Transaction information tampering risk identification method and device, equipment and storage medium

Abstract

The invention discloses a transaction information tampering risk identification method and device, equipment and a storage medium, and belongs to the technical field of Internet. The method is executed by computer equipment and comprises the steps that transaction information of a merchant is acquired, and the transaction information is used for indicating transaction details of at least one selling transaction executed by the merchant; feature extraction processing is conducted on the transaction information, transaction feature representation and feature description information in the transaction information are obtained, the transaction feature representation is used for indicating features of transaction resources in the selling transaction executed by the merchant, and the feature description information is used for describing attribute information of the transaction feature representation; obtaining feature processing information and reference identification features corresponding to the feature description information; performing character editing processing on the transaction feature representation based on the feature processing information to obtain a to-be-identified feature; and comparing the reference identification feature with the to-be-identified feature to obtain an identification result of the transaction information.

Description

Technical Field

[0001] The present application relates to the field of Internet technology, and in particular to a method, device, equipment and storage medium for identifying the risk of transaction information tampering. Background Art

[0002] With the development of Internet technology, there exist online e-commerce platforms for online shopping.

[0003] In related technologies, when merchants enter a new e-commerce platform, they need to upload screenshots of their transaction flows to prove the online shopping stores they have operated in the past, in order to verify their experience of entering the e-commerce platform. The e-commerce platform needs to conduct manual review to check whether there are any signs of image modification in the screenshots of the transaction flows, and to determine the authenticity of the screenshots of the transaction flows.

[0004] However, the authenticity of transaction flow screenshots cannot be accurately determined simply by checking whether there are traces of image modification in the screenshots. How to improve the accuracy of identifying transaction flow screenshots is an urgent problem to be solved. Summary of the invention

[0005] The present application provides a method, device, equipment and storage medium for identifying the risk of transaction information tampering. The technical solution is as follows:

[0006] According to one aspect of the present application, a method for identifying the risk of transaction information tampering is provided, the method being executed by a computer device, the method comprising:

[0007] Acquire the transaction information of the merchant, where the transaction information is used to indicate transaction details of at least one sales transaction performed by the merchant;

[0008] Performing feature extraction processing on the transaction information to obtain a transaction feature representation and feature description information in the transaction information, wherein the transaction feature representation is used to indicate the characteristics of the transaction resources in the sales transaction performed by the merchant, and the feature description information is used to describe the attribute information of the transaction feature representation;

[0009] Acquire feature processing information and reference identification features corresponding to the feature description information, wherein the feature processing information is used to indicate a character editing method for the transaction feature representation, and the reference identification feature is an original information feature in a secure transaction;

[0010] Based on the feature processing information, character editing processing is performed on the transaction feature representation to obtain a feature to be identified, where the feature to be identified is a character string that needs to be tampered risk identified and is obtained based on the transaction feature representation;

[0011] The reference identification feature and the feature to be identified are compared to obtain an identification result of the transaction information.

[0012] According to another aspect of the present application, a device for identifying the risk of transaction information tampering is provided, the device comprising:

[0013] An acquisition module, used to acquire the transaction information of the merchant, wherein the transaction information is used to indicate transaction details of at least one sales transaction performed by the merchant;

[0014] a processing module, configured to perform feature extraction processing on the transaction information to obtain a transaction feature representation and feature description information in the transaction information, wherein the transaction feature representation is used to indicate the characteristics of the transaction resources in the sales transaction performed by the merchant, and the feature description information is used to describe the attribute information of the transaction feature representation;

[0015] The acquisition module is further used to acquire feature processing information and reference identification features corresponding to the feature description information, wherein the feature processing information is used to indicate a character editing method for the transaction feature representation, and the reference identification feature is an original information feature in a secure transaction;

[0016] The processing module is further used to perform character editing processing on the transaction feature representation based on the feature processing information to obtain a feature to be identified, wherein the feature to be identified is a character string that needs to be tampered risk identified based on the transaction feature representation;

[0017] The processing module is further used to compare the reference identification feature with the feature to be identified to obtain an identification result of the transaction information.

[0018] In an optional implementation of the application, the feature processing information includes extraction instruction information, the extraction instruction information indicates a character extraction method for representing the transaction feature; the reference identification feature includes a reference format feature, the reference format feature is used to indicate the arrangement format characteristics of the reference transaction value in the security transaction;

[0019] The processing module is also used for:

[0020] Based on the extraction indication information, character extraction processing is performed on the transaction feature representation to obtain a character format feature, wherein the transaction feature representation is used to describe a transaction-related value in the sales transaction, and the character format feature is used to indicate an arrangement format feature of the transaction-related value;

[0021] A character string comparison operation is performed on the reference format feature and the character format feature to obtain the identification result.

[0022] In an optional implementation of the application, the processing module is further used for at least one of the following:

[0023] Based on first extraction indication information, performing character extraction processing on a decimal point symbol in the transaction feature representation, and obtaining a numerical precision of the transaction-related numerical value based on a position of the decimal point symbol, wherein the first extraction indication information indicates a symbol style of the decimal point symbol;

[0024] Based on the second extraction instruction information, character extraction processing is performed at the head position or the tail position of the transaction feature representation to obtain the currency symbol of the transaction-related value, and the second extraction instruction information indicates the extraction position of the currency symbol;

[0025] Based on the third extraction indication information, character extraction processing is performed on the transaction feature representation to obtain a separator symbol of the transaction-related numerical value, and the third extraction indication information indicates a preset interval for extracting the separator symbol.

[0026] In an optional implementation of the application, the feature processing information includes statistical indication information, the statistical indication information indicates a statistical calculation method and a numerical comparison method for at least two transaction feature representations; the reference identification feature includes a reference consistency feature, the reference consistency feature is used to indicate the consistency of at least two transaction numerical features of the secure transaction;

[0027] The processing module is also used for:

[0028] Based on the statistical indication information, statistical calculation processing is performed on the at least two transaction feature representations to obtain at least two transaction numerical features;

[0029] Based on the statistical indication information, performing a numerical comparison operation on the at least two transaction numerical features to obtain a data consistency feature;

[0030] A string comparison operation is performed on the reference consistency feature and the data consistency feature to obtain the identification result.

[0031] In an optional implementation of the application, the at least two transaction feature representations correspond to the same feature description information, and the at least two transaction feature representations both have numerical features;

[0032] The processing module is also used for:

[0033] Based on the statistical indication information, performing value extraction processing on at least two transaction feature representations one by one to obtain the at least two transaction value features, wherein the at least two transaction value features correspond to the at least two transaction feature representations in a one-to-one manner;

[0034] The reference consistency feature is used to indicate the data consistency of transaction feature representations having the same feature description information at different locations in the secure transaction.

[0035] In an optional implementation of the application, the statistical indication information is used to perform statistical calculation processing on at least three transaction feature representations, the at least three transaction feature representations corresponding to different feature description information, and the statistical calculation processing includes numerical calculation processing and numerical extraction processing;

[0036] The processing module is also used for:

[0037] Based on the operation information in the statistical indication information, the numerical calculation process is performed on at least two transaction feature representations to obtain a first transaction numerical feature, where the first transaction numerical feature is used to indicate a transaction numerical value implicitly carried by the at least two transaction feature representations in the transaction information;

[0038] Based on the extraction information in the statistical indication information, performing the value extraction process on a transaction feature representation to obtain a second transaction value feature, where the second transaction value feature is used to indicate a transaction value explicitly carried by the transaction feature representation in the transaction information;

[0039] The reference consistency feature is used to indicate the data consistency between the transaction value implicitly carried and the transaction value explicitly carried in the secure transaction.

[0040] In an optional implementation of the application, the at least two transaction feature representations correspond to the same feature description information, the first transaction feature representation of the at least two transaction feature representations does not have a numerical feature, and the at least two transaction feature representations do have a numerical feature;

[0041] The processing module is also used for:

[0042] Based on the numerical mapping relationship in the statistical indication information, searching for a mapping numerical range corresponding to the first transaction feature representation;

[0043] Based on the extraction information in the statistical indication information, performing numerical extraction processing on the second transaction feature representation to obtain a third transaction numerical feature;

[0044] Based on the statistical indication information, a numerical comparison operation is performed on the mapped value range and the third transaction value feature to obtain a data consistency feature, where the data consistency feature is used to indicate whether the third transaction value belongs to the mapped value range;

[0045] The reference consistency feature is used to indicate the numerical value attribution relationship between the reference numerical value range and the reference transaction value in the security transaction.

[0046] In an optional implementation of the application, the acquisition module is further used to:

[0047] Acquire a type tag of the transaction information, where the type tag is used to indicate a statistical method for the information of the at least one sales transaction;

[0048] The feature processing information is determined based on the feature description information and the type label.

[0049] In an optional implementation of the application, the acquisition module is further used to:

[0050] By comparing the transaction information and the sample information, obtaining similarity information for the sample information;

[0051] When the similarity information exceeds a trust threshold, the sample label of the sample information is determined as the type label of the transaction information.

[0052] In an optional implementation of the application, the acquisition module is further used to:

[0053] Retrieving preset keywords in the transaction information;

[0054] In the case where the preset keyword exists in the transaction information, the sample label corresponding to the preset keyword is determined as the type label of the transaction information.

[0055] In an optional implementation of the application, the transaction information includes a transaction picture of at least one transaction detail, and the processing module is further used to:

[0056] Performing text recognition on the transaction image to obtain transaction data carried by the transaction image;

[0057] The transaction feature representation and the feature description information are determined in the target transaction data corresponding to the preset position in the transaction picture.

[0058] In an optional implementation of the application, the transaction image includes table information, and the processing module is further used to:

[0059] Determine the first transaction data corresponding to the first preset position in the transaction image as the feature description information, the first preset position is the header position of the table information, and the feature description information is used to indicate the attribute information represented by the transaction feature;

[0060] The second transaction data corresponding to the second preset position in the transaction image is determined as the transaction feature representation, and the second preset position and the first preset position have the same horizontal coordinate position or vertical coordinate position.

[0061] In an optional implementation of the application, the processing module is further used for:

[0062] The transaction picture is subjected to image optimization processing to obtain an optimized transaction picture. The text recognition is used to recognize the optimized transaction picture. The image optimization processing includes at least one of grayscale processing, sharpening processing and contrast enhancement processing.

[0063] According to another aspect of the present application, a computer device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the method for identifying the risk of transaction information tampering as described above.

[0064] According to another aspect of the present application, a computer-readable storage medium is provided, in which at least one instruction, at least one program, a code set or an instruction set is stored, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the method for identifying the risk of transaction information tampering as described above.

[0065] According to another aspect of the present application, a computer program product is provided, which includes computer instructions stored in a computer-readable storage medium, and a processor reads and executes the computer instructions from the computer-readable storage medium to implement the method for identifying the risk of transaction information tampering as described above.

