One-card transaction security management method, system and storage medium

Abstract

The present application discloses a one-card transaction security management method, system and storage medium, which extracts various information items from transaction data generated by users in the one-card transaction process, and analyzes these information items through data processing and semantic understanding algorithms based on artificial intelligence and natural language processing technology, so as to extract the embedded semantic deep essential correlation characteristics of these transaction data information items, thereby realizing data type identification of these transaction data information items, so as to identify which data is sensitive data, and apply different encryption strategies to sensitive data of different security levels, and can adaptively encrypt and manage the information items in the one-card transaction data in an automated manner, thereby realizing more intelligent one-card transaction security management, helping to improve system security and user experience, and enhancing users' trust and support for the one-card transaction security management system.

Description

Technical Field

[0001] The present application relates to the field of intelligent management, and more specifically, to a method, system and storage medium for secure management of transactions via a one-card system. Background Art

[0002] With the rapid development and widespread application of information technology, the one-card system, as a comprehensive service platform integrating identity authentication, payment, access control management and other functions, has been widely used in schools, enterprises, communities and other fields. However, the popularity of the one-card system has also brought new security challenges, especially the security protection of user personal information and transaction data has become increasingly prominent. In this context, it is particularly important to build an efficient and secure one-card transaction security management system.

[0003] At present, traditional one-card security management systems often do not classify data in detail. All data may be treated equally without distinguishing between sensitive data and non-sensitive data. This practice leads to a waste of resources because the same protection measures are taken for all data, increasing the complexity and cost of the system. At the same time, it may also lead to insufficient protection of sensitive data, while non-sensitive data may be overprotected, affecting the user's sense of use. In addition, some other traditional one-card security management systems usually rely on simple rule matching or keyword filtering to identify sensitive information, which is limited in effectiveness when dealing with complex and diverse data. Rule matching and keyword filtering are prone to missed detection or false detection, resulting in sensitive information not being correctly identified and protected, thereby increasing the risk of data leakage.

[0004] Therefore, an optimized one-card transaction security management method and system are used to solve the above technical problems. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a one-card transaction security management method, system and storage medium, which extracts each information item in the transaction data generated by the user in the one-card transaction process, and analyzes these information items through data processing and semantic understanding algorithms based on artificial intelligence and natural language processing technology, so as to extract the embedded semantic deep essential association features of these transaction data information items, thereby realizing data type identification of these transaction data information items, so as to identify which data is sensitive data, and apply different encryption strategies to sensitive data of different security levels, and can perform adaptive encryption management on the information items in the one-card transaction data in an automated manner, thereby realizing more intelligent one-card transaction security management, helping to improve system security and user experience, and enhancing users' trust and support for the one-card transaction security management system.

[0006] According to one aspect of the present application, a one-card transaction security management system is provided, which includes:

[0007] One-card transaction data collection module, used to obtain transaction data generated by users during the one-card transaction process;

[0008] A transaction information item extraction module, used to extract multiple information items from the transaction data;

[0009] A transaction information item semantic embedding coding module, used for performing semantic embedding coding on each of the plurality of information items to obtain semantic embedding coding features of a plurality of transaction data information items;

[0010] An information item embedding semantics enhancement coding module is used to perform semantic field effect distribution enhancement analysis on the semantic embedding coding features of the multiple transaction data information items to obtain multiple transaction data information item enhanced semantic embedding coding features;

[0011] The data type identification and encryption module is used to identify the data type of each transaction data information item enhanced semantic embedded coding feature in the multiple transaction data information item enhanced semantic embedded coding features to obtain an identification result and determine an encryption strategy.

[0012] According to another aspect of the present application, a method for managing the security of a one-card transaction is provided, which includes:

[0013] Get the transaction data generated by the user during the one-card transaction;

[0014] extracting a plurality of information items from the transaction data;

[0015] Performing semantic embedding coding on each of the plurality of information items to obtain semantic embedding coding features of a plurality of transaction data information items;

[0016] Performing enhanced analysis of semantic field effect distribution on the semantic embedding coding features of the multiple transaction data information items to obtain enhanced semantic embedding coding features of the multiple transaction data information items;

[0017] The data type of each transaction data information item enhanced semantic embedded coding feature of the plurality of transaction data information item enhanced semantic embedded coding features is identified to obtain an identification result, and an encryption strategy is determined.

[0018] According to another aspect of the present application, there is provided a computer storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the one-card transaction security management method as described above.

