Transaction identification method, device, electronic device and storage medium

Abstract

The transaction identification method, device, electronic device and storage medium provided by the present invention can analyze the degree of attribute difference between the attribute information of the current transaction and the historical transaction each time the target customer initiates a transaction, thereby determining the difference between the target customer and his / her transaction habits, and then accurately identifying the fraudulent nature of the current transaction by considering the target customer's profile, thereby effectively identifying that each transaction is normally initiated by the person himself / herself. Based on the present invention, fraudulent transactions can be accurately identified, the occurrence of fraudulent transactions can be reduced or even avoided, and the property safety of customers can be guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of software technology, and more specifically, to a transaction identification method, device, electronic device and storage medium. Background Art

[0002] In life, mobile phones are closely related to our lives, but bank card fraud caused by lost mobile phones is a common occurrence, which causes great harm to both the bank's customers and the bank itself.

[0003] Therefore, how to safely and securely prevent transaction fraud has become an urgent problem that needs to be solved. Summary of the invention

[0004] In view of this, in order to solve the above problems, the present invention provides a transaction identification method, device, electronic device and storage medium, and the technical solution is as follows:

[0005] In one aspect, the present invention provides a transaction identification method, the method comprising:

[0006] In response to a target transaction currently initiated by a target customer, obtaining the profile information of the target customer and the current first attribute information of the target transaction;

[0007] Retrieving the second attribute information of the target customer when initiating the target transaction in history, and analyzing the degree of attribute difference between the first attribute information and the second attribute information;

[0008] According to the degree of attribute difference and the portrait information of the target customer, the fraudulent nature of the target transaction currently initiated by the target customer is identified.

[0009] Optionally, the analyzing the degree of attribute difference between the first attribute information and the second attribute information includes:

[0010] Taking the target transaction as an entity and the first attribute information as an edge, constructing a first transaction knowledge graph of the target customer;

[0011] Taking the target transaction as an entity and the second attribute information as an edge, constructing a second transaction knowledge graph of the target customer;

[0012] The first transaction knowledge graph and the second transaction knowledge graph are compared to calculate the degree of attribute difference.

[0013] Optionally, the target customer has initiated multiple target transactions in history, and the second transaction knowledge graph is multiple;

[0014] The comparing the first transaction knowledge graph and the second transaction knowledge graph to calculate the degree of attribute difference includes:

[0015] Aggregate the plurality of the second transaction knowledge graphs to obtain an average transaction knowledge graph of the target customer, where the entity of the average transaction knowledge graph is the target transaction and the edge is an aggregation result of the second attribute information;

[0016] The first transaction knowledge graph and the average transaction knowledge graph are compared to calculate the degree of difference of the first attribute.

[0017] Optionally, the comparing the first transaction knowledge graph and the second transaction knowledge graph to calculate the degree of attribute difference further includes:

[0018] Filtering a target transaction knowledge graph that meets a preset condition from the plurality of second transaction knowledge graphs;

[0019] The first transaction knowledge graph and the target transaction knowledge graph are compared to calculate the degree of difference of the second attribute.

[0020] Optionally, the identifying the fraudulent nature of the target transaction currently initiated by the target customer according to the attribute difference degree and the profile information of the target customer includes:

[0021] Retrieving a fraud prediction model, wherein the fraud prediction model is obtained by training a basic model in advance using the historical attribute difference degree when the target transaction is initiated by the designated customer and the profile information of the designated customer as samples, with the goal of making the fraud prediction result of the sample close to the fraud labeling result;

[0022] Inputting the attribute difference degree and the target customer's profile information into the fraud prediction model to obtain the fraud probability output by the fraud prediction model;

[0023] Determining whether the fraud probability is greater than a preset probability threshold;

[0024] If so, fraud is determined to have occurred;

[0025] If not, it is determined that no fraud has occurred.

[0026] Optionally, the method further includes:

[0027] In the event of fraud, transaction confirmation is performed to the target customer.

[0028] Another aspect of the present invention provides a transaction identification device, the device comprising:

[0029] An information acquisition module, used to respond to a target transaction currently initiated by a target customer, and acquire the portrait information of the target customer and the current first attribute information of the target transaction;

[0030] an attribute analysis module, used to retrieve the second attribute information of the target customer when initiating the target transaction in history, and analyze the degree of attribute difference between the first attribute information and the second attribute information;

[0031] The fraud identification module is used to identify the fraudulent nature of the target transaction currently initiated by the target customer based on the degree of attribute difference and the portrait information of the target customer.