[0066] The beneficial effects of the technical solution provided by this application include at least:

[0067] By performing feature extraction processing on the transaction information, characteristic information about the transaction resources in the sales transaction is obtained in the transaction details, and the identification dimension of the transaction information is expanded to the transaction items and resource transfer conditions in the transaction details; the transaction feature representation is processed by character editing to obtain the features to be identified, and the editing and extraction of the character strings that need to be identified for tampering risks is realized. Even if the tampering means of the transaction information does not leave explicit modification traces, the solution of the present application can accurately identify the tampering risk of the transaction information from the implicit information at the semantic level such as the transaction items and resource transfer conditions; that is, the solution of the present application does not rely on obvious modification traces to identify the tampering risk, but identifies the tampering risk through the inherent characteristics implicitly carried in the transaction information; it can identify the tampering risk caused by the modification made by the background controller during the transaction information generation stage, and increases the scope of application of effectively identifying the tampering risk of the transaction information. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0069] Figure 1 is a schematic diagram of a computer system provided by an exemplary embodiment of the present application;

[0070] Figure 2 is a schematic diagram of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application;

[0071] Figure 3 is a flow chart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application;

[0072] Figure 4 is a flow chart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application;

[0073] Figure 5 is a schematic diagram of transaction information provided by an exemplary embodiment of the present application;

[0074] Figure 6 is a flow chart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application;

[0075] Figure 7 is a schematic diagram of transaction information provided by an exemplary embodiment of the present application;

[0076] Figure 8 is a schematic diagram of transaction information provided by an exemplary embodiment of the present application;

[0077] Fig. 9 is a schematic diagram of transaction information provided by an exemplary embodiment of the present application;

[0078] Fig.10 is a flow chart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application;

[0079] Fig.11 is a schematic diagram of transaction information provided by an exemplary embodiment of the present application;

[0080] Fig.12 is a flow chart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application;

[0081] Fig.13 is a schematic diagram of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application;

[0082] Fig.14 is a flow chart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application;

[0083] Fig.15 It is a structural block diagram of a device for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application;

[0084] Fig.16 It is a structural block diagram of a server provided by an exemplary embodiment of the present application.

[0085] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application. DETAILED DESCRIPTION

[0086] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0087] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0088] The terms used in this disclosure are for the purpose of describing specific embodiments only and are not intended to limit the disclosure. The singular forms of "a", "said" and "the" used in this disclosure and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0089] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions. For example, the transaction information involved in this application is obtained with full authorization.

[0090] It should be understood that although the terms first, second, etc. may be used in the present disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present disclosure, the first parameter may also be referred to as the second parameter, and similarly, the second parameter may also be referred to as the first parameter. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0091] Figure 1 A schematic diagram of a computer system provided by an embodiment of the present application is shown. The computer system can be implemented as a system architecture of a method for identifying the risk of transaction information tampering. The computer system may include: a merchant terminal 110, a data server 120, and a computer device 200, wherein the computer device 200 includes an identification terminal 210 and an identification server 220 for identifying the risk of transaction information tampering.

[0092] The merchant terminal 110 may be an electronic device such as a mobile phone, a tablet computer, a personal computer (PC), a personal digital assistant (PDA), etc. The merchant terminal 110 may directly provide the merchant's transaction information to the computer device 200, or store the merchant's transaction information in the data server 120, and the data server 120 provides the merchant's transaction information to the computer device 200. The data server 120 may be an independent physical server with data storage function, or a server cluster or distributed system composed of multiple physical servers, which is used to batch store the transaction information of multiple merchants.

[0093] The method for identifying the risk of transaction information tampering provided by the embodiment of the present application can be performed by a computer device 200, which refers to an electronic device with data calculation, processing and storage capabilities. Figure 1 Taking the implementation environment of the scheme shown as an example, the method for identifying the risk of transaction information tampering can be executed by the identification terminal 210 (such as the client of the target application installed and running in the identification terminal 210 executes the method for identifying the risk of transaction information tampering), or the method for identifying the risk of transaction information tampering can be executed by the identification server 220, or the identification terminal 210 and the identification server 220 can interact and cooperate to execute it, and this application does not limit this.

[0094] The authentication terminal 210 may be a client of an electronic device such as a mobile phone, a tablet computer, a personal computer, a PDA, etc., in which a target application may be installed and run. The target application may be an authentication application for transaction information, or other application that provides an authentication function for transaction information. In addition, the present application does not limit the form of the target application, including but not limited to App (Application) installed in the terminal 100, applets, etc., and may also be in the form of a web page.

[0095] The authentication server 220 may be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The authentication server 220 may be a background server of the target application installed by the authentication terminal 210, and is used to provide background services for the client of the target application. The authentication server 220 may also be a server that independently performs authentication of transaction information.

[0096] In addition, the technical solution of the present application can be combined with blockchain technology. For example, in the identification method of transaction information tampering risk disclosed in the present application, some data involved (such as merchant transaction information and other data) can be stored on the blockchain, and accordingly, the data server 120 is a node server in the blockchain system. The merchant terminal 110, the data server 120 and the computer device 200 can communicate through a network, such as a wired or wireless network.

[0097] Figure 2 A schematic diagram of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application is provided. In this embodiment, a transaction picture 302 is used to indicate at least one transaction detail executed by an e-commerce store as a transaction object; a preset keyword 304 is retrieved from the transaction picture 302, and when the preset keyword 304 exists in the transaction picture 302, a sample label 305 corresponding to the preset keyword 304 is determined as the type label 303 of the transaction picture 302.

[0098] The transaction image 302 is subjected to image optimization processing 310 to obtain an optimized transaction image 302a, wherein the image optimization processing 310 includes at least one of grayscale processing, sharpening processing, and contrast enhancement processing. The optimized transaction image 302a is subjected to text recognition 320 to obtain transaction data 322 carried by the optimized transaction image 302a; the transaction data 322 includes table information, and the first transaction data corresponding to the first preset position 322a in the transaction image 302 is determined as feature description information 324b, the first preset position 322a is the header position of the table information, and the feature description information 324b is used to indicate the data attribute of the transaction feature representation 324a; the second transaction data corresponding to the second preset position 332b in the transaction image is determined as the transaction feature representation 324a, and the second preset position 332b and the first preset position 322a have the same vertical coordinate position.

[0099] According to the feature description information 324b and the type tag 303, the corresponding pre-set feature processing information 332 and reference identification feature 334 are obtained; wherein the feature processing information 332 is used to indicate the character editing method for the transaction feature representation 324a, and the reference identification feature 334 is an information feature in a secure transaction without the risk of tampering.

[0100] Based on the feature processing information 332, the transaction feature representation 324a is subjected to character editing processing to obtain the feature to be identified 326, which is a character string obtained based on the transaction feature representation and needs to be identified for tampering risk. In one example, the feature to be identified is a character format feature used to indicate the arrangement format characteristics of the transaction-related numerical value, such as: at least one of the thousandths of the transaction-related numerical value, the decimal point precision of the amount numerical value, and the currency symbol identifier of the amount numerical value; in another example, the feature to be identified is a data consistency feature used to indicate whether at least two transaction numerical features are the same, such as: at least one of the data consistency of the graphic part and the table part, the data consistency of the product of the transaction quantity and the transaction unit price and the transaction amount, and the data consistency between the merchant's growth level and the transaction quantity. The reference identification feature 334 and the feature to be identified 326 are subjected to a string comparison operation to obtain the identification result 328 of the transaction information.

[0101] Figure 3 A flowchart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application is shown. The method can be executed by a computer device. The method includes:

[0102] Step 510: Obtaining merchant transaction information;

[0103] Exemplarily, transaction information is used to indicate the transaction details of at least one sales transaction performed by the merchant; the transaction details include at least one of the transaction amount and the transaction quantity of the transaction items in the sales transaction performed by the merchant, and may also include but not be limited to at least one of the item information, transaction time, and transaction unit price of the transaction items. In one example, the merchant is an e-commerce merchant corresponding to the e-commerce platform, and the transaction information is the detailed information of the online transaction order provided by the e-commerce platform; in another example, the merchant is a physical merchant corresponding to a physical store, and the transaction record is the transaction record flow information provided by a financial institution, such as the flow information of the physical merchant's registered account; this application does not limit the form of the merchant's sales transaction.

[0104] Exemplarily, the acquired transaction information may be at least one of picture information, document information, and data table information. The present application does not impose any restrictive provisions on the format of the transaction information.

[0105] Step 520: performing feature extraction processing on the transaction information to obtain transaction feature representation and feature description information in the transaction information;

[0106] Exemplarily, the transaction feature representation is used to indicate the characteristics of the transaction resources in the sales transaction performed by the merchant, and the transaction feature representation includes but is not limited to a string consisting of at least one of numbers, symbols, Chinese characters, and letters. The feature description information is used to describe the attribute information of the transaction feature representation. Exemplarily, the corresponding positions of the transaction feature representation and the feature description information in the transaction information are usually different.

[0107] Step 530: Acquire feature processing information and reference identification features corresponding to the feature description information;

[0108] Exemplarily, the feature processing information is used to indicate a character editing method for representing a transaction feature; exemplary, through a pre-set correspondence, corresponding feature processing information and reference identification features are obtained according to the feature description information.

[0109] Exemplarily, the reference identification feature is an original information feature in a secure transaction, which is used to indicate information features in a secure transaction, such as at least one of the features of the arrangement format of numerical information and the consistency of at least two transaction numerical features. Exemplarily, a secure transaction can be a transaction executed by a merchant in a trusted whitelist, or a transaction determined by manually verifying transaction data; Exemplarily, a secure transaction is a transaction without the risk of information tampering.

[0110] Step 540: Based on the feature processing information, character editing processing is performed on the transaction feature representation to obtain features to be identified;

[0111] Exemplarily, based on the feature processing information indicating the character editing method, the character editing processing includes but is not limited to at least one of the following: extracting some characters in the transaction feature representation, performing statistical calculations on the digital part of the transaction feature representation, and editing some characters in the transaction feature representation.

[0112] Exemplarily, the feature to be identified is a character string that needs to be identified for tampering risk based on the transaction feature representation; the feature to be identified is obtained by character editing processing of the transaction feature representation, and the feature to be identified includes but is not limited to a character string composed of at least one of numbers, symbols, Chinese characters, and letters.

[0113] Step 550: Compare the reference identification feature and the feature to be identified to obtain an identification result of the transaction information;

[0114] Exemplarily, comparing the reference identification feature and the feature to be identified may be performing a string comparison operation on the character strings carried by the above two features, and comparing whether the character content and character order of the reference identification feature and the feature to be identified are the same; comparing the reference identification feature and the feature to be identified may also be comparing feature verification information of the hash value, parity check value, etc. of the reference identification feature and the feature to be identified, and comparing whether the feature verification information corresponding to the reference identification feature and the feature to be identified are the same; by comparing the reference identification feature and the feature to be identified, it is determined whether the transaction information is at risk of tampering.

[0115] Exemplarily, the reference authentication feature is an information feature in a secure transaction. When the reference authentication feature and the feature to be authenticated are the same, the authentication result is used to indicate that the transaction information is not at risk of tampering; when the reference authentication feature and the feature to be authenticated are different, the authentication result is used to indicate that the transaction information is at risk of tampering.

[0116] In summary, the method provided in this embodiment obtains characteristic information about transaction resources in the sales transaction in the transaction details by performing feature extraction processing on the transaction information, and expands the identification dimension of the transaction information to the transaction items and resource transfer conditions in the transaction details; performs character editing processing on the transaction feature representation to obtain the features to be identified, and realizes the editing and extraction of the character string that needs to be identified for tampering risk. Even if the tampering means of the transaction information does not leave explicit modification traces, the solution of the present application can accurately identify the tampering risk of the transaction information from the implicit information at the semantic level such as the transaction items and resource transfer conditions; that is, the solution of the present application does not rely on obvious modification traces to identify the tampering risk, but identifies the tampering risk through the inherent characteristics implicitly carried in the transaction information; it can identify the tampering risk caused by the modification made by the background controller during the transaction information generation stage, and increases the scope of application of effectively identifying the tampering risk of the transaction information.

[0117] Figure 4 A flowchart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application is shown. The method can be executed by a computer device. Figure 3 In the illustrated embodiment, step 540 may be implemented as step 542, and step 550 may be implemented as step 552:

[0118] Step 542: Based on the extraction indication information, character extraction processing is performed on the transaction feature representation to obtain character format features;

[0119] Exemplarily, the transaction feature representation is used to describe the transaction-related numerical value in the sales transaction, such as: at least one of the transaction unit price, transaction amount, transaction quantity, and transaction time of the sold item. It should be noted that the transaction feature representation in this embodiment is used to describe the transaction-related numerical value, but it is not limited to only including numbers in the transaction feature representation; for example, the transaction feature representation is the transaction unit price in the sales transaction, and the transaction feature representation is ￥13.99, and the transaction feature representation includes a decimal point symbol "." and a currency symbol "￥"; it can be seen that the transaction feature representation includes characters other than numbers, but can be used to indicate that the transaction unit price in the sales transaction is 13.99 yuan.