[0019] The present application has at least the following technical effects: Compared with the prior art, the present application provides a one-card transaction security management method, system and storage medium, which extracts each information item from the transaction data generated by the user in the one-card transaction process, and analyzes these information items through data processing and semantic understanding algorithms based on artificial intelligence and natural language processing technology, so as to extract the embedded semantic deep essential association characteristics of these transaction data information items, thereby realizing data type identification of these transaction data information items, so as to identify which data is sensitive data, and apply different encryption strategies to sensitive data of different security levels, and can adaptively encrypt and manage the information items in the one-card transaction data in an automated manner, thereby realizing more intelligent one-card transaction security management, which helps to improve system security and user experience, and enhance users' trust and support for the one-card transaction security management system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0021] Figure 1 A block diagram of a card transaction security management system according to an embodiment of the present application;

[0022] Figure 2 A schematic diagram of data flow of a one-card transaction security management system according to an embodiment of the present application;

[0023] Figure 3 A block diagram of an information item embedding semantic enhancement coding module in a one-card transaction security management system according to an embodiment of the present application;

[0024] Figure 4 It is a flowchart of the one-card transaction security management method according to the embodiment of the present application.

[0025] Reference numerals:

[0026] 300. One-card transaction security management system; 310. One-card transaction data collection module; 320. Transaction information item extraction module; 330. Transaction information item semantic embedding coding module; 340. Information item embedded semantic reinforcement coding module; 350. Data type identification and encryption module; 341. Information item semantic importance quantitative representation and ranking unit; 342. Information item semantic global association coding unit. DETAILED DESCRIPTION

[0027] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0028] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0029] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0030] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0031] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0032] In view of the above technical problems, a one-card transaction security management system is proposed in the technical solution of the present application. Accordingly, in the process of one-card transaction security management, first of all, it is necessary to classify the transaction data generated by the user during the one-card transaction in detail to clarify which ones are sensitive information. These sensitive information usually include but are not limited to personal identity information (such as name, ID number, telephone number), sensitive payment information (such as bank card number, payment password), personal bills and consumption habits. Then, for different categories of sensitive information, different security levels should be set according to security specifications, and they should be clearly identified within the system so that different encryption strategies can be adopted for these data information to improve the security of users using the one-card. Specifically, for highly sensitive data, such as personal identity information, bank card numbers, etc., strong encryption algorithms (such as AES, RSA) should be used for encrypted storage. In this way, even if the data is illegally accessed, the plaintext information cannot be directly read for highly sensitive data. For some sensitive information that needs to be displayed to users (such as bill details), desensitization technology can be used to hide some characters. For example, when displaying a bank card number, asterisks can be used. Instead of showing the middle digits, only the first and last digits are displayed.

[0033] It should be noted that the sensitive information involved in this application, including but not limited to personal identity information (such as name, ID number, telephone number), sensitive payment information (such as bank card number, payment password), personal bills and consumption habits, as well as transaction data, including but not limited to financial transactions, personal consumption data, etc., are all information and data fully authorized by all parties. The collection, use and processing of relevant information and data are carried out in compliance with national and relevant regional laws, regulations and relevant standards.

[0034] In particular, in the above-mentioned one-card transaction security management system, it is crucial to accurately identify the sensitive data in the transaction data generated by the user during the one-card transaction process. This step is the basis for sensitive data encryption and one-card transaction security management. Sensitive data includes personal identity information, sensitive payment information, personal bills, and consumption habits, etc., which involve the privacy and property security of users. Therefore, by classifying and identifying each information item in the one-card transaction data in detail, sensitive data can be distinguished from other non-sensitive data, and different protection measures can be taken for these information items and sensitive data. For example, personal identity information may need to be encrypted and stored using a strong encryption algorithm (such as AES, RSA), while consumption habit data may only require simple desensitization processing. Accurately identifying sensitive data helps to take targeted protection measures, which helps prevent sensitive data from being accidentally leaked during storage, transmission and processing, and improves the overall security of the one-card transaction management system. In addition, by classifying and identifying transaction data, the use of resources can be optimized. For example, it is not necessary to perform high-intensity encryption processing on all data, but to focus resources on highly sensitive data, thereby improving the efficiency and performance of the system.

[0035] Based on this, in the technical solution of this application, a one-card transaction security management system is proposed. Figure 1 It is a block diagram of the one-card transaction security management system according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of the one-card transaction security management system according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to the one-card transaction security management system 300 of the embodiment of the present application, it includes: a one-card transaction data collection module 310, which is used to obtain the transaction data generated by the user in the one-card transaction process; a transaction information item extraction module 320, which is used to extract multiple information items from the transaction data; a transaction information item semantic embedding coding module 330, which is used to perform semantic embedding coding on each of the multiple information items to obtain multiple transaction data information item semantic embedding coding features; an information item embedded semantic enhancement coding module 340, which is used to perform enhanced analysis of the semantic field effect distribution of the semantic embedding coding features of the multiple transaction data information items to obtain multiple transaction data information item enhanced semantic embedding coding features; a data type identification and encryption module 350, which is used to perform data type identification on the enhanced semantic embedding coding features of each transaction data information item in the multiple transaction data information item enhanced semantic embedding coding features to obtain an identification result, and determine an encryption strategy.