[0032] Optionally, the attribute analysis module is specifically used to:

[0033] A first transaction knowledge graph of the target customer is constructed with the target transaction as an entity and the first attribute information as an edge; a second transaction knowledge graph of the target customer is constructed with the target transaction as an entity and the second attribute information as an edge; and the first transaction knowledge graph and the second transaction knowledge graph are compared to calculate the degree of attribute difference.

[0034] Another aspect of the present invention provides an electronic device, comprising: at least one memory and at least one processor; the memory stores a program, the processor calls the program stored in the memory, and the program is used to implement any one of the transaction identification methods described.

[0035] Another aspect of the present invention provides a storage medium, wherein the storage medium stores computer executable instructions, and the computer executable instructions are used to execute any one of the transaction identification methods described above.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The transaction identification method, device, electronic device and storage medium provided by the present invention can analyze the degree of attribute difference between the attribute information of the current transaction and the historical transaction each time the target customer initiates a transaction, thereby determining the difference between the target customer and his / her transaction habits, and then accurately identifying the fraudulent nature of the current transaction by considering the target customer's profile, thereby effectively identifying that each transaction is normally initiated by the person himself / herself. Based on the present invention, fraudulent transactions can be accurately identified, the occurrence of fraudulent transactions can be reduced or even avoided, and the property safety of customers can be guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0039] Figure 1A method flow chart of a transaction identification method provided by an embodiment of the present invention;

[0040] Figure 2 A partial method flow chart of a transaction identification method provided by an embodiment of the present invention;

[0041] Figure 3 A flowchart of another part of the transaction identification method provided by an embodiment of the present invention;

[0042] Figure 4 A flowchart of another part of the transaction identification method provided by an embodiment of the present invention;

[0043] Figure 5 A schematic diagram of the structure of a transaction identification device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] The embodiment of the present invention provides a transaction identification method, the method flow chart of which is as follows: Figure 1 As shown, the following steps are included:

[0047] S10, responding to the target transaction currently initiated by the target customer, obtaining the portrait information of the target customer and the current first attribute information of the target transaction.

[0048] In an embodiment of the present invention, for a transaction, multiple related attributes will be assigned when it is initiated, such as transaction amount, transaction date, transaction time, transaction week number (the day of the week), transaction type (other bank, same bank), transaction time limit (real-time, ordinary), transaction location, and transaction channel (mobile banking, ATM, etc.).

[0049] Therefore, each time a customer initiates a transaction, multiple attributes related to the transaction have corresponding values, and the values ​​of these attributes can constitute the attribute information of the transaction.

[0050] The portrait information can describe some of the customer’s own attributes, such as gender, age, marital status, place of residence, years of work, annual income, etc.

[0051] S20, retrieve the second attribute information of the target customer when initiating the target transaction in history, and analyze the degree of attribute difference between the first attribute information and the second attribute information.

[0052] In the embodiment of the present invention, the attribute information of each transaction initiated by the customer is counted, and the attribute information of the same transaction initiated by the customer multiple times is stored together. Therefore, when the target customer currently initiates a target transaction, the attribute information of the target transaction initiated by the target customer in the past, that is, the second attribute information, can be obtained. When a target transaction is initiated once, a set of second attribute information is obtained.

[0053] Since the first attribute information is the attribute information of the target transaction currently initiated by the target customer, the first attribute information and the second attribute information have the same attribute. By aligning the attributes, the numerical difference between the first attribute information and the second attribute information under different attributes can be obtained. Under one attribute, the greater the numerical difference, the greater the attribute difference. In addition, the more attributes with different numerical values, the greater the attribute difference.

[0054] Of course, a group of second attribute information has a corresponding attribute difference degree with the first attribute information, and multiple groups of second attribute information correspond to multiple attribute difference degrees with the first attribute information. Therefore, when there are multiple groups of second attribute information, multiple attribute difference degrees of multiple groups of second attribute information can be processed to obtain a final attribute difference degree for identifying fraud, such as randomly selecting a group of attribute difference degrees of second attribute information, or selecting the attribute difference degree of a second attribute information whose transaction time is closest to the current one, which is not limited in the embodiment of the present invention.