[0120] Exemplarily, the character format feature is used to indicate the arrangement format feature of the transaction-related numerical value; the character format feature is obtained based on some characters in the transaction feature representation. The arrangement format feature includes, but is not limited to, at least one of the length, line break position, digital precision, unit symbol, and separator symbol in the number of the transaction-related numerical value. The character format feature can be some characters in the transaction feature representation, such as the unit symbol and the separator symbol in the number; it can also be a quantitative description of the digital length, position information, and other information of the characters in the transaction feature representation. The character format feature extracts the arrangement format feature of the transaction-related numerical value and represents it as a number or a symbol.

[0121] In this embodiment, the feature processing information includes extraction instruction information, the extraction instruction information indicates a character extraction method for the transaction feature representation, and based on the extraction method indicated by the extraction instruction information, the transaction feature representation is processed to obtain the character format feature.

[0122] Step 552: performing a string comparison operation on the reference format feature and the character format feature to obtain an identification result;

[0123] Exemplarily, the reference authentication feature includes a reference format feature, and the reference format feature is used to indicate the arrangement format characteristics of the reference transaction value in the security transaction.

[0124] Exemplarily, in this embodiment, it is necessary to identify the difference between the extracted character format features and the reference identification features, and use the reference format features to indicate the arrangement format characteristics of the reference transaction value in the secure transaction, and then determine the identification result for the transaction-related value; for example, when the reference format features and the character format features are the same, the identification result is used to indicate that the transaction information does not have a risk of tampering; when the reference format features and the character format features are different, the identification result is used to indicate that the transaction information has a risk of tampering. Furthermore, the identification result is also used to indicate the type of tampering risk. The identification result in this embodiment is used to indicate whether the character format in the transaction information matches the pre-set format features, and whether the transaction information has a risk of tampering is determined by identifying the format information in the transaction information.

[0125] It should be noted that the string comparison operation can also be used to compare whether the reference format feature and the character format feature are the same, and can also be used to indicate whether the compared character format feature belongs to the reference format feature, or whether each character in the compared character format feature is the same as the reference format feature.

[0126] Since the character format feature quantitatively describes the arrangement format characteristics of the transaction information and is represented as numbers or symbols, string comparison operations can be used to determine whether the arrangement format characteristics of the transaction information and security information match, thereby obtaining the authentication result of the transaction information.

[0127] In one implementation, the string comparison operation is to compare the encoding information of two strings. Taking the string comparison operation between the reference format feature and the character format feature as an example, the reference format code corresponding to the reference format feature and the character format code corresponding to the character format feature are obtained based on the eight-bit Unicode Transformation Format-8 (UTF8) or the American Standard Code for Information Interchange (ASCII), and the string is encoded into binary or hexadecimal information; the binary or hexadecimal information is numerically compared. It should be noted that the present application does not limit the encoding method of the string.

[0128] In summary, the method provided in this embodiment obtains characteristic information about transaction resources in the sales transaction in the transaction details by performing feature extraction processing on the transaction information, and expands the identification dimension of the transaction information to the transaction items and resource transfer conditions in the transaction details; performs character extraction processing on the transaction feature representation to obtain the character format feature to be identified, and realizes the extraction of the arrangement format characteristics of the transaction-related numerical values. Even if the means of tampering with the transaction information do not leave explicit traces of modification, the scheme of the present application can accurately identify the tampering risk of the transaction information from the format features related to the character arrangement implicitly carried by the transaction information; that is, the scheme of the present application does not rely on obvious traces of modification to identify the tampering risk, but identifies the tampering risk through the characteristics of the character arrangement in the transaction information; performs string comparison operations on the reference format features and the character format features to determine whether the transaction information has a tampering risk; and identifies the character format differences generated when the transaction information is modified by the background controller during the transaction information generation stage, thereby increasing the scope of application of effectively identifying the tampering risk of the transaction information.

[0129] Next, the extraction instruction information is further introduced.

[0130] In an optional implementation of the present application, step 542 above can be implemented as at least one of sub-steps 21, 22, and 23. The extraction instruction information in step 542 is used to indicate the way to perform character extraction processing. The specific extraction instruction information indicates at least one of the following: the symbol style of the character to be extracted, the position of the character to be extracted, and the interval between the characters to be extracted. The character extraction processing will be further introduced in the sub-steps below:

[0131] Sub-step 21: Based on the first extraction instruction information, perform character extraction processing on the decimal point symbol in the transaction feature representation, and obtain the numerical precision of the transaction-related numerical value based on the position of the decimal point symbol;

[0132] Exemplarily, the first extraction indication information indicates a symbol style with a decimal point symbol, which is used to instruct character extraction processing of the decimal point symbol in the transaction feature representation; exemplary, the decimal point symbol is a full-width symbol or a half-width symbol, and the interval between the integer part and the decimal part of the number needs to be increased based on the information layout. The decimal point symbol is a period symbol and a space symbol, such as: ".".

[0133] Exemplarily, the decimal point symbol is subjected to character extraction processing in the transaction feature representation through the symbol style of the decimal point symbol indicated by the first extraction indication information. In one implementation, the symbol style of the decimal point symbol corresponds to symbol encoding information, and the symbol encoding information is searched in the transaction encoding information corresponding to the transaction feature representation to determine the position of the decimal point symbol in the transaction feature representation. The symbol encoding information and the transaction encoding information are encoded based on UTF8 or ASCII, and characters such as the decimal point symbol ".", letters or numbers are encoded as binary or hexadecimal information. This application does not limit the encoding method of the character information.

[0134] Furthermore, the numerical precision of the transaction-related value is determined according to the position of the decimal point symbol in the transaction feature representation. For example, if the transaction feature representation is: ￥13.99, the number of digits after the decimal point symbol "." is 2, and the numerical precision of the transaction-related value is two decimal places, also known as a precision of 10 -2 In one implementation, when no decimal point is found in the transaction feature representation, the numerical precision of the transaction-related value is determined to be units, also known as a precision of 1 or a precision of 10. 0 .

[0135] Accordingly, the reference format feature is used to indicate the reference precision information in the security transaction. For example, when the feature description information is the transaction amount or the transaction unit price, the reference precision information indicated by the reference format feature corresponding to the feature description information is two decimal places. When the feature description information is the transaction quantity, the reference precision information indicated by the reference format feature corresponding to the feature description information is the unit place.

[0136] Sub-step 22: Based on the second extraction instruction information, character extraction processing is performed at the head position or the tail position of the transaction feature representation to obtain the currency symbol of the transaction-related value;

[0137] Exemplarily, the extraction position of the currency symbol is indicated by the second extraction indication information, and the currency symbol is subjected to character extraction processing in the transaction feature representation. The extraction position indicated by the second extraction indication information is the head position or the tail position of the transaction feature representation, and the extracted transaction currency symbol includes but is not limited to any one of: "￥", "€", "￥", and "$".

[0138] Accordingly, the reference format feature is used to indicate a reference symbol in a security transaction. For example, when the feature description information contains a country or region name and the feature description information also includes a transaction amount or a transaction unit price, the reference symbol in the reference format feature corresponds to the country or region name.

[0139] Sub-step 23: Based on the third extraction instruction information, character extraction processing is performed on the transaction feature representation to obtain a separator of the transaction-related numerical value;

[0140] Exemplarily, the preset interval of the extraction separator is indicated by the third extraction instruction information, and the separator in the transaction feature representation is subjected to character extraction processing. In one example, the separator is at least one of a comma symbol or a space symbol, and the separator is a thousands separator. By adding a thousands separator to a number, it is possible to avoid difficulty in distinguishing the numerical value due to a large number of digits; Exemplarily, the preset interval of the thousands separator is three digits; Further, the length of the number before the decimal point exceeds three digits, and the interval between two adjacent thousands separators is three digits, and the interval between the first thousands separator and the decimal point at the end of the number is three digits.

[0141] Accordingly, the reference format feature is used to indicate the symbol style of the separator symbol in the security transaction. Exemplarily, the identification result is determined by comparing whether the symbol style in the reference format feature and the extracted separator symbol are the same.

[0142] In summary, the method provided in this embodiment obtains characteristic information about transaction resources in the sales transaction in the transaction details by performing feature extraction processing on the transaction information, and expands the identification dimension of the transaction information to the transaction items and resource transfer conditions in the transaction details; performs character extraction processing on the transaction feature representation to obtain the character format features to be identified. Even if the means of tampering with the transaction information do not leave explicit traces of modification, the solution of the present application can accurately identify the risk of tampering with the transaction information from the character format features such as the numerical precision implicitly carried by the transaction information, the currency symbol of the transaction-related numerical value, and the separator symbol of the transaction-related numerical value; performs string comparison operations on the reference format features and the character format features to determine whether the transaction information has a risk of tampering; and effectively identifies the character format differences generated when the transaction information is modified by the background controller during the transaction information generation stage, thereby increasing the scope of application of effectively identifying the risk of tampering with the transaction information.

[0143] In a specific example, Figure 5 A schematic diagram of transaction information provided by an exemplary embodiment of the present application is shown. The transaction information carries a first transaction table 410, which includes 4 rows and 5 columns of data, and is used to indicate the transaction details of three sales transactions numbered 1 to 3 executed by the merchant.

[0144] Taking the transaction details with serial number 1 as an example, this embodiment will introduce the identification process of the tampering risk of the third column information 413, the fourth column information 414 and the fifth column information 415 in the first transaction table 410; any one of the above three information corresponds to transaction feature representation and feature description information; in an example, the transaction feature representation and feature description information are information corresponding to the preset position of the first transaction table 410. Exemplarily, the feature description information corresponds to the header position of the first transaction table 410, and the preset position and the header position corresponding to the transaction feature representation have the same vertical coordinate. The detailed introduction of determining the transaction feature representation and feature description information in the transaction information will be introduced in a separate embodiment below after this embodiment.

[0145] The feature description information of the third column 413 of the first transaction table 410 is the transaction unit price, and the transaction feature is represented as ￥55.0. After character extraction, the numerical precision is one decimal place and the currency symbol is "￥".

[0146] Specifically, the first extraction instruction information is used to instruct to perform character extraction processing on the transaction feature representation containing numbers, that is, to instruct to search for the decimal point symbol "." in the transaction feature representation ￥55.0, and according to the number of digits after the decimal point symbol being one, the numerical precision is determined to be one decimal place, and the numerical precision of the third column information 413 is recorded as 10. -1 In one example, when the feature description information is the transaction amount or the transaction unit price, the reference precision information indicated by the reference format feature corresponding to the feature description information is two decimal places, that is, the reference format feature is recorded as 10. -2 . Perform a string comparison operation on the reference format feature and the character format feature describing the numerical precision of the third column information 413 to obtain 10 -1 and 10 -2 There are differences in character strings, and the identification result of the numerical precision of the third column information 413 is that there is a risk of tampering.

[0147] The second extraction instruction information is used to instruct the character extraction processing of the transaction feature representation including the transaction amount or the transaction unit price of the feature description information, that is, to instruct the extraction of the first character at the head position of the transaction feature representation ￥55.0, and to extract the currency symbol "￥". In one example, the reference currency symbols indicated by the reference format feature corresponding to the feature description information include "￥", "€", "￥" and "$". The reference format feature and the character format feature describing the currency symbol in the third column of information 413 are compared by a string comparison operation, and the currency symbol "￥" in the third column of information 413 is obtained in the reference format feature. The identification result of the currency symbol in the third column of information 413 is that there is no risk of tampering. In one example, the string comparison operation is to compare the encoding information of the four currency symbols "￥", "€", "￥" and "$" in the reference format feature based on the UTF8 method, to see whether there is the encoding information of the currency symbol "￥" extracted from the third column of information 413.