[0036] In particular, the one-card transaction data acquisition module 310 is used to obtain transaction data generated by the user during the one-card transaction process. In the one-card transaction process, transaction data refers to various information records generated when the user conducts a transaction.

[0037] In particular, the transaction information item extraction module 320 is used to extract multiple information items from the transaction data. In one example, the information items cover various aspects of the transaction, including transaction identification, user information, transaction amount, transaction time, transaction location, transaction type, payment method, terminal information, merchant information and transaction status. In the technical solution of the present application, it is expected to identify the data type of the transaction data information item in order to identify which data is sensitive data and apply different encryption strategies to sensitive data of different security levels.

[0038] In particular, the transaction information item semantic embedding coding module 330 is used to perform semantic embedding coding on each of the multiple information items to obtain semantic embedding coding features of multiple transaction data information items. In a specific example of the present application, a semantic embedding encoder based on the Bert model is used to perform semantic embedding coding on each of the multiple information items to obtain multiple transaction data information item semantic embedding coding vectors as the semantic embedding coding features of the multiple transaction data information items. It should be understood that in the one-card transaction data, these information items in the transaction data may contain text data in various forms, such as user name, transaction description, consumption location, etc. Therefore, it is necessary to perform semantic embedding coding on these different information items to unify these data types into the same comparable data format and capture the embedded coding semantic features therein. Among them, the Bert model is a pre-trained language model based on the Transformer architecture, which can understand the meaning of words in context. Compared with traditional word embedding methods (such as Word2Vec, GloVe), the Bert model can capture more complex semantic relationships and contextual information. Therefore, by encoding each information item using a semantic embedding encoder based on the Bert model, we can better understand the deep semantics of the text data of these information items, thereby more accurately identifying the sensitive information therein.

[0039] In particular, the information item embedding semantics enhancement coding module 340 is used to perform an enhanced analysis of the semantic field effect distribution of the semantic embedding coding features of the multiple transaction data information items to obtain multiple transaction data information item enhanced semantic embedding coding features. It should be understood that since each transaction data information item semantic embedding coding vector in the multiple transaction data information item semantic embedding coding vectors respectively contains semantic embedding coding features related to different information items in the transaction data, there are some problems in the representation of the embedded semantics of these information items, such as the original transaction data information item semantic embedding coding vector may have a semantic sparsity problem, that is, some words lack sufficient information in a specific context to accurately represent their meaning. In addition, the importance of different transaction data information item embedding coding semantics in different contexts is different, and the traditional word embedding method may not be able to balance these semantic features well, resulting in some key information being ignored, affecting the subsequent type identification and encryption process of sensitive data. This causes a semantic understanding bias when performing contextual semantic coding of transaction data information items. Based on this, in the technical solution of the present application, the multiple transaction data information item semantic embedding coding features are further subjected to an enhanced analysis of the semantic field effect distribution to obtain multiple transaction data information item enhanced semantic embedding coding features. In particular, the enhanced analysis and processing of semantic field effect distribution can perform refined feature extraction and semantic coding of the semantic embedding coding features of multiple transaction data information items, so as to realize deep analysis and semantic enhancement coding of the semantic embedding coding vectors of multiple transaction data information items. It not only overcomes the limitations of traditional methods in dealing with long-distance dependencies and deep semantic mining, but also significantly improves the ability to understand the essence of transaction data content. Specifically, the semantic enhancement coding module based on sequence field effect distribution analysis can better capture the deep semantic information in the text of different information items by performing multi-dimensional processing and enhancement on the semantic embedding coding vectors of transaction data information items, and it combines the advantages of the bidirectional LSTM model, not only considering the unidirectional semantic flow, but also taking into account the reverse contextual influence, thereby enhancing the understanding of the overall meaning of the transaction data. Specifically, in a specific example of the present application, such as Figure 3 As shown, the information item embedded semantic enhancement coding module 340 includes: an information item semantic importance quantification representation sorting unit 341, which is used to calculate the semantic depth implicit representation value of the multiple transaction data information item semantic embedding coding features, and reorder the multiple transaction data information item semantic embedding coding features based on the semantic depth implicit representation value to obtain an updated sequence of transaction data information item semantic embedding coding features; an information item semantic global association coding unit 342, which is used to perform global semantic association coding on the updated sequence of the transaction data information item semantic embedding coding features to obtain the multiple transaction data information item enhanced semantic embedding coding features.