[0055] In the specific implementation process, the "analyzing the attribute difference between the first attribute information and the second attribute information" in step S20 can be performed using the following steps. The method flow chart is as follows: Figure 2 As shown:

[0056] S201, constructing a first transaction knowledge graph of a target customer with the target transaction as an entity and the first attribute information as an edge.

[0057] S202, constructing a second transaction knowledge graph of the target customer with the target transaction as the entity and the second attribute information as the edge.

[0058] S203, comparing the first transaction knowledge graph and the second transaction knowledge graph to calculate the degree of attribute difference.

[0059] In the embodiment of the present invention, the transaction knowledge graph is constructed for the first attribute information and the second attribute information respectively by using the knowledge graph technology. Of course, since the transaction time corresponding to the first attribute information and the second attribute information is different, the embodiment of the present invention constructs the transaction knowledge graph of the target customer according to the time axis.

[0060] Therefore, the basic architecture of the first transaction knowledge graph and the second transaction knowledge graph is the same, that is, the target transaction is the entity and the attributes related to the target transaction are the edges. The difference is that the values ​​of the attributes corresponding to the edges will be different.

[0061] When the target customer has initiated target transactions multiple times in the history, the second attribute information is in multiple groups. The grouped second attribute information can also be processed to determine a second attribute information for calculating the degree of attribute difference, such as randomly selecting a second attribute information, or selecting a second attribute information with a transaction time closest to the current one. The embodiment of the present invention is not limited to this.

[0062] After determining the second attribute information used to calculate the degree of attribute difference, based on the knowledge graph technology, a corresponding knowledge graph, i.e., a second transaction knowledge graph, can be constructed. For the second transaction knowledge graph and the first transaction knowledge graph, by comparing the edges with the same attributes, the degree of difference between different edges can be accurately obtained, and the degree of attribute difference can be obtained by summarizing the degree of difference of all edges or screening the degree of difference of some edges.

[0063] In some other embodiments, when the target customer has initiated target transactions multiple times in the history, there are multiple groups of second attribute information, and a corresponding transaction knowledge graph is constructed for each group of second attribute information, that is, multiple second transaction knowledge graphs are obtained.

[0064] At this time, in the process of executing step S203 "comparing the first transaction knowledge graph and the second transaction knowledge graph to calculate the degree of attribute difference", the following steps can be adopted. The method flow chart is as follows: Figure 3 As shown:

[0065] S2031, aggregate multiple second transaction knowledge graphs to obtain an average transaction knowledge graph of the target customer, where the entity of the average transaction knowledge graph is the target transaction and the edge is the aggregation result of the second attribute information.

[0066] In an embodiment of the present invention, based on the edges, i.e., attributes, of multiple second transaction knowledge graphs, the values ​​under different attributes can be aggregated. The aggregation processing includes but is not limited to operations such as taking the mean or taking the mode, and multiple values ​​under the same attribute are aggregated into one, thereby aggregating the edges with the same attributes of multiple second transaction knowledge graphs into one edge, thereby obtaining an average transaction knowledge graph with the target transaction as the entity.

[0067] Taking transaction amount as an example, when aggregating the values ​​under this attribute, since it is a continuous variable, the mean can be taken as the aggregation result corresponding to this attribute. Taking transaction type as an example, when aggregating the values ​​under this attribute, since it is a discrete variable, the mode, that is, the value with the most occurrences, can be taken as the aggregation result corresponding to this attribute.

[0068] S2032, comparing the first transaction knowledge graph and the average transaction knowledge graph to calculate the degree of difference of the first attribute.

[0069] In the embodiment of the present invention, for the first transaction knowledge graph and the average transaction knowledge graph, the difference degree of different edges can be vectorized and calculated, so as to summarize the difference degree of all edges or select the difference degree of some edges as the attribute difference degree of the first aspect, that is, the first attribute difference degree. This can reduce the amount of calculation of the graph difference.

[0070] On this basis, in order to improve the comprehensiveness of the calculation of the attribute difference degree, the embodiment of the present invention may further include the following steps:

[0071] A target transaction knowledge graph that meets preset conditions is selected from multiple second transaction knowledge graphs; and the first transaction knowledge graph and the target transaction knowledge graph are compared to calculate the degree of difference of the second attribute.