[0148] The third extraction instruction information is used to instruct to perform character extraction processing on the transaction feature representation with digits, that is, to instruct to perform character extraction processing on the transaction feature representation with a digit length of more than three digits before the decimal point when the transaction feature representation has a decimal point; or to perform character extraction processing on the transaction feature representation with a digit length of more than three digits when the transaction feature representation does not have a decimal point. Since the transaction feature representation of ￥55.0 in the third column of information 413 has a decimal point, and the digit length before the decimal point is two digits, it is not necessary to perform character extraction processing based on the third extraction instruction information, and it is not necessary to identify the separator.

[0149] The feature description information of the fourth column 414 of the first transaction table 410 is the number of items, and the transaction feature is represented as 1,031. After character extraction, the numerical precision is the unit digit and the separator is “,”.

[0151] Specifically, the first extraction instruction information is used to instruct to perform character extraction processing on the transaction feature representation containing numbers, similar to the character extraction processing of the third column information 413. The numerical precision of the fourth column information 414 is recorded as 10 0 In one example, when the feature description information is the transaction amount, the reference precision information of the reference format feature corresponding to the feature description information is the unit digit. That is, the reference format feature is recorded as 10 0 Perform a string comparison operation on the reference format feature and the character format feature describing the numerical precision of the fourth column information 414 to obtain 10 0 and 10 0 There is no difference in the character strings, and the identification result of the numerical precision of the fourth column information 414 is that there is no risk of tampering.

[0152] The second extraction indication information is used to instruct character extraction processing to be performed on the transaction feature representation whose feature description information includes the transaction amount or the transaction unit price; since the feature description information corresponding to the fourth column information 414 does not contain the transaction amount or the transaction unit price, there is no need to perform character extraction processing based on the second extraction indication information, and there is no need to identify the currency symbol.

[0153] The third extraction instruction information is used to instruct character extraction processing for transaction feature representations with numbers. As described above, since the transaction feature representation of the fourth column of information 414 is 1,031; there is no decimal point, and the number length exceeds three digits. Based on the third extraction instruction information, character extraction processing is performed at intervals of three digits, that is, starting from the first digit at the end of the transaction feature representation, the characters before the three digits are extracted, that is, the character "," before "031" is extracted; since the number before the character "," is one digit long, which is less than the interval distance of three digits, the character extraction is stopped. In one example, the reference delimiter symbol indicated by the reference format feature corresponding to the feature description information is ",". A string comparison operation is performed on the reference format feature and the character format feature describing the delimiter symbol in the fourth column of information 414, and each character in the character format feature is obtained to be the same as the reference format feature. The identification result of the delimiter symbol in the fourth column of information 414 is that there is no risk of tampering.

[0154] The characteristic description information of the fifth column 415 of the first transaction table 410 is the transaction amount, and the transaction characteristic is represented as ￥56,705.00. After character extraction, the numerical precision is two decimal places, the currency symbol is "￥", and the separator is ",".

[0155] Specifically, the first extraction instruction information is used to instruct to perform character extraction processing on the transaction feature representation containing numbers, similar to the character extraction processing of the third column information 413. The numerical precision of the fifth column information 415 is recorded as 10. -2 In one example, when the feature description information is the transaction amount or the transaction unit price, the reference precision information indicated by the reference format feature corresponding to the feature description information is two decimal places, that is, the reference format feature is recorded as 10. -2 Perform a string comparison operation on the reference format feature and the character format feature describing the numerical precision of the fifth column information 415 to obtain 10 -2 and 10 -2 There is no difference in the character strings, and the identification result of the numerical precision of the fifth column information 415 is that there is no risk of tampering.

[0156] The second extraction instruction information is used to instruct to perform character extraction processing on the transaction feature representation of the feature description information including the transaction amount or the transaction unit price. Similar to the character extraction processing of the third column of information 413, the fifth column of information 415 extracts the currency symbol "￥". In one example, the reference currency symbols indicated by the reference format feature corresponding to the feature description information include "￥", "€", "￥" and "$". A character string comparison operation is performed on the reference format feature and the character format feature describing the currency symbol in the fifth column of information 415, and the currency symbol "￥" in the fifth column of information 415 is obtained in the reference format feature. The identification result of the currency symbol in the fifth column of information 415 is that there is no risk of tampering.

[0157] The third extraction instruction information is used to instruct character extraction processing for transaction feature representations that contain numbers. As described above, since the transaction feature representation of the fifth column information 415 is RMB 56,705.00; there is a decimal point, and the length of the number before the decimal point exceeds three digits, character extraction processing is performed at intervals of three digits based on the third extraction instruction information, that is, starting from the first digit before the decimal point of the transaction feature representation, the characters before the three digits are extracted, that is, the characters "," before "075" are extracted; since the length of the number before the character "," is two digits, the interval distance is less than three digits, so the character extraction is stopped. It should be noted that, in the case where the length of the number before the character "," exceeds three digits, the character extraction processing continues, and the characters before the three digits continue to be extracted until the length of the number before the extracted character does not exceed three digits. In one example, the reference separator symbol indicated by the reference format feature corresponding to the feature description information is ",". A string comparison operation is performed on the reference format feature and the character format feature describing the separator symbol in the fifth column of information 415, and each character in the character format feature is found to be the same as the reference format feature. The identification result of the separator symbol in the fifth column of information 415 is that there is no risk of tampering.

[0158] Figure 6 A flowchart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application is shown. The method can be executed by a computer device. Figure 3 In the illustrated embodiment, step 540 may be implemented as step 544 and step 545, and step 550 may be implemented as step 554:

[0159] Step 544: Based on the statistical indication information, statistical calculation processing is performed on at least two transaction feature representations to obtain at least two transaction numerical features;

[0160] Exemplarily, the feature processing information includes statistical indication information, and the statistical indication information indicates a statistical calculation method and a numerical comparison method for the transaction feature representation; it should be noted that the statistical indication information corresponds to at least two feature description information. The statistical calculation processing is to process the transaction feature representation corresponding to the at least two feature description information by at least one of numerical extraction, basic four arithmetic operations, statistical data analysis, etc., to obtain at least two transaction numerical features. The at least two feature description information in this embodiment can be the same or different, and this embodiment does not limit this.

[0161] Exemplarily, the transaction value feature can be information explicitly carried in the transaction information, such as the transaction value feature is the total transaction amount extracted from the transaction information; or it can be information implicitly carried in the transaction information, such as the transaction value feature is the product of the transaction quantity and the transaction unit price in the transaction information. This application does not limit the way in which the transaction value feature is carried in the transaction information.

[0162] Exemplarily, the statistical calculation processing may be to process at least two transaction feature representations separately, such as: extracting the numerical part of the transaction feature representation; or it may be to process at least two transaction feature representations collaboratively, such as: calculating the sum of at least two transaction feature representations; or the above two methods may be combined; the present application does not limit the method of statistical calculation processing.

[0163] Step 545: Based on the statistical indication information, perform a numerical comparison operation on at least two transaction numerical features to obtain a data consistency feature;

[0164] Exemplarily, the data consistency feature is used to indicate whether at least two transaction numerical features are the same; Exemplarily, the data consistency feature can be represented as a number, such as the first feature value is used to indicate that at least two transaction numerical features are the same, the second feature value is used to indicate that at least two transaction numerical features are different, or the difference between two transaction numerical features or the standard deviation between at least three transaction numerical features is determined as the data consistency feature. The data consistency feature can also be represented as a string, Exemplarily, the first feature field (such as true) is used to indicate that at least two transaction numerical features are the same, and the second feature field (such as false) is used to indicate that at least two transaction numerical features are different; This application does not limit the representation method of the data consistency feature.

[0165] Step 554: performing a string comparison operation on the reference consistency feature and the data consistency feature to obtain an identification result;

[0166] Exemplarily, the reference identification feature includes a reference consistency feature, and the reference consistency feature is used to indicate the consistency of at least two transaction numerical features of a secure transaction. The reference consistency feature is a consistency condition between at least two transaction numerical features in a secure transaction. Exemplarily, the reference consistency feature includes, but is not limited to, at least one of the following: at least two transaction numerical features are the same, different, the difference between the two transaction numerical features is less than a first preset threshold, and the standard deviation between the three transaction numerical features is less than a second preset threshold.

[0167] Exemplarily, in this embodiment, it is necessary to identify the difference between the data consistency feature obtained by comparison and the reference consistency feature, and use the reference consistency feature to indicate the consistency of at least two transaction numerical features in the secure transaction, and then determine the identification result for the transaction feature representation; for example, when the reference consistency feature and the data consistency feature are the same, the identification result is used to indicate that the transaction information does not have a tampering risk; when the reference consistency feature and the data consistency feature are different, the identification result is used to indicate that the transaction information has a tampering risk, and further, the identification result is also used to indicate the tampering risk type. The identification result in this embodiment is used to indicate whether there is a logical problem directly between at least two transaction feature representations implicitly or explicitly carried in the transaction information, and to determine whether the transaction information has a tampering risk by mining the connection between the transaction feature representations.

[0168] In summary, the method provided in this embodiment obtains characteristic information about transaction resources in the sales transaction in the transaction details by performing feature extraction processing on the transaction information, and expands the identification dimension of the transaction information to the transaction items and resource transfer in the transaction details; performs statistical calculation processing on the transaction feature representation to obtain the data consistency feature to be identified, and realizes the determination of whether at least two transaction numerical features are the same. Even if the tampering means of the transaction information does not leave explicit modification traces, the scheme of the present application can also accurately identify the tampering risk of the transaction information from the consistency feature of whether at least two transaction numerical features implicitly carried by the transaction information are the same; that is, the scheme of the present application does not rely on obvious modification traces to identify the tampering risk, but identifies the tampering risk by whether different transaction feature representations in the transaction information are consistent; performs string comparison operations on the reference consistency feature and the data consistency feature to determine whether the transaction information has a tampering risk; identifies the data consistency differences generated when the transaction information is modified by the background controller in the transaction information generation stage, thereby increasing the scope of application of effectively identifying the tampering risk of the transaction information.

[0169] Next, the statistical indication information is further introduced.

[0170] In an optional implementation of the present application, there are at least three implementation methods for statistical indication information:

[0171] Implementation method 1: Based on the statistical indication information, numerical value extraction processing is performed on at least two transaction feature representations one by one;

[0172] Implementation method 2: Based on the statistical indication information, numerical calculation processing and numerical extraction processing are performed on at least three transaction feature representations;

[0173] Implementation method three: based on the statistical indication information, searching for the mapping value range corresponding to the first transaction feature representation, and performing value extraction processing on the second transaction feature representation;

[0174] The following three implementation methods are described one by one through embodiments:

[0175] Implementation method 1: Based on the statistical indication information, numerical value extraction processing is performed on at least two transaction feature representations one by one;

[0176] In implementation mode 1, step 544 above can be implemented as sub-step 41:

[0177] Sub-step 41: Based on the statistical indication information, perform value extraction processing on at least two transaction feature representations one by one to obtain at least two transaction value features;

[0178] Exemplarily, the statistical indication information in this embodiment is used to perform numerical extraction processing on data features located at different positions in the transaction information; accordingly, by performing numerical comparison operations on at least two transaction numerical features, the data consistency feature obtained is used to indicate the data consistency of at least two transaction numerical features explicitly carried at different positions in the transaction information. For example, the transaction information includes a statistical table and a statistical line graph, and the data consistency feature is used to indicate whether at least two transaction feature representations with the same attribute information are consistent in the statistical table and the statistical line graph.

[0179] Exemplarily, at least two transaction feature representations correspond to the same feature description information, and at least two transaction feature representations both have numerical features. Exemplarily, the statistical indication information is used to indicate a processing method for numerical extraction processing, such as extracting a numeric part from a character string, or extracting a transaction numerical feature based on at least one of the horizontal coordinate information or the vertical coordinate information of a feature point, which is not limited in this embodiment.