[0040] Specifically, the information item semantic importance quantification representation sorting unit 341 is used to calculate the semantic depth implicit representation value of the semantic embedding coding features of the multiple transaction data information items, and reorder the multiple transaction data information item semantic embedding coding features based on the semantic depth implicit representation value to obtain an update sequence of the transaction data information item semantic embedding coding features. In an embodiment of the present application, first, a semantic reconstructor is used to perform feature reconstruction on each transaction data information item semantic embedding coding vector in the multiple transaction data information item semantic embedding coding vectors to obtain multiple reconstructed transaction data information item semantic embedding coding feature vectors; specifically, a semantic reconstructor is used to perform domain transformation on each element in the original multiple transaction data information item semantic embedding coding vectors to generate multiple new reconstructed transaction data information item semantic embedding coding feature vectors. The semantic reconstructor enables the original semantic embedding coding vectors of each transaction data information item to achieve dimensional unification, and enhances the representation ability of the embedded coding semantics of each information item in a specific context, so that the feature vector is more in line with the meaning in the actual transaction data language environment. Next, the semantic depth implicit representation value of each reconstructed transaction data information item semantic embedding coding feature vector in the multiple reconstructed transaction data information item semantic embedding coding feature vectors is calculated to obtain multiple transaction data information item semantic embedding coding semantic depth implicit representation values; specifically, by calculating the semantic depth implicit representation value of these reconstructed transaction data information item semantic embedding coding feature vectors, the information density or complexity carried by the single information item embedding coding semantics in the entire transaction data is reflected; specifically, the semantic depth implicit representation value can quantify the importance and influence of each information item embedding coding feature in the transaction data in which it is located, and according to these values, the order of the transaction data information item semantic embedding coding vector is further adjusted to ensure that key information is given priority. For example, in the transaction description, some information items may be more important than other information items, and the order can ensure that these key information receive more attention in the encoding process. Further, based on the order of the multiple transaction data information item semantic embedding coding semantic depth implicit representation values, the multiple transaction data information item semantic embedding coding vectors are reordered to obtain the update sequence of the transaction data information item semantic embedding enhanced coding vector as the update sequence of the transaction data information item semantic embedding coding feature. Specifically, the semantic depth implicit representation values ​​corresponding to the semantic embedding coding vectors of each transaction data information item are sorted to generate an updated sequence of new transaction data information item semantic embedding enhanced coding vectors. In this way, the semantic structure is reorganized according to the importance of the semantics of the information item embedding coding, which can improve the quality and consistency of the global semantic coding.

[0041] Among them, the specific steps of using a semantic reconstructor to reconstruct the features of the semantic embedding coding vectors of each transaction data information item in the multiple transaction data information item semantic embedding coding vectors include: multiplying the first weight matrix by the semantic embedding coding vector of the transaction data information item, and then multiplying the obtained feature vector by the second weight matrix to obtain the reconstructed transaction data information item semantic embedding coding feature vector.

[0042] More specifically, the specific steps of calculating the semantic depth implicit representation value of each reconstructed transaction data information item semantic embedding coding feature vector in the multiple reconstructed transaction data information item semantic embedding coding feature vectors include: calculating the square of the one-norm of the reconstructed transaction data information item semantic embedding coding feature vector to obtain the transaction data information item semantic embedding coding mapping feature norm representation; adding and subtracting a constant one from the transaction data information item semantic embedding coding mapping feature norm representation to obtain a first transaction data information item semantic embedding coding mapping feature norm modulation value and a second transaction data information item semantic embedding coding mapping feature norm modulation value, and dividing the first transaction data information item semantic embedding coding mapping feature norm modulation value by the second transaction data information item semantic embedding coding mapping feature norm modulation value to obtain the transaction data information item semantic embedding coding mapping feature norm modulation value; calculating the logarithmic function value of the transaction data information item semantic embedding coding mapping feature norm modulation value with base two to obtain the transaction data information item semantic embedding coding semantic depth implicit representation value.

[0043] Specifically, the information item semantic global association encoding unit 342 is used to perform global semantic association encoding on the update sequence of the transaction data information item semantic embedding encoding features to obtain the multiple transaction data information item enhanced semantic embedding encoding features. In an embodiment of the present application, first, the update sequence of the transaction data information item semantic embedding enhanced encoding vector is input into the first semantic encoder based on the forward LSTM model to obtain the sequence of transaction data information item semantic forward global semantic encoding feature vectors; and the update sequence of the transaction data information item semantic embedding enhanced encoding vector is input into the second semantic encoder based on the backward LSTM model to obtain the sequence of transaction data information item semantic backward global semantic encoding feature vectors; specifically, a bidirectional long short-term memory (LSTM) model is used to encode the update sequence of the updated transaction data information item semantic embedding enhanced encoding vector from the forward and backward directions respectively. The forward LSTM captures the information flow pattern from the beginning to the end of the sentence, while the backward LSTM operates in reverse to identify the dependency relationship from the end of the sentence to the beginning of the sentence. This dual-perspective encoding mechanism effectively captures the complex semantic structure information between each information item in the transaction data and provides a more comprehensive semantic understanding. Next, the forward semantic encoding feature vectors of the transaction data information items and the backward semantic encoding feature vectors of the transaction data information items at corresponding positions in the sequence of the forward semantic encoding feature vectors of the transaction data information items and the sequence of the backward semantic encoding feature vectors of the transaction data information items are fused to obtain multiple transaction data information item enhanced semantic embedding encoding vectors as the multiple transaction data information item enhanced semantic embedding encoding features. In this case, the bidirectional semantic information is integrated by fusing the global semantic encoding results from the forward and backward directions to form a richer and more comprehensive enhanced semantic embedding encoding vector for each transaction data information item, supporting more complex transaction data semantic understanding and different information item data type recognition tasks.