[0072] In an embodiment of the present invention, a second transaction knowledge graph or multiple second transaction graphs with the closest transaction time to the current one can be selected as the target transaction knowledge graph, and a second transaction knowledge graph whose value under a specified attribute meets the corresponding threshold can also be selected as the target transaction knowledge graph. For example, the second transaction knowledge graph with the largest transaction amount can be used as the target transaction knowledge graph. This embodiment of the present invention is not limited to this.

[0073] For the first transaction knowledge graph and each target transaction knowledge graph, the degree of difference of different edges can be obtained through vectorized calculation, so as to summarize the degree of difference of all edges or filter the degree of difference of some edges as the attribute difference degree of the second aspect, that is, the second attribute difference degree.

[0074] Thus, a set of the first attribute degree and the second attribute degree is taken as the attribute difference degree between the first attribute information and the second attribute information.

[0075] S30, identifying the fraudulent nature of the target transaction currently initiated by the target customer based on the degree of attribute difference and the profile information of the target customer.

[0076] In the embodiment of the present invention, the difference in transaction attribute direction and the customer's profile are comprehensively considered to identify whether the initiated target transaction is fraudulent.

[0077] In the specific implementation process, step S30 "identifying the fraudulent nature of the target transaction currently initiated by the target customer based on the degree of attribute difference and the target customer's profile information" can be implemented by the following steps. The method flow chart is as follows: Figure 4 As shown:

[0078] S301, retrieve the fraud prediction model. The fraud prediction model is obtained by training the basic model in advance using the historical attribute difference degree when the target transaction is initiated by the designated customer and the portrait information of the designated customer as samples, with the goal of making the fraud prediction results of the samples close to the fraud labeling results.

[0079] In an embodiment of the present invention, the basic model of the fraud prediction model can be a fully connected multi-layer deep neural network, which takes the degree of attribute difference and portrait information as characteristics. By annotating samples, the basic model is continuously iteratively trained to finally obtain a fraud prediction model.

[0080] S302, inputting the attribute difference degree and the target customer's profile information into the fraud prediction model to obtain the fraud probability output by the fraud prediction model.

[0081] In an embodiment of the present invention, a fraud prediction model is based on the degree of attribute difference of the target transaction currently initiated by the target customer and the portrait information of the target customer as features input into the fraud prediction model, and the fraud prediction model outputs a predicted fraud probability. The higher the fraud probability, the higher the possibility of fraud in this transaction.

[0082] S303, determine whether the fraud probability is greater than a preset probability threshold; if so, execute step S304; if not, execute step S305.

[0083] S304, determine that fraud exists.

[0084] S305, determine that there is no fraud.

[0085] On this basis, when fraud is determined, the embodiment of the present invention can also use telephone, email, or text message to confirm the transaction with the target customer, and then execute the target transaction after obtaining the confirmation information fed back by the target customer.

[0086] The transaction identification method provided by the embodiment of the present invention can analyze the degree of attribute difference between the attribute information of the current transaction and the historical transaction each time the target customer initiates a transaction, thereby determining the difference between the target customer and his / her transaction habits, and then accurately identifying the fraudulent nature of the current transaction by considering the target customer's profile, thereby effectively identifying that each transaction is normally initiated by the person himself / herself. Based on the present invention, fraudulent transactions can be accurately identified, the occurrence of fraudulent transactions can be reduced or even avoided, and the property safety of customers can be guaranteed.

[0087] Based on the transaction identification method provided in the above embodiment, the embodiment of the present invention also provides a device for executing the above transaction identification method. The structural diagram of the device is shown in FIG. Figure 5 As shown, including:

[0088] The information acquisition module 10 is used to respond to the target transaction currently initiated by the target customer and acquire the portrait information of the target customer and the current first attribute information of the target transaction;

[0089] The attribute analysis module 20 is used to retrieve the second attribute information of the target customer when initiating the target transaction in history, and analyze the attribute difference between the first attribute information and the second attribute information;

[0090] The fraud identification module 30 is used to identify the fraudulent nature of the target transaction currently initiated by the target customer based on the degree of attribute difference and the profile information of the target customer.

[0091] Optionally, the attribute analysis module 20 performs a process of analyzing the attribute difference between the first attribute information and the second attribute information, including:

[0092] Taking the target transaction as the entity and the first attribute information as the edge, construct the first transaction knowledge graph of the target customer; taking the target transaction as the entity and the second attribute information as the edge, construct the second transaction knowledge graph of the target customer; compare the first transaction knowledge graph and the second transaction knowledge graph to calculate the degree of attribute difference.