[0180] In one implementation, the numerical extraction process is to search for numerical codes corresponding to the numbers 0 to 9 in the transaction feature code corresponding to the transaction feature representation, so as to extract the numerical part of the transaction feature representation. Exemplarily, the transaction feature code can be encoded based on UTF8 or ASCII.

[0181] Exemplarily, the at least two transaction value features and the at least two transaction feature representations correspond one to one;

[0182] In this embodiment, a string comparison operation is performed on the reference consistency feature and the data consistency feature, and the reference consistency feature is used to indicate the data consistency of transaction features having the same feature description information in a secure transaction at different locations.

[0183] In one example, Figure 7 A schematic diagram of transaction information provided by an exemplary embodiment of the present application is shown. The transaction information carries sales information 420, and the sales information 420 includes overview information 422 and line chart information 424; the overview information 422 is used to indicate the sales information of the e-commerce merchant in this year. The line chart information 424 is used to describe the sales information of each month in this year through a dot-line graph. Further, the line chart information 424 compares the information of this year and the information of the previous year through different line types. Exemplarily, the line chart information 424 shows a dot-line graph of the sales quantity of items in each month. Exemplarily, the line chart 424 also includes a comparison information description 425. A numerical comparison operation is performed on the transaction numerical feature corresponding to the first transaction feature representation 422a in the overview information 422 and the second transaction feature representation 425a in the comparison information description 425 to obtain a data consistency feature. Determine the data consistency of at least two transaction numerical features explicitly carried at different positions in the transaction information.

[0184] In summary, the method provided in this embodiment obtains characteristic information about transaction resources in the sales transaction in the transaction details by performing feature extraction processing on the transaction information, and expands the identification dimension of the transaction information to the transaction items and resource transfer conditions in the transaction details; performs numerical extraction processing on the transaction feature representation to obtain the data consistency feature to be identified, thereby realizing the data consistency of at least two transaction numerical features explicitly carried at different positions in the transaction information; performs string comparison operation on the reference consistency feature and the data consistency feature to determine whether the transaction information has the risk of tampering; and identifies the data consistency differences generated when the transaction information is modified by the background controller during the transaction information generation stage, thereby improving the accuracy of identification of the transaction information.

[0185] Implementation method 2: Based on the statistical indication information, numerical calculation processing and numerical extraction processing are performed on at least three transaction feature representations;

[0186] In the second implementation, step 544 above can be implemented as sub-step 42a and sub-step 42b:

[0187] Sub-step 42a: performing numerical calculation processing on at least two transaction feature representations based on the operation information in the statistical indication information to obtain a first transaction numerical feature;

[0188] Exemplarily, the statistical indication information in this embodiment is used to indicate the numerical calculation process and the numerical extraction process; accordingly, the data consistency feature is obtained by performing a numerical comparison operation on two transaction numerical features. The data consistency feature is used to indicate the data consistency of the first transaction numerical feature implicitly carried in the transaction information and the second transaction numerical feature explicitly carried. For example: the first transaction numerical feature is the product of the transaction quantity and the transaction unit price, and the second transaction numerical feature is the transaction amount; the data consistency feature is used to indicate whether at least three transaction feature representations with different attribute information are consistent in the transaction information.

[0189] Exemplarily, the statistical indication information is used to perform statistical calculation processing on at least three transaction feature representations, and the at least three transaction feature representations have different feature description information. The statistical calculation processing includes numerical calculation processing and numerical extraction processing. Among them, the first transaction numerical feature is obtained by numerical calculation processing indicated by the operation information. The first transaction numerical feature is used to indicate the transaction numerical value implicitly carried by at least two transaction feature representations in the transaction information. The transaction numerical value implicitly carried in at least two transaction feature representations needs to be processed by statistics, comparison, operation, etc. on the at least two transaction feature representations to obtain the transaction numerical value. Exemplarily, as shown above, the first transaction numerical feature is the product of two transaction feature representations, and the feature description information corresponding to the two transaction feature representations is the transaction quantity and the transaction unit price.

[0190] Sub-step 42b: performing numerical extraction processing on a transaction feature representation based on the extraction information in the statistical indication information to obtain a second transaction numerical feature;

[0191] Exemplarily, the second transaction value feature is obtained by extracting the value indicated by the extraction information, and the second transaction value feature is used to indicate the transaction value explicitly carried by a transaction feature representation in the transaction information, and the transaction value explicitly carried by the transaction feature representation is directly recorded in the transaction feature representation, which is part or all of the characters of the transaction feature representation. Exemplarily, as shown above, the second transaction value feature is the digital part of the transaction feature representation corresponding to the transaction amount in the feature description information.

[0192] In this embodiment, a string comparison operation is performed on the reference consistency feature and the data consistency feature, and the reference consistency feature is used to indicate the data consistency between the transaction value implicitly carried and the transaction value explicitly carried in the secure transaction.

[0193] In one example, Figure 8A schematic diagram of transaction information provided by an exemplary embodiment of the present application is shown. The transaction information carries a second transaction table 430, which includes 2 rows and 5 columns of data, and is used to indicate the transaction details of a sales transaction with a serial number of 4 executed by an e-commerce merchant. Exemplarily, the third transaction feature representation 433 and the fourth transaction feature representation 434 in the second transaction table 430 are multiplied, that is, the operation information in the statistical indication information is used to indicate the multiplication operation of the two transaction feature representations corresponding to the feature description information of the transaction quantity and the transaction unit price, and the first transaction numerical feature is obtained as 199.99*314=62796.86. The fifth transaction feature representation 435 in the second transaction table 430 is subjected to numerical extraction processing, that is, the extraction information in the statistical indication information is used to indicate the numerical extraction processing of the transaction feature representation whose feature description information is the transaction amount, and the second transaction numerical feature is obtained as 62796.86. By comparing the first transaction numerical feature and the second transaction numerical feature, the data consistency of the first transaction numerical feature implicitly carried in the transaction information and the second transaction numerical feature explicitly carried is determined.

[0194] In an optional implementation, the ratio of the transaction feature representation corresponding to the total sales of goods (Ordered product sales) and the transaction feature representation corresponding to the total number of sales orders (Total order items) is calculated to be determined as the first transaction numerical feature; the transaction feature representation corresponding to the average sales per order (Avg. sales order item) is subjected to numerical extraction processing to be determined as the second transaction numerical feature. That is, the operation information in the statistical indication information is used to indicate the division operation of the two transaction feature representations corresponding to the feature description information of the total sales of goods and the total number of sales orders, and the extraction information in the statistical indication information is used to indicate the numerical extraction processing of the transaction feature representation whose feature description information is the average sales per order.

[0195] In another optional implementation, the ratio of the transaction feature representation corresponding to the total number of units sold (Units ordered) and the transaction feature representation corresponding to the total number of sales orders is determined as the first transaction numerical feature; the transaction feature representation corresponding to the average number of units sold per order (Avg. units order item) is subjected to numerical extraction processing and determined as the second transaction numerical feature. That is, the operation information in the statistical indication information is used to indicate the division operation of the two transaction feature representations corresponding to the feature description information of the total number of units sold and the total number of sales orders, and the extraction information in the statistical indication information is used to indicate the numerical extraction processing of the transaction feature representation corresponding to the feature description information of the average number of units sold per order.

[0196] In summary, the method provided in this embodiment obtains characteristic information about transaction resources in the sales transaction in the transaction details by performing feature extraction processing on the transaction information, and expands the identification dimension of the transaction information to the transaction items and resource transfer conditions in the transaction details; performs numerical calculation processing and numerical extraction processing on the transaction feature representation to obtain the data consistency feature to be identified, thereby realizing the data consistency of the first transaction numerical feature implicitly carried and the second transaction numerical feature explicitly carried in the transaction information; performs string comparison operation on the reference consistency feature and the data consistency feature to determine whether the transaction information has a risk of tampering; and identifies the data consistency differences generated when the transaction information is modified by the background controller during the transaction information generation stage, thereby improving the accuracy of identifying the transaction information.

[0197] Implementation method three: based on the statistical indication information, searching for the mapping value range corresponding to the first transaction feature representation, and performing value extraction processing on the second transaction feature representation;

[0198] In implementation mode 3, step 544 in the above text can be implemented as sub-step 43a and sub-step 43b, and step 545 can be implemented as sub-step 43c:

[0199] Sub-step 43a: based on the numerical mapping relationship in the statistical indication information, searching for a mapping numerical range corresponding to the first transaction feature representation;

[0200] Exemplarily, the statistical indication information in this embodiment is used to indicate the numerical mapping relationship and the numerical extraction process; accordingly, the data consistency feature is obtained through the numerical comparison operation. The data consistency feature is used to indicate the numerical attribution relationship between the mapping numerical range implicitly carried by the first transaction feature representation and the third transaction numerical feature explicitly carried by the second transaction feature representation. For example: the first transaction feature representation is the growth level of the merchant, implicitly carrying the numerical range of the transaction quantity; the second transaction feature representation is the transaction quantity, explicitly carrying the transaction feature representation of the transaction quantity; the data consistency feature is used to indicate the numerical attribution relationship between the numerical range and the third transaction numerical feature extracted from the transaction feature representation.

[0201] Exemplarily, at least two transaction feature representations correspond to the same feature description information, and no numerical feature exists in the first transaction feature representation of at least two transaction feature representations; the numerical mapping relationship is used to indicate the candidate numerical range corresponding to the candidate feature representation, and the first transaction feature representation is searched in the candidate feature representations, and the corresponding candidate numerical range is determined as the mapping numerical range corresponding to the first transaction feature representation.

[0202] Sub-step 43b: performing numerical extraction processing on the second transaction feature representation based on the extraction information in the statistical indication information to obtain a third transaction numerical feature;

[0203] Exemplarily, the third transaction value feature is obtained by extracting the value indicated by the extraction information, and the third transaction value feature is used to indicate the transaction value explicitly carried by the second transaction feature in the transaction information.

[0204] Sub-step 43c: Based on the statistical indication information, perform a numerical comparison operation on the mapped numerical range and the third transaction numerical feature to obtain a data consistency feature;

[0205] Exemplarily, the data consistency feature is used to indicate whether the third transaction value belongs to the mapped value range; in this embodiment, a string comparison operation is performed on the reference consistency feature and the data consistency feature, and the reference consistency feature is used to indicate the numerical attribution relationship between the reference value range and the reference transaction value in the secure transaction.

[0206] In one example, Fig. 9A schematic diagram of transaction information provided by an exemplary embodiment of the present application is shown. The transaction information carries the introduction information 441 of the e-commerce merchant, the growth level 442 of the e-commerce merchant, and the sales volume 443 of the e-commerce merchant; based on the numerical mapping relationship, the mapping numerical range 445 corresponding to the "diamond"-level high-quality store indicated by the growth level 442 of the e-commerce merchant is found to be 1,300 to 2,000 pieces of monthly sales. A numerical comparison operation is performed on the numerical parts extracted from the mapping numerical range 445 and the sales volume 443 of the e-commerce merchant, and it is determined that the mapping numerical range 445 implicitly carried by the growth level 442 of the e-commerce merchant (i.e., the monthly sales volume is 1,300 to 2,000 pieces) contains the numerical feature explicitly carried by the sales volume 443 of the e-commerce merchant, i.e., 1,536 items of goods are sold per month; in the case where there is a numerical attribution relationship between the mapping numerical range 445 and the sales volume 443 of the e-commerce merchant, the data consistency feature is recorded as the first value. Correspondingly, the reference consistency feature is used to indicate whether there is a numerical attribution relationship in a secure transaction. When the value of the reference consistency feature is the same as the value of the data consistency feature, there is no risk of tampering with the transaction information; otherwise, there is a risk of tampering with the transaction information. In summary, the method provided in this embodiment obtains characteristic information about transaction resources in the sales transaction in the transaction details by performing feature extraction processing on the transaction information, and expands the identification dimension of the transaction information to the transaction items and resource transfer in the transaction details; performs search processing and numerical extraction processing on the transaction feature representation to obtain the data consistency feature to be identified, and realizes the determination of the numerical attribution relationship between the mapping numerical range implicitly carried by the first transaction feature representation and the third transaction numerical feature explicitly carried by the second transaction feature representation; performs string comparison operations on the reference consistency feature and the data consistency feature to determine whether the transaction information has a risk of tampering; and identifies the data consistency differences generated when the transaction information is modified by the background controller during the transaction information generation stage, thereby improving the accuracy of identifying the transaction information.