[0044] Among them, the specific steps of fusing the forward semantic coding feature vectors of the transaction data information items and the backward semantic coding feature vectors of the transaction data information items at corresponding positions in the sequence of the forward semantic coding feature vectors of the transaction data information items and the sequence of the backward semantic coding feature vectors of the transaction data information items include: multiplying the forward semantic coding feature vectors of the transaction data information items at the corresponding positions with the forward weight matrix of the transaction data information items to obtain the forward semantic modulation vectors of the transaction data information items; multiplying the backward semantic coding feature vectors of the transaction data information items at the corresponding positions with the backward weight matrix of the transaction data information items to obtain the backward semantic modulation vectors of the transaction data information items; adding the forward semantic modulation vectors of the transaction data information items and the backward semantic modulation vectors of the transaction data information items by position, and then adding them by position with the transaction data information item bias vector to obtain the enhanced semantic embedding coding vectors of the transaction data information items.

[0045] In summary, in the above embodiment, the semantic embedding coding features of the multiple transaction data information items are subjected to an enhanced analysis of the semantic field effect distribution to obtain the enhanced semantic embedding coding features of the multiple transaction data information items, including: performing an enhanced analysis of the semantic field effect distribution on the semantic embedding coding features of the multiple transaction data information items using the following semantic enhancement formula to obtain the enhanced semantic embedding coding features of the multiple transaction data information items; wherein the semantic enhancement formula is:

[0046]

[0047]

[0048]

[0049]

[0050]

[0051]

[0052]

[0053] in, semantically embedding encoding vectors for the plurality of transaction data information items, are respectively the first, second, ..., first, second, third, fourth, fifth, fifth, sixth, sixth, fifth, sixth, sixth, seventh ... ,..., The semantic embedding encoding vector of transaction data information items, and are the first weight matrix and the second weight matrix respectively, The semantic embedding encoding feature vector of multiple reconstructed transaction data information items is The semantic embedding encoding feature vector of the reconstructed transaction data information item, is the square of the one-norm of the vector, represents the logarithmic function value with base 2, It is the first implicit representation value of the semantic embedding encoding semantic depth of multiple transaction data information items. The semantic embedding of transaction data information items encodes the semantic depth implicit representation value. is the update sequence of the feature vector of the transaction data information item, is the first, second and third in the update sequence of the semantic embedding enhanced coding vector of the transaction data information item transaction data information item word feature vector, is the forward LSTM encoding, is the forward global semantic encoding feature vector of the transaction data information item, is the backward LSTM encoding, is the backward global semantic encoding feature vector of the transaction data information item, and are the forward weight matrix of transaction data information item and the backward weight matrix of transaction data information item, respectively. is the transaction data information item bias vector, A semantic embedding encoding vector is enhanced for the transaction data information item.

[0054] In particular, the data type identification and encryption module 350 is used to identify the data type of each transaction data information item in the enhanced semantic embedding coding features of the multiple transaction data information items to obtain an identification result, and determine the encryption strategy. In a specific example of the present application, each transaction data information item in the enhanced semantic embedding coding vector is input into a data type identifier based on a classifier to obtain the identification result; specifically, the enhanced semantic embedding representation of each information item in the transaction data is used to perform classification processing respectively, so as to perform data type identification to obtain the identification result of each information item. Among them, the identification result is the data category of each information item, which can identify which transaction data information items are sensitive data, thereby realizing the encryption of sensitive data. Furthermore, based on the identification results of each information item, the encryption strategy is determined. Specifically, in the process of data type identification, the security level labels of these transaction data information items can also be obtained, and the encryption strategy is determined based on the security level labels.

[0055] Specifically, personal identity information (such as name, ID number, phone number): This type of information is usually used to verify the identity of the user and ensure that only legitimate users can access their accounts. In order to protect this type of information, it is recommended to use high-strength encryption algorithms, such as AES (Advanced Encryption Standard) or RSA (public key encryption algorithm) to ensure that even if the data is illegally obtained, it cannot be easily decrypted. Sensitive payment information (such as bank card number, payment password): Information involving financial transactions, such as bank card numbers, should be protected at the highest level. In addition to using the above encryption methods, additional security measures can be implemented, such as two-factor authentication, transaction limit control, etc. to reduce potential risks. Personal bills and spending habits: Although the sensitivity of this type of information is relatively low, privacy protection still needs to be paid attention to. You can consider using data desensitization technology, that is, partially masking specific fields without affecting the value of the data. For example, when displaying a bank card number, only the first and last few digits are exposed, and the rest are represented by asterisks. In this way, the information items in the one-card transaction data can be adaptively encrypted and managed in an automated manner, thereby realizing more intelligent one-card transaction security management, helping to improve system security and user experience, and enhancing users' trust and support for the one-card transaction security management system.