[0093] Optionally, the target customer has initiated multiple target transactions in the past, and the second transaction knowledge graph is multiple;

[0094] The attribute analysis module 20 performs a process of comparing the first transaction knowledge graph and the second transaction knowledge graph to calculate the degree of attribute difference, including:

[0095] Aggregate multiple second transaction knowledge graphs to obtain an average transaction knowledge graph of the target customer, where the entities of the average transaction knowledge graph are the target transactions and the edges are the aggregation results of the second attribute information; compare the first transaction knowledge graph and the average transaction knowledge graph to calculate the degree of difference in the first attribute.

[0096] Optionally, the attribute analysis module 20 performs a process of comparing the first transaction knowledge graph and the second transaction knowledge graph to calculate the degree of attribute difference, further comprising:

[0097] A target transaction knowledge graph that meets preset conditions is selected from multiple second transaction knowledge graphs; and the first transaction knowledge graph and the target transaction knowledge graph are compared to calculate the degree of difference of the second attribute.

[0098] Optionally, the fraud identification module 30 performs a process of identifying the fraudulent nature of the target transaction currently initiated by the target customer based on the degree of attribute difference and the profile information of the target customer, including:

[0099] Retrieve a fraud prediction model, which is obtained by training a basic model in advance using the historical attribute difference degree when a target transaction is initiated by a designated customer and the portrait information of the designated customer as samples, with the goal of making the fraud prediction results of the samples close to the fraud labeling results; input the attribute difference degree and the portrait information of the target customer into the fraud prediction model to obtain the fraud probability output by the fraud prediction model; determine whether the fraud probability is greater than a preset probability threshold; if so, determine that fraud exists; if not, determine that fraud does not exist.

[0100] Optionally, the fraud identification module 30 is further used for:

[0101] In case of fraud, transaction confirmation is provided to the target customer.

[0102] The transaction identification device provided by the embodiment of the present invention can analyze the degree of attribute difference between the attribute information of the current transaction and the historical transaction each time the target customer initiates a transaction, thereby determining the difference between the target customer and his / her transaction habits, and then accurately identifying the fraudulent nature of the current transaction by considering the target customer's profile, thereby effectively identifying that each transaction is normally initiated by the person himself / herself. Based on the present invention, fraudulent transactions can be accurately identified, the occurrence of fraudulent transactions can be reduced or even avoided, and the property safety of customers can be guaranteed.

[0103] Based on the transaction identification method provided in the above embodiments, an embodiment of the present invention further provides an electronic device, including: at least one memory and at least one processor; the memory stores a program, the processor calls the program stored in the memory, and the program is used to implement the transaction identification method of any embodiment.

[0104] Based on the transaction identification method provided in the above embodiments, an embodiment of the present invention further provides a storage medium, in which computer executable instructions are stored. The computer executable instructions are used to execute the transaction identification method of any embodiment.

[0105] The transaction identification method, device, electronic device and storage medium provided by the present invention are introduced in detail above. The principles and implementation methods of the present invention are explained in this article using specific examples. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0106] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0107] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device that includes a series of elements is inherent to the elements, or also includes elements inherent to these processes, methods, articles or devices. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the presence of other identical elements in the process, method, article or device that includes the elements.

[0108] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A transaction identification method, characterized in that: The method comprises: In response to a target transaction currently initiated by a target customer, obtaining the profile information of the target customer and the current first attribute information of the target transaction; Retrieving the second attribute information of the target customer when initiating the target transaction in history, and analyzing the degree of attribute difference between the first attribute information and the second attribute information; According to the attribute difference degree and the profile information of the target customer, identifying the fraudulent nature of the target transaction currently initiated by the target customer; The analyzing the degree of attribute difference between the first attribute information and the second attribute information includes: Taking the target transaction as an entity and the first attribute information as an edge, constructing a first transaction knowledge graph of the target customer; Taking the target transaction as an entity and the second attribute information as an edge, constructing a second transaction knowledge graph of the target customer; The first transaction knowledge graph and the second transaction knowledge graph are compared to calculate the degree of attribute difference.

2. The method according to claim 1, characterized in that The target customer has initiated multiple target transactions in the