[0207] Fig.10 A flowchart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application is shown. The method can be executed by a computer device. Figure 3 Based on the illustrated embodiment, step 515 is also included:

[0208] Step 515: Obtain the type tag of the transaction information;

[0209] Exemplarily, the type tag is used to indicate the information statistics method for at least one sales transaction; the information statistics method includes but is not limited to at least one of the display method of the transaction information and the time range of the transaction information. Exemplarily, the type tag of the transaction information can be directly obtained or determined based on the transaction information. This application does not limit the determination method of the type tag.

[0210] Further, the feature processing information in this embodiment is determined based on the feature description information and the type tag. Through the pre-set correspondence, the corresponding feature processing information is found according to the feature description information and the type tag. For the introduction of the feature processing information, please refer to the various embodiments above, which will not be repeated here.

[0211] In one example, Fig.11 The diagram shows a schematic diagram of transaction information provided by an exemplary embodiment of the present application. The first transaction information 450a shown in the figure indicates at least one sales transaction performed by the merchant in a table format. The merchant sells multiple transaction items. In the first transaction information 450a, the transaction items are not distinguished one by one, but time is used as a statistical dimension to determine the sales transactions occurring in different time periods. For example, the above Figure 5 Also shown is at least one sales transaction performed by the merchant in a table format, with different transaction items as statistical dimensions to display at least one sales transaction. Figure 7 It also shows that at least one sales transaction is displayed in a manner of comparing the information of this year with the information of the previous year through a line graph.

[0212] In an optional implementation, step 515 may be implemented as follows:

[0213] By comparing the transaction information and the sample information, similarity information for the sample information is obtained;

[0214] When the similarity information exceeds the trust threshold, the sample label of the sample information is determined as the type label of the transaction information;

[0215] Exemplarily, the sample label corresponding to the sample information is used to indicate the statistical method of at least one sales transaction in the sample information. Exemplarily, this embodiment can directly compare the similarity between the transaction information and the sample information; it can also train an artificial neural network (ANN) based on the sample information, and predict the transaction information by calling the artificial neural network to achieve direct and indirect comparison between the transaction information and the sample information; this embodiment does not limit this.

[0216] Exemplarily, the trust threshold is a trust lower limit for determining that the transaction information and the sample information have the same type label, and the sample label of the sample information is determined as the type label of the transaction information to achieve determination of the type label of the transaction information.

[0217] In another optional implementation, step 515 may be implemented as follows:

[0218] Search for preset keywords in transaction information;

[0219] When there is a preset keyword in the transaction information, the sample label corresponding to the preset keyword is determined as the type label of the transaction information;

[0220] Exemplarily, the preset keyword corresponds to a sample tag, and the preset keyword is a key string that appears in the transaction information corresponding to the sample tag. The number of preset keywords can be one or more, and the preset environment words are usually preset based on experience, but it is not excluded that they are determined by statistical methods.

[0221] Exemplarily, when there is a preset keyword in the search for transaction information, the transaction information and the sample information have the same type tag, and the sample tag of the sample information is determined as the type tag of the transaction information to achieve determination of the type tag of the transaction information.

[0222] To summarize, the method provided in this embodiment obtains the type label of the transaction information, jointly determines the feature processing information and the reference identification feature based on the type label and the feature description information, performs targeted processing on the transaction information based on the type label, and realizes the editing and extraction of the character string that needs to be identified for tampering risk; performs string comparison operations on the reference identification feature and the feature to be identified to determine whether the transaction information has a tampering risk; and can identify the tampering risk caused by the modification made by the background controller during the transaction information generation stage, that is, it realizes the accurate identification of the tampering risk of the transaction information when there is no modification trace on the transaction information, increases the scope of application of the effective identification of the tampering risk of the transaction information, and improves the accuracy of the identification of the transaction information.

[0223] Fig.12 A flowchart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application is shown. The method can be executed by a computer device. Figure 3 In the illustrated embodiment, the transaction information includes at least one transaction picture of transaction details, and step 520 can be implemented as steps 522 and 524:

[0224] Step 522: Perform text recognition on the transaction image to obtain transaction data carried by the transaction image;

[0225] For example, the transaction image carries transaction data, and through character recognition (Character Recognition), the string information carried in the transaction image is obtained. In this embodiment, character recognition is usually implemented based on an artificial neural network, but it does not exclude the case where character recognition is implemented through similarity comparison.

[0226] Step 524: determining transaction feature representation and feature description information in the target transaction data corresponding to the preset position in the transaction image;

[0227] Exemplarily, the transaction feature representation and feature description information are part or all of the transaction data in the transaction image. In one implementation, the preset position is usually a picture position preset based on experience. Further, the transaction image has a type label, and different type labels have different preset positions. For the introduction of type labels, please refer to step 515 above, which will not be repeated here.

[0228] In an optional implementation, the transaction image includes table information, and step 524 can be implemented as follows:

[0229] Determine the first transaction data corresponding to the first preset position in the transaction image as feature description information;

[0230] ·determine the second transaction data corresponding to the second preset position in the transaction image as the transaction feature representation;

[0231] Exemplarily, the first preset position is the header position of the table information, and the feature description information is used to indicate the attribute information represented by the transaction feature; the second preset position and the first preset position have the same horizontal coordinate position or vertical coordinate position. Exemplarily, the header position carries an explanation of the meaning of the information in the table and / or a sorting sequence number of the information in the table, and determining the first transaction data at the header position as the feature description information realizes accurate description of the attribute information represented by the transaction feature.

[0232] In another optional implementation, the following steps are further included before step 522 of this embodiment:

[0233] Perform image optimization processing on the transaction picture to obtain an optimized transaction picture;

[0234] Exemplarily, the transaction image is optimized and preprocessed before text recognition. The image optimization includes at least one of grayscale processing, sharpening processing, and contrast enhancement processing. Image optimization of the transaction image ensures the accuracy of text recognition and reduces the impact of differences in imaging quality of the transaction image on text recognition. In the optional implementation shown in this embodiment, text recognition is used to identify the optimized transaction image.

[0235] In summary, the method provided in this embodiment performs text recognition processing on the transaction image, determines the transaction feature representation in the transaction data obtained by text recognition, screens and determines the characteristic information about the transaction resources in the sales transaction in the transaction details, and expands the identification dimension of the transaction image to the transaction items and resource transfer in the transaction details; performs character editing processing on the transaction feature representation to obtain the feature to be identified, and realizes the editing and extraction of the character string that needs to be identified for tampering risk; performs string comparison operation on the reference identification feature and the feature to be identified to determine whether the transaction image has a tampering risk; can identify the tampering risk caused by the modification made by the background controller during the transaction information generation stage, that is, it realizes the accurate identification of the tampering risk of the transaction information when there is no modification trace on the transaction information, increases the scope of application of the effective identification of the tampering risk of the transaction information, and improves the accuracy of the identification of the transaction information.

[0236] Fig.13 A schematic diagram of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application is shown.

[0237] In this embodiment, the merchant terminal 110 reports transaction information to the data server 120, and the transaction information is used to indicate the transaction details of at least one sales transaction performed by the merchant using the merchant terminal 110. The data server 120 provides object storage (Cloud Object Storage, COS) for the transaction information. It can be understood that the data server 120 can store transaction information uploaded by multiple merchant terminals. This embodiment is described by taking only one merchant terminal as an example.

[0238] Exemplarily, the authentication server 220 obtains transaction information from the data server 120 at regular intervals, for example, the authentication server 220 obtains transaction information from the data server 120 at 12-hour intervals. In one example, the authentication server obtains transaction information from the data server 120 by running a business statistics service (Know Your Business, KYB).

[0239] The authentication server 220 is used to verify the identity information carried in the transaction information; illustratively, the identity information is used to indicate the operating qualifications of the merchant, and the identity information verification is used to determine whether the operating qualifications provided by the merchant meet the preset qualification requirements.

[0240] The authentication server 220 is used to authenticate transaction information to determine whether the transaction information has a risk of being tampered with.

[0241] Next, an example of how the authentication server 220 authenticates transaction information will be introduced.

[0242] Fig.14A flow chart of a method for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application is shown.

[0243] This embodiment is executed by an authentication server, which runs a business statistics service, an information authentication service, an image comparison service, a text recognition service, and a model processing service. The following steps will be further described:

[0244] Step 602: The business statistics service sends the transaction picture to the information identification service;

[0245] Exemplarily, the transaction images sent by the business statistics service are regularly obtained from the data server, and the data server provides object storage for the transaction images. Exemplarily, the transaction images are detailed information of online transaction orders uploaded by e-commerce merchants and provided by the e-commerce platform.

[0246] Step 604: the information identification service sends the transaction picture to the picture comparison service;

[0247] Exemplarily, the information authentication service is used to authenticate transaction images and determine whether there is a risk of tampering with the transaction images. In this embodiment, the information authentication service authenticates the transaction images by calling image comparison services, text recognition services, model processing services, etc.

[0248] Exemplarily, by sending the transaction picture to the picture comparison service, the picture comparison service is called to process the transaction picture.

[0249] Step 606: The picture comparison service compares the transaction picture and the sample picture;

[0250] Exemplarily, the image comparison service is used to compare the similarity between the transaction image and the sample image, and the sample label corresponding to the sample image is used to indicate the statistical method of information of at least one sales transaction in the sample image; for example, the transaction image shows at least one sales transaction by comparing the information of this year with the information of the previous year through a line chart. Exemplarily, when the similarity information between the transaction image and the sample image exceeds the trust threshold, the transaction image and the sample image have the same type label, and the trust threshold is the lower limit of the trust for judging that the transaction image and the sample image have the same type label.

[0251] Step 608: The image comparison service sends the type tag to the information identification service;

[0252] Exemplarily, when the similarity information between the transaction image and the sample image exceeds a trust threshold, the sample label corresponding to the sample image is determined as a type label, and the type label is fed back to the information identification service.

[0253] Step 610: The information identification service sends the transaction image to the text recognition service;

[0254] Exemplarily, by sending the transaction image to the text recognition service, the text recognition service is called to perform recognition processing on the transaction image.

[0255] Step 612: The text recognition service performs text recognition on the transaction image;

[0256] Exemplarily, the text recognition service performs text recognition on the transaction image to obtain the string information carried in the transaction image. The target transaction data corresponding to the preset position in the transaction image is determined as the transaction feature representation and feature description information. In one example, the transaction image received by the text recognition service also carries a type tag, and the preset position for the text recognition service to perform text recognition is indicated by the type tag; the type tag corresponds to one or more preset positions, which are used to indicate the position information of the transaction feature representation and feature description information. It should be noted that in order to ensure text recognition of transaction images with different image resolutions, the preset position can indicate the relative position of the transaction image, for example, one-third to two-thirds of the width of the transaction image; it can also indicate the pixel position of the sample image, and the resolution of the transaction image is converted to be consistent with the sample image by upsampling or downsampling the pixels of the transaction image, and then text recognition is performed based on the pixel position.