[0056] Among them, each transaction data information item semantic embedding coding vector in the set of transaction data information item semantic embedding coding vectors respectively represents the text semantic embedding coding features of each transaction data information item. When performing enhanced analysis of the semantic field effect distribution, it is possible to enhance the semantic features of each transaction data information item semantic embedding coding vector based on local association semantics based on the semantic field effect distribution of the set of transaction data information item semantic embedding coding vectors. However, due to the field dynamics of the semantic field effect distribution, each transaction data information item enhanced semantic embedding coding vector in the set of transaction data information item enhanced semantic embedding coding vectors will have significant feature explicit spatial structural differences, which will affect the convergence consistency of the classifier and the accuracy of the classification results.

[0057] Preferably, inputting the transaction data information item enhanced semantic embedding coding vector into a classifier-based data type identifier to obtain an identification result comprises:

[0058] The absolute value sum of all eigenvalues ​​of the enhanced semantic embedding coding vector of the transaction data information item and the square root of the sum of squares are calculated to obtain the first transaction data information item enhanced semantic embedding coding space structure value and the second transaction data information item enhanced semantic embedding coding space structure value, that is:

[0059]

[0060]

[0061] in, The first eigenvalues, Indicates the first transaction data information item strengthens the semantic embedding coding space structure value, Indicates that the second transaction data information item strengthens the semantic embedding encoding space structure value;

[0062] Determine the number of eigenvalues ​​of all eigenvalues ​​of the enhanced semantic embedding encoding vector of the transaction data information item , i.e., the length of the transaction data information item enhanced semantic embedding encoding vector;

[0063] For each eigenvalue of the transaction data information item enhanced semantic embedding coding vector, the first transaction data information item enhanced semantic embedding coding spatial structure value minus the product of the eigenvalue and the number of eigenvalues ​​is calculated to obtain the first transaction data information item enhanced semantic embedding coding long-range dependency value ,in, Indicates the first transaction data information item strengthens the semantic embedding coding space structure value, The first eigenvalues, The number of eigenvalues ​​representing all eigenvalues ​​of the enhanced semantic embedding encoding vector of the transaction data information item, Indicates that the first transaction data information item strengthens the semantic embedding encoding long-range dependency value;

[0064] Calculate the product of the square root of the number of eigenvalues ​​multiplied by the eigenvalue minus the second transaction data information item enhanced semantic embedding coding spatial structure value to obtain the second transaction data information item enhanced semantic embedding coding long-range dependency value ,in, Indicates the second transaction data information item strengthens the semantic embedding encoding space structure value, The first eigenvalues, The number of eigenvalues ​​representing all eigenvalues ​​of the enhanced semantic embedding encoding vector of the transaction data information item, Indicating that the second transaction data information item strengthens the semantic embedding encoding long-range dependency value;

[0065] The index value obtained by taking the long-range dependency value of the enhanced semantic embedding coding of the first transaction data information item as the exponent of the natural constant and the inverse of the long-range dependency value of the enhanced semantic embedding coding of the second transaction data information item are weighted summed to obtain the optimized eigenvalue corresponding to each eigenvalue ,in, Indicates that the first transaction data information item strengthens the semantic embedding encoding long-range dependency value, Indicates that the second transaction data information item strengthens the semantic embedding encoding long-range dependency value, represents a natural constant, and represents the weighted hyperparameter, Indicates the optimized eigenvalue corresponding to each eigenvalue;

[0066] Combining the optimized feature values ​​into an optimized transaction data information item to strengthen the semantic embedding coding vector;

[0067] The optimized transaction data information item reinforced semantic embedding coding vector is input into a classifier-based data type identifier to obtain a recognition result.

[0068] That is, in view of the possible spatial structure loss in the high-dimensional space of the feature set of the transaction data information item enhanced semantic embedding coding vector, which leads to the weight of the classifier implicitly inferring the spatial structure information based on the feature, resulting in inconsistent convergence. By establishing a long-distance feature dependency relationship based on the overall feature scale of the transaction data information item enhanced semantic embedding coding vector relative to the spatial structure representation of the transaction data information item enhanced semantic embedding coding vector, the feature local connectivity of the transaction data information item enhanced semantic embedding coding vector is established, and the spatial ambiguity information of the object feature value is captured by the unstructured feature value point prediction of the transaction data information item enhanced semantic embedding coding vector, thereby improving the spatial inductive bias perception ability of the feature set of the transaction data information item enhanced semantic embedding coding vector, improving the convergence consistency of the classifier, and improving the accuracy of the recognition result obtained by the data type identifier based on the classifier input of the transaction data information item enhanced semantic embedding coding vector. In this way, the data type of these transaction data information items can be automatically identified to identify which information items are sensitive data, and different encryption strategies are applied to sensitive data of different security levels, thereby realizing more intelligent one-card transaction security management.

