting whether there is inconsistency in the historical transaction data, processing the inconsistency in the historical transaction data, and outputting the analyzed historical transaction data; S4. Construct a payment intrusion detection module, and perform intrusion detection on the analyzed historical transaction data through the constructed payment intrusion detection module to obtain the detected historical transaction data; S5. Set a data processing method to process the detected historical transaction data.

2. A payment platform transaction data processing method based on big data according to claim 1, characterized in that: The collecting of historical transaction data in the payment platform and processing the historical transaction data by means of data filtering to obtain filtered historical transaction data comprises the following steps: S11, filtering the collected historical transaction data; Traverse the collected historical transaction data and set the complete historical transaction data including: addresses of both parties to the transaction, order number, transaction amount, transaction time and transaction method; Identify missing content in historical transaction data and delete the historical transaction data with missing content: S12. Perform secondary filtering on the filtered historical transaction data through a filtering algorithm.

3. A payment platform transaction data processing method based on big data according to claim 2, characterized in that: The secondary filtering of the filtered historical transaction data by the filtering algorithm comprises the following steps: Create a length of Array, select A hash function is used to calculate each data in the historical transaction data, and the calculation results are saved in an array, where: represents the number of hash functions, Indicates the amount of data in the collected historical transaction data; When the hash function calculates the historical transaction data, if the calculation results of two sets of data are consistent, each data in the two sets of data is compared; When the comparison results are consistent, the two sets of data are set to be the same data; When two sets of data are the same, compare the collection time of the two sets of data and keep the most recent set of data; After the calculation is completed, the calculated historical transaction data is summarized to obtain the filtered historical transaction data.

4. A payment platform transaction data processing method based on big data according to claim 1, characterized in that: The method of clustering the filtered historical transaction data based on the analysis of the filtered historical transaction data comprises the following steps: S21. Initialize the collected historical transaction data and build a historical transaction data set ={ , , ..., }; in, Indicates Historical transaction data, Indicates the first historical transaction data; Select Historical Transaction Data Center Historical transaction data as the initial historical transaction data center ={ , , ..., }, set each historical transaction data center as the representative data of a historical transaction data type; S22, clustering the collected historical transaction data; Based on historical transaction data in the historical transaction dataset , according to the distance formula, Calculate historical transaction data arrive The distance between the historical transaction data centers is calculated, and the data is divided into the class corresponding to the historical transaction data center with the smallest distance; Set to add 1 each time calculation until all historical transaction data are clustered, then use the result of this classification as the input of step S23, and then execute step S23; Set to add 1 each time calculation until all historical transaction data are clustered, then use the result of this classification as the input of step S23, and then execute step S23; in, Represents historical transaction data Go to the Historical Trading Data Center The distance Indicates Historical transaction data center, set The value range is [1, ]; S23, updating the historical transaction data center; S24. Repeat steps S22-S23 until the historical transaction data center type and the historical transaction data in the category no longer change, and then output the historical transaction data type, historical transaction data, and historical transaction data center.

5. A payment platform transaction data processing method based on big data according to claim 4, characterized in that: The updating of the historical transaction data center comprises the following steps: When all historical transaction data are classified, recalculate the historical transaction data center of each category based on the current classification results ; in, Indicates the distance from the rth historical transaction data in the category to the current historical transaction data center. Indicates the number of historical transaction data in the current category; Based on recalculating the historical transaction data center of each category, the historical transaction data in each category is recalculated.

6. A payment platform transaction data processing method based on big data according to claim 1, characterized in that: The method of performing mutual exclusive analysis on the clustered historical transaction data, detecting whether there is inconsistency in the historical transaction data, processing the inconsistency in the historical transaction data, and outputting the analyzed historical transaction data comprises the following steps: S31, traversing the historical transaction data in each historical transaction data type after clustering, and determining whether there is inconsistency between the transaction amount and the corresponding transaction quantity; S32. When it is detected that the historical transaction data in the historical transaction data type contains a transaction amount and a corresponding transaction quantity, a transaction lock is set for each transaction item, and each transaction item is set to allow only one set of transaction data to be accessed during the transaction process; S33. Recalculate the historical transaction data based on the set transaction lock, and summarize the calculation results to obtain the analyzed historical transaction data.

7. A payment platform transaction data processing method based on big data according to claim 1, characterized in that: The construction of the payment intrusion detection module, performing intrusion detection on the analyzed historical transaction data through the constructed payment intrusion detection module, and obtaining the detected historical transaction data includes the following steps: S41, constructing a payment intrusion detection module; S42: Perform intrusion detection on the analyzed historical transaction data based on the constructed payment intrusion detection module, and output the detected historical transaction data.

8. A payment platform transaction data processing method based on big data according to claim 7, characterized in that: The construction of the payment intrusion detection module includes the following steps: S411, collecting four types of payment transaction data to construct a training set, and performing feature extraction on the four types of payment transaction data in the training set; Preprocess the four types of payment transaction data in the training set; Summarize the preprocessed four types of payment transaction data, and input the preprocessed four types of payment transaction data into the convolutional neural network for feature extraction; After receiving the preprocessed four types of payment transaction data, the convolutional neural network moves in the input preprocessed four types of payment transaction data according to the set step size through the convolution kernel, and performs convolution calculation with the feature value of the area on the corresponding area of ​​each step, so as to realize feature extraction of the preprocessed four types of payment transaction data; S412, training the features extracted by the convolutional neural network; The features extracted by the convolutional neural network are used as the input of the BP neural network; The training process of BP neural network includes two stages: forward propagation of signal and back propagation of error; The calculation formula for the forward propagation of the signal is: ； ； in, represents the input of the hth hidden layer neuron, represents the weight of the i-th input of the input layer to the h-th hidden layer neuron, and m represents the number of input data of the input layer; Represents the i-th input of the input layer; Represents the bias value of the hth hidden layer neuron; represents the input of the jth output layer neuron, represents the weight from the hth hidden layer neuron to the jth output layer neuron, and q represents the number of hidden layer neurons; represents the bias value of the jth output layer neuron, represents the output of the jth output layer neuron; Back propagation phase of error: Calculate the error between the output layer and the expected value, set the error threshold between the output layer and the expected value, and when the error between the output layer and the expected value is greater than or equal to the set error threshold, perform weight adjustment through the error back propagation stage; The weights are adjusted by updating the current weights with weight adjustment values, and the adjusted weights are used as new weights for forward propagation calculation of the signal for iterative calculation until a determined BP neural network model is obtained; The payment intrusion detection module is set as a combination of a BP neural network model and a convolutional neural network, where the convolutional neural network is the front end of the payment intrusion detection module and the BP neural network model is the back end of the payment intrusion detection module.

9. A system for implementing the payment platform transaction data processing method based on big data according to any one of claims 1 to 8, characterized in that: include: Data collection module, data processing module, mutual exclusion analysis module, clustering module, payment intrusion detection module and transaction data processing module; The data collection module is used to collect historical transaction data in the payment platform; The data processing module is used to process historical transaction data by means of data filtering; The clustering module is used to cluster the processed historical transaction data by clustering; The mutual exclusion analysis module is used to perform mutual exclusion analysis on the clustered historical transaction data; The payment intrusion detection module is used to perform payment intrusion detection on the historical transaction data after mutual exclusion analysis; The transaction data processing module is used to perform special processing according to the payment intrusion detection result.

Images