[0257] Step 614: the text recognition service sends the transaction feature representation and feature description information in the transaction image to the information authentication service;

[0258] The text recognition service returns the transaction feature representation and feature description information carried in the transaction image to the information identification service.

[0259] Step 616: Based on the type tag, the information identification service sends the transaction feature representation and feature description information to the model processing service;

[0260] Exemplarily, one or more candidate model processing services are running in the identification server, and the candidate model processing services correspond to the candidate type labels one by one. Based on the type label of the transaction image, the model processing service is determined from the one or more candidate model processing services; and the transaction feature representation and feature description information are sent to the model processing service.

[0261] Step 618: The model processing service searches for identification information corresponding to the type tag;

[0262] Exemplarily, the model processing service searches for corresponding identification information based on the feature description information; the identification information includes feature processing information and reference identification features, the feature processing information is used to indicate the character editing method for the transaction feature representation, and the reference identification feature is an information feature in a secure transaction without the risk of tampering. Exemplarily, the information format of the identification information can be a JavaScript code specification or a programming language in other formats.

[0263] Step 620: The model processing service identifies whether the transaction feature indicates a risk of tampering;

[0264] Exemplarily, the model processing service determines whether the transaction feature representation has a risk of tampering based on the identification information. Exemplarily, based on the feature processing information corresponding to the feature description information, the transaction feature representation is subjected to character editing processing to obtain the feature to be identified, and a string comparison operation is performed on the reference identification feature corresponding to the feature description information and the feature to be identified to obtain the identification result of the transaction information. Exemplarily, the string comparison operation performs a numerical comparison on the reference format code corresponding to the reference format feature obtained by encoding based on UTF8 or ASCII and the character format code corresponding to the character format feature.

[0265] Step 622: The model processing service sends the identification result to the information identification service;

[0266] The model processing service returns the identification result corresponding to the transaction image to the information identification service.

[0267] Step 624: The information identification service sends the identification result to the business statistics service;

[0268] The information authentication service sends the authentication result to the business statistics service, and instructs the business statistics service to write the authentication result into the data server.

[0269] To summarize, the method provided in this embodiment obtains characteristic information about transaction resources in the sales transaction in the transaction details by performing feature extraction on the transaction image, and expands the identification dimension of the transaction image to the transaction items and resource transfer conditions in the transaction details; performs character editing on the transaction feature representation to obtain the features to be identified, and realizes the editing and extraction of the character string that needs to be identified for tampering risk. Even if the means of tampering with the transaction image does not leave explicit traces of image smearing or image editing records, the scheme of the present application can also accurately identify the tampering risk of the transaction image from the implicit information at the semantic level such as the transaction items and resource transfer conditions; that is, the scheme of the present application does not rely on obvious modification traces to identify tampering risks, but identifies tampering risks through the implicit intrinsic features carried in the transaction image; it can identify tampering risks caused by modifications made by the background controller during the transaction information generation stage, thereby increasing the scope of application of effectively identifying tampering risks of transaction information.

[0270] Those skilled in the art can understand that the above embodiments can be implemented independently, or the above embodiments can be freely combined to form new embodiments to implement the transaction information tampering risk identification method of the present application.

[0271] Fig.15 The structural block diagram of a device for identifying the risk of transaction information tampering provided by an exemplary embodiment of the present application is shown. The device includes:

[0272] An acquisition module 810 is used to acquire the transaction information of the merchant, where the transaction information is used to indicate transaction details of at least one sales transaction performed by the merchant;

[0273] A processing module 820 is used to perform feature extraction processing on the transaction information to obtain a transaction feature representation and feature description information in the transaction information, wherein the transaction feature representation is used to indicate the characteristics of the transaction resources in the sales transaction performed by the merchant, and the feature description information is used to describe the attribute information of the transaction feature representation;

[0274] The acquisition module 810 is further used to acquire feature processing information and reference identification features corresponding to the feature description information, wherein the feature processing information is used to indicate a character editing method for the transaction feature representation, and the reference identification feature is an original information feature in a secure transaction;

[0275] The processing module 820 is further configured to perform character editing processing on the transaction feature representation based on the feature processing information to obtain a feature to be identified, wherein the feature to be identified is a character string that needs to be tampered risk identified based on the transaction feature representation;

[0276] The processing module 820 is further configured to compare the reference identification feature with the feature to be identified to obtain an identification result of the transaction information.

[0277] In an optional implementation of the application, the feature processing information includes extraction instruction information, the extraction instruction information indicates a character extraction method for representing the transaction feature; the reference identification feature includes a reference format feature, the reference format feature is used to indicate the arrangement format characteristics of the reference transaction value in the security transaction;

[0278] The processing module 820 is further configured to:

[0279] Based on the extraction indication information, character extraction processing is performed on the transaction feature representation to obtain a character format feature, wherein the transaction feature representation is used to describe a transaction-related value in the sales transaction, and the character format feature is used to indicate an arrangement format feature of the transaction-related value;

[0280] A character string comparison operation is performed on the reference format feature and the character format feature to obtain the identification result.

[0281] In an optional implementation of the application, the processing module 820 is further used for at least one of the following:

[0282] Based on first extraction indication information, performing character extraction processing on a decimal point symbol in the transaction feature representation, and obtaining a numerical precision of the transaction-related numerical value based on a position of the decimal point symbol, wherein the first extraction indication information indicates a symbol style of the decimal point symbol;

[0283] Based on the second extraction instruction information, character extraction processing is performed at the head position or the tail position of the transaction feature representation to obtain the currency symbol of the transaction-related value, and the second extraction instruction information indicates the extraction position of the currency symbol;

[0284] Based on the third extraction indication information, character extraction processing is performed on the transaction feature representation to obtain a separator symbol of the transaction-related numerical value, and the third extraction indication information indicates a preset interval for extracting the separator symbol.

[0285] In an optional implementation of the application, the feature processing information includes statistical indication information, the statistical indication information indicates a statistical calculation method and a numerical comparison method for at least two transaction feature representations; the reference identification feature includes a reference consistency feature, the reference consistency feature is used to indicate the consistency of at least two transaction numerical features of the secure transaction;

[0286] The processing module 820 is further configured to:

[0287] Based on the statistical indication information, statistical calculation processing is performed on the at least two transaction feature representations to obtain at least two transaction numerical features;

[0288] Based on the statistical indication information, performing a numerical comparison operation on the at least two transaction numerical features to obtain a data consistency feature;

[0289] A string comparison operation is performed on the reference consistency feature and the data consistency feature to obtain the identification result.

[0290] In an optional implementation of the application, the at least two transaction feature representations correspond to the same feature description information, and the at least two transaction feature representations both have numerical features;

[0291] The processing module 820 is further configured to:

[0292] Based on the statistical indication information, performing value extraction processing on at least two transaction feature representations one by one to obtain the at least two transaction value features, wherein the at least two transaction value features correspond to the at least two transaction feature representations in a one-to-one manner;

[0293] The reference consistency feature is used to indicate the data consistency of transaction feature representations having the same feature description information at different locations in the secure transaction.

[0294] In an optional implementation of the application, the statistical indication information is used to perform statistical calculation processing on at least three transaction feature representations, the at least three transaction feature representations corresponding to different feature description information, and the statistical calculation processing includes numerical calculation processing and numerical extraction processing;

[0295] The processing module 820 is further configured to:

[0296] Based on the operation information in the statistical indication information, the numerical calculation process is performed on at least two transaction feature representations to obtain a first transaction numerical feature, where the first transaction numerical feature is used to indicate a transaction numerical value implicitly carried by the at least two transaction feature representations in the transaction information;

[0297] Based on the extraction information in the statistical indication information, performing the value extraction process on a transaction feature representation to obtain a second transaction value feature, where the second transaction value feature is used to indicate a transaction value explicitly carried by the transaction feature representation in the transaction information;

[0298] The reference consistency feature is used to indicate the data consistency between the transaction value implicitly carried and the transaction value explicitly carried in the secure transaction.

[0299] In an optional implementation of the application, the at least two transaction feature representations correspond to the same feature description information, the first transaction feature representation of the at least two transaction feature representations does not have a numerical feature, and the at least two transaction feature representations do have a numerical feature;

[0300] The processing module 820 is further configured to:

[0301] Based on the numerical mapping relationship in the statistical indication information, searching for a mapping numerical range corresponding to the first transaction feature representation;

[0302] Based on the extraction information in the statistical indication information, performing numerical extraction processing on the second transaction feature representation to obtain a third transaction numerical feature;

[0303] Based on the statistical indication information, a numerical comparison operation is performed on the mapped value range and the third transaction value feature to obtain a data consistency feature, where the data consistency feature is used to indicate whether the third transaction value belongs to the mapped value range;

[0304] The reference consistency feature is used to indicate the numerical value attribution relationship between the reference numerical value range and the reference transaction value in the security transaction.

[0305] In an optional implementation of the application, the acquisition module 810 is further used to:

[0306] Acquire a type tag of the transaction information, where the type tag is used to indicate a statistical method for the information of the at least one sales transaction;

[0307] The feature processing information is determined based on the feature description information and the type label.

[0308] In an optional implementation of the application, the acquisition module 810 is further used to:

[0309] By comparing the transaction information and the sample information, obtaining similarity information for the sample information;

[0310] When the similarity information exceeds a trust threshold, the sample label of the sample information is determined as the type label of the transaction information.

[0311] In an optional implementation of the application, the acquisition module 810 is further used to:

[0312] Retrieving preset keywords in the transaction information;

[0313] In the case where the preset keyword exists in the transaction information, the sample label corresponding to the preset keyword is determined as the type label of the transaction information.

[0314] In an optional implementation of the application, the transaction information includes a transaction picture of at least one transaction detail, and the processing module 820 is further used to:

[0315] Performing text recognition on the transaction image to obtain transaction data carried by the transaction image;

[0316] The transaction feature representation and the feature description information are determined in the target transaction data corresponding to the preset position in the transaction picture.

[0317] In an optional implementation of the application, the transaction image includes table information, and the processing module 820 is further used to:

[0318] Determine the first transaction data corresponding to the first preset position in the transaction image as the feature description information, the first preset position is the header position of the table information, and the feature description information is used to indicate the attribute information represented by the transaction feature;

[0319] The second transaction data corresponding to the second preset position in the transaction image is determined as the transaction feature representation, and the second preset position and the first preset position have the same horizontal coordinate position or vertical coordinate position.

[0320] In an optional implementation of the application, the processing module 820 is further used to:

[0321] The transaction picture is subjected to image optimization processing to obtain an optimized transaction picture. The text recognition is used to recognize the optimized transaction picture. The image optimization processing includes at least one of grayscale processing, sharpening processing and contrast enhancement processing.

[0322] One point that needs to be explained is that the device provided in the above embodiment only uses the division of the above-mentioned functional modules as an example to implement its functions. In actual applications, the above-mentioned functions can be assigned to different functional modules according to actual needs, that is, the content structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0323] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method; the technical effects achieved by each module performing operations are the same as those in the embodiment of the method, and will not be elaborated here.

[0324] An embodiment of the present application also provides a computer device, which includes: a processor and a memory, wherein a computer program is stored in the memory; the processor is used to execute the computer program in the memory to implement the transaction information tampering risk identification method provided by the above-mentioned method embodiments.

[0325] Optionally, the computer device is a server. Fig.16 It is a structural block diagram of a server provided by an exemplary embodiment of the present application.

[0326] Typically, the server 2300 includes: a processor 2301 and a memory 2302 .

[0327] The processor 2301 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 2301 may be implemented in at least one hardware form of digital signal processing (DSP), field programmable gate array (FPGA), and programmable logic array (PLA). The processor 2301 may also include a main processor and a coprocessor. The main processor is a processor for processing data in an awake state, also known as a central processing unit (CPU); the coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 2301 may be integrated with a graphics processing unit (GPU), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 2301 may also include an artificial intelligence (AI) processor, which is used to process computing operations related to machine learning.