[0069] As described above, the one-card transaction security management system 300 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a one-card transaction security management algorithm. In a possible implementation, the one-card transaction security management system 300 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the one-card transaction security management system 300 can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the one-card transaction security management system 300 can also be one of the many hardware modules of the wireless terminal.

[0070] Alternatively, in another example, the one-card transaction security management system 300 and the wireless terminal may also be separate devices, and the one-card transaction security management system 300 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0071] Furthermore, a method for managing the security of one-card transactions is also provided.

[0072] Figure 4 Flow chart of the one-card transaction security management method according to the embodiment of the present application. Figure 4As shown, according to the one-card transaction security management method of the embodiment of the present application, the steps include: S1, obtaining transaction data generated by the user in the one-card transaction process; S2, extracting multiple information items from the transaction data; S3, semantically embedding and coding each of the multiple information items to obtain semantic embedding coding features of multiple transaction data information items; S4, performing enhanced analysis of the semantic field effect distribution of the semantic embedding coding features of the multiple transaction data information items to obtain strengthened semantic embedding coding features of multiple transaction data information items; S5, performing data type identification on the strengthened semantic embedding coding features of each transaction data information item in the strengthened semantic embedding coding features of the multiple transaction data information items to obtain an identification result, and determining an encryption strategy.

[0073] In summary, according to the embodiment of the present application, the one-card transaction security management method is clarified, which extracts each information item from the transaction data generated by the user in the one-card transaction process, and analyzes these information items through data processing and semantic understanding algorithms based on artificial intelligence and natural language processing technology, so as to extract the embedded semantic deep essential correlation characteristics of these transaction data information items, thereby realizing data type identification of these transaction data information items, so as to identify which data is sensitive data, and apply different encryption strategies to sensitive data of different security levels, and can adaptively encrypt and manage the information items in the one-card transaction data in an automated manner, thereby realizing more intelligent one-card transaction security management, which helps to improve system security and user experience, and enhance users' trust and support for the one-card transaction security management system.

[0074] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps in the functions of the one-card transaction security management method according to various embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0075] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0076] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. One-card transaction security management system, characterized by: include: One-card transaction data collection module, used to obtain transaction data generated by users during the one-card transaction process; A transaction information item extraction module, used to extract multiple information items from the transaction data; A transaction information item semantic embedding coding module, used for performing semantic embedding coding on each of the plurality of information items to obtain semantic embedding coding features of a plurality of transaction data information items; An information item embedding semantic enhancement coding module is used to perform semantic field effect distribution enhancement analysis on the semantic embedding coding features of the multiple transaction data information items to obtain multiple transaction data information item enhanced semantic embedding coding features; The information item embedded semantics enhancement coding module includes: an information item semantic importance quantification representation sorting unit, which is used to calculate the semantic depth implicit representation values ​​of the semantic embedded coding features of the multiple transaction data information items, and re-sort the multiple transaction data information item semantic embedded coding features based on the semantic depth implicit representation values ​​to obtain an update sequence of the transaction data information item semantic embedded coding features; an information item semantic global association coding unit, which is used to perform global semantic association coding on the update sequence of the transaction data information item semantic embedded coding features to obtain the multiple transaction data information item enhanced semantic embedded coding features; The data type identification and encryption module is used to identify the data type of each transaction data information item enhanced semantic embedded coding feature in the multiple transaction data information item enhanced semantic embedded coding features to obtain an identification result and determine an encryption strategy.

2. The one-card transaction security management system according to claim 1 is characterized in that: The transaction information item semantic embedding coding module is used to: use a semantic embedding encoder based on the Bert model to perform semantic embedding coding on each of the multiple information items to obtain multiple transaction data information item semantic embedding coding vectors as the semantic embedding coding features of the multiple transaction data information items.

3. The one-card transaction security management system according to claim 2 is characterized in that: The information item semantic importance quantitative representation sorting unit comprises: a feature reconstruction subunit, configured to use a semantic reconstructor to perform feature reconstruction on each of the plurality of transaction data information item semantic embedding coding vectors to obtain a plurality of reconstructed transaction data information item semantic embedding coding feature vectors; A semantic depth implicit representation value calculation subunit, used to calculate the semantic depth implicit representation value of each reconstructed transaction data information item semantic embedding encoding feature vector in the plurality of reconstructed transaction data information item semantic embedding encoding feature vectors to obtain a plurality of transaction data information item semantic embedding encoding semantic depth implicit representation values; A sequence updating subunit is used to reorder the multiple transaction data information item semantic embedding coding vectors based on the order of the semantic depth implicit representation values ​​of the semantic embedding coding of the multiple transaction data information items to obtain an updated sequence of transaction data information item semantic embedding enhanced coding vectors as an updated sequence of the transaction data information item semantic embedding coding features.