[0328] The memory 2302 may include one or more computer-readable storage media, which may be non-transitory. The memory 2302 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 2302 is used to store at least one instruction, which is used to be executed by the processor 2301 to implement the transaction information tampering risk identification method provided in the method embodiment of the present application.

[0329] In some embodiments, the server 2300 may also optionally include: an input interface 2303 and an output interface 2304. The processor 2301, the memory 2302, the input interface 2303, and the output interface 2304 may be connected via a bus or a signal line. Each peripheral device may be connected to the input interface 2303 and the output interface 2304 via a bus, a signal line, or a circuit board. The input interface 2303 and the output interface 2304 may be used to connect at least one peripheral device related to input / output (I / O) to the processor 2301 and the memory 2302. In some embodiments, the processor 2301, the memory 2302, the input interface 2303, and the output interface 2304 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 2301, the memory 2302, the input interface 2303, and the output interface 2304 may be implemented on a separate chip or circuit board, which is not limited in the embodiments of the present application.

[0330] Those skilled in the art will appreciate that the structure shown above does not constitute a limitation on the server 2300 , and may include more or fewer components than shown in the figure, or combine certain components, or adopt a different component arrangement.

[0331] In an exemplary embodiment, a chip is also provided, which includes a programmable logic circuit and / or program instructions. When the chip is run on a computer device, it is used to implement the method for identifying the risk of transaction information tampering as described in the above aspects.

[0332] In an exemplary embodiment, a computer program product is also provided, the computer program product includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor reads and executes the computer instructions from the computer-readable storage medium to implement the transaction information tampering risk identification method provided by the above-mentioned method embodiments.

[0333] In an exemplary embodiment, a computer-readable storage medium is also provided, in which a computer program is stored. The computer program is loaded and executed by a processor to implement the transaction information tampering risk identification method provided by the above-mentioned method embodiments.

[0334] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0335] Those skilled in the art should be aware that in one or more of the above examples, the functions described in the embodiments of the present application can be implemented with hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein the communication media include any media that facilitates the transmission of a computer program from one place to another. The storage medium can be any available medium that a general or special-purpose computer can access.

[0336] The above description is only an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for identifying the risk of transaction information tampering, characterized in that: The method is performed by a computer device, and the method comprises: Acquire the transaction information of the merchant, where the transaction information is used to indicate transaction details of at least one sales transaction performed by the merchant; Performing feature extraction processing on the transaction information to obtain a transaction feature representation and feature description information in the transaction information, wherein the transaction feature representation is used to indicate the characteristics of the transaction resources in the sales transaction performed by the merchant, and the feature description information is used to describe the attribute information of the transaction feature representation; Acquire feature processing information and reference identification features corresponding to the feature description information, wherein the feature processing information is used to indicate a character editing method for the transaction feature representation, and the reference identification feature is an original information feature in a secure transaction; Based on the feature processing information, character editing processing is performed on the transaction feature representation to obtain a feature to be identified, where the feature to be identified is a character string that needs to be tampered risk identified and is obtained based on the transaction feature representation; The reference identification feature and the feature to be identified are compared to obtain an identification result of the transaction information.

2. The method according to claim 1, characterized in that The feature processing information includes extraction instruction information, and the extraction instruction information indicates a character extraction method for representing the transaction feature; the reference identification feature includes a reference format feature, and the reference format feature is used to indicate the arrangement format characteristics of the reference transaction value in the security transaction; The step of performing character editing processing on the transaction feature representation based on the feature processing information to obtain the feature to be identified includes: Based on the extraction indication information, character extraction processing is performed on the transaction feature representation to obtain a character format feature, wherein the transaction feature representation is used to describe a transaction-related value in the sales transaction, and the character format feature is used to indicate an arrangement format feature of the transaction-related value; The comparing the reference identification feature with the feature to be identified to obtain the identification result of the transaction information includes: A character string comparison operation is performed on the reference format feature and the character format feature to obtain the identification result.

3. The method according to claim 2, characterized in that The step of performing character extraction processing on the transaction feature representation based on the extraction instruction information to obtain character format features comprises at least one of the following steps: Based on first extraction indication information, performing character extraction processing on a decimal point symbol in the transaction feature representation, and obtaining a numerical precision of the transaction-related numerical value based on a position of the decimal point symbol, wherein the first extraction indication information indicates a symbol style of the decimal point symbol; Based on the second extraction instruction information, character extraction processing is performed at the head position or the tail position of the transaction feature representation to obtain the currency symbol of the transaction-related value, and the second extraction instruction information indicates the extraction position of the currency symbol; Based on the third extraction indication information, character extraction processing is performed on the transaction feature representation to obtain a separator symbol of the transaction-related numerical value, and the third extraction indication information indicates a preset interval for extracting the separator symbol.

4. The method according to claim 1, characterized in that: The feature processing information includes statistical indication information, wherein the statistical indication information indicates a statistical calculation method and a numerical comparison method for at least two transaction feature representations; the reference identification feature includes a reference consistency feature, wherein the reference consistency feature is used to indicate the consistency of at least two transaction numerical features of the secure transaction; The step of performing character editing processing on the transaction feature representation based on the feature processing information to obtain the feature to be identified includes: Based on the statistical indication information, statistical calculation processing is performed on the at least two transaction feature representations to obtain at least two transaction numerical features; Based on the statistical indication information, performing a numerical comparison operation on the at least two transaction numerical features to obtain a data consistency feature; The comparing the reference identification feature with the feature to be identified to obtain the identification result of the transaction information includes: A string comparison operation is performed on the reference consistency feature and the data consistency feature to obtain the identification result.

5. The method according to claim 4, characterized in that The at least two transaction feature representations correspond to the same feature description information, and the at least two transaction feature representations both have numerical features; The step of performing statistical calculation processing on the at least two transaction feature representations based on the statistical indication information to obtain at least two transaction numerical features includes: Based on the statistical indication information, performing value extraction processing on at least two transaction feature representations one by one to obtain the at least two transaction value features, wherein the at least two transaction value features correspond to the at least two transaction feature representations in a one-to-one manner; The reference consistency feature is used to indicate the data consistency of transaction feature representations having the same feature description information at different locations in the secure transaction.

6. The method according to claim 4, characterized in that The statistical indication information is used to perform statistical calculation processing on at least three transaction feature representations, the at least three transaction feature representations corresponding to different feature description information, and the statistical calculation processing includes numerical calculation processing and numerical extraction processing; The step of performing statistical calculation processing on the at least two transaction feature representations based on the statistical indication information to obtain at least two transaction numerical features includes: Based on the operation information in the statistical indication information, the numerical calculation process is performed on at least two transaction feature representations to obtain a first transaction numerical feature, where the first transaction numerical feature is used to indicate a transaction numerical value implicitly carried by the at least two transaction feature representations in the transaction information; Based on the extraction information in the statistical indication information, performing the value extraction process on a transaction feature representation to obtain a second transaction value feature, where the second transaction value feature is used to indicate a transaction value explicitly carried by the transaction feature representation in the transaction information; The reference consistency feature is used to indicate the data consistency between the transaction value implicitly carried and the transaction value explicitly carried in the secure transaction.

7. The method according to claim 4, characterized in that The at least two transaction feature representations correspond to the same feature description information, the first transaction feature representation of the at least two transaction feature representations does not have a numerical feature, and the at least two transaction feature representations have a numerical feature; The step of performing statistical calculation processing on the at least two transaction feature representations based on the statistical indication information to obtain at least two transaction numerical features includes: Based on the numerical mapping relationship in the statistical indication information, searching for a mapping numerical range corresponding to the first transaction feature representation; Based on the extraction information in the statistical indication information, performing numerical extraction processing on the second transaction feature representation to obtain a third transaction numerical feature; The performing a numerical comparison operation on the at least two transaction numerical features based on the statistical indication information to obtain a data consistency feature includes: Based on the statistical indication information, a numerical comparison operation is performed on the mapped value range and the third transaction value feature to obtain a data consistency feature, where the data consistency feature is used to indicate whether the third transaction value belongs to the mapped value range; The reference consistency feature is used to indicate the numerical value attribution relationship between the reference numerical value range and the reference transaction value in the security transaction.

8. The method according to any one of claims 1 to 7, characterized in that: The method further comprises: Acquire a type tag of the transaction information, where the type tag is used to indicate a statistical method for the information of the at least one sales transaction; The feature processing information is determined based on the feature description information and the type label.

9. The method according to claim 8, characterized in that The step of obtaining the type tag of the transaction information includes: By comparing the transaction information and the sample information, obtaining similarity information for the sample information; When the similarity information exceeds a trust threshold, the sample label of the sample information is determined as the type label of the transaction information.

10. The method according to claim 8, characterized in that The step of obtaining the type tag of the transaction information includes: Retrieving preset keywords in the transaction information; In the case where the preset keyword exists in the transaction information, the sample label corresponding to the preset keyword is determined as the type label of the transaction information.

11. The method according to any one of claims 1 to 7, characterized in that: The transaction information includes at least one transaction picture of transaction details, and the feature extraction process is performed on the transaction information to obtain the transaction feature representation and feature description information in the transaction information, including: Performing text recognition on the transaction image to obtain transaction data carried by the transaction image; The transaction feature representation and the feature description information are determined in the target transaction data corresponding to the preset position in the transaction picture.

12. The method according to claim 11, characterized in that The transaction picture includes table information, and the determining of the transaction feature representation and the feature description information in the target transaction data corresponding to the preset position in the transaction picture includes: Determine the first transaction data corresponding to the first preset position in the transaction image as the feature description information, the first preset position is the header position of the table information, and the feature description information is used to indicate the attribute information represented by the transaction feature; The second transaction data corresponding to the second preset position in the transaction image is determined as the transaction feature representation, and the second preset position and the first preset position have the same horizontal coordinate position or vertical coordinate position.

13. The method according to claim 11, characterized in that The method further comprises: The transaction picture is subjected to image optimization processing to obtain an optimized transaction picture. The text recognition is used to recognize the optimized transaction picture. The image optimization processing includes at least one of grayscale processing, sharpening processing and contrast enhancement processing.

14. A device for identifying the risk of transaction information tampering, characterized in that: The device comprises: An acquisition module, used to acquire the transaction information of the merchant, wherein the transaction information is used to indicate transaction details of at least one sales transaction performed by the merchant; a processing module, configured to perform feature extraction processing on the transaction information to obtain a transaction feature representation and feature description information in the transaction information, wherein the transaction feature representation is used to indicate the characteristics of the transaction resources in the sales transaction performed by the merchant, and the feature description information is used to describe the attribute information of the transaction feature representation; The acquisition module is further used to acquire feature processing information and reference identification features corresponding to the feature description information, wherein the feature processing information is used to indicate a character editing method for the transaction feature representation, and the reference identification feature is an original information feature in a secure transaction; The processing module is further used to perform character editing processing on the transaction feature representation based on the feature processing information to obtain a feature to be identified, wherein the feature to be identified is a character string that needs to be tampered risk identified based on the transaction feature representation; The processing module is further used to compare the reference identification feature with the feature to be identified to obtain an identification result of the transaction information.

15. A computer device, characterized in that: The computer device includes: a processor and a memory, wherein the memory stores at least one program; the processor is used to execute the at least one program in the memory to implement the method for identifying the risk of transaction information tampering as described in any one of claims 1 to 13.

16. A computer-readable storage medium, characterized in that: The readable storage medium stores executable instructions, and the executable instructions are loaded and executed by a processor to implement the method for identifying the risk of transaction information tampering as described in any one of claims 1 to 13.

17. A computer program product, characterized in that The computer program product includes computer instructions, which are stored in a computer-readable storage medium. A processor reads and executes the computer instructions from the computer-readable storage medium to implement the method for identifying the risk of transaction information tampering as described in any one of claims 1 to 13.

Images