4. The one-card transaction security management system according to claim 3 is characterized in that: The feature reconstruction subunit is used to: multiply the first weight matrix by the transaction data information item semantic embedding coding vector, and then multiply the obtained feature vector by the second weight matrix to obtain the reconstructed transaction data information item semantic embedding coding feature vector.

5. The one-card transaction security management system according to claim 4 is characterized in that: The semantic depth implicit representation value calculation subunit is used to: Calculating the square of a norm of the semantic embedding coding feature vector of the reconstructed transaction data information item to obtain a norm representation of the semantic embedding coding mapping feature of the transaction data information item; Adding and subtracting a constant one from the transaction data information item semantic embedding coding mapping feature norm representation to obtain a first transaction data information item semantic embedding coding mapping feature norm modulation value and a second transaction data information item semantic embedding coding mapping feature norm modulation value, and dividing the first transaction data information item semantic embedding coding mapping feature norm modulation value by the second transaction data information item semantic embedding coding mapping feature norm modulation value to obtain a transaction data information item semantic embedding coding mapping feature norm modulation value; With base two, the logarithmic function value of the transaction data information item semantic embedding coding mapping feature norm modulation value is calculated to obtain the transaction data information item semantic embedding coding semantic depth implicit representation value.

6. The one-card transaction security management system according to claim 5 is characterized in that: The information item semantic global association encoding unit comprises: A transaction data information item semantic forward global semantic encoding subunit, used for inputting the update sequence of the transaction data information item semantic embedding enhanced encoding vector into a first semantic encoder based on a forward LSTM model to obtain a sequence of transaction data information item semantic forward global semantic encoding feature vectors; A transaction data information item semantic backward global semantic encoding subunit, used for inputting the update sequence of the transaction data information item semantic embedding enhanced encoding vector into a second semantic encoder based on a backward LSTM model to obtain a sequence of transaction data information item semantic backward global semantic encoding feature vectors; The transaction data information item enhanced semantic embedding coding subunit is used to fuse the transaction data information item forward semantic coding feature vectors and the transaction data information item backward semantic coding feature vectors at corresponding positions in the sequence of the transaction data information item forward semantic coding feature vectors and the sequence of the transaction data information item backward semantic coding feature vectors to obtain multiple transaction data information item enhanced semantic embedding coding vectors as the multiple transaction data information item enhanced semantic embedding coding features.

7. The one-card transaction security management system according to claim 6 is characterized in that: The transaction data information item strengthens the semantic embedding coding subunit, and is used to: Multiplying the forward semantic encoding feature vector of the transaction data information item at the corresponding position by the forward weight matrix of the transaction data information item to obtain a forward semantic modulation vector of the transaction data information item; Multiplying the backward semantic encoding feature vector of the transaction data information item at the corresponding position by the backward weight matrix of the transaction data information item to obtain a backward semantic modulation vector of the transaction data information item; After adding the forward semantic modulation vector of the transaction data information item and the backward semantic modulation vector of the transaction data information item by position, the vector is then added by position with the bias vector of the transaction data information item to obtain the enhanced semantic embedding coding vector of the transaction data information item.

8. The one-card transaction security management system according to claim 7 is characterized in that: Data type identification and encryption module for: Inputting each transaction data information item enhanced semantic embedding coding vector of the plurality of transaction data information item enhanced semantic embedding coding vectors into a classifier-based data type identifier to obtain the identification result; Based on the identification results of the respective information items, an encryption strategy is determined.

9. A method for managing the security of one-card transactions, characterized in that: include: Obtain transaction data generated by users during the one-card transaction process; extracting a plurality of information items from the transaction data; Performing semantic embedding coding on each of the plurality of information items to obtain semantic embedding coding features of a plurality of transaction data information items; The semantic embedding coding features of the multiple transaction data information items are subjected to enhanced analysis of semantic field effect distribution to obtain enhanced semantic embedding coding features of the multiple transaction data information items, which includes: Calculating semantic depth implicit representation values ​​of the semantic embedding coding features of the multiple transaction data information items, and reordering the semantic embedding coding features of the multiple transaction data information items based on the semantic depth implicit representation values ​​to obtain an update sequence of the semantic embedding coding features of the transaction data information items; performing global semantic association coding on the update sequence of the semantic embedding coding features of the transaction data information items to obtain enhanced semantic embedding coding features of the multiple transaction data information items; The data type of each transaction data information item enhanced semantic embedded coding feature of the plurality of transaction data information item enhanced semantic embedded coding features is identified to obtain an identification result, and an encryption strategy is determined.

10. A computer storage medium having computer program instructions stored thereon, wherein when the computer program instructions are executed by a processor, the processor is enabled to execute the one-card transaction security management system as claimed in any one of claims 1 to 8.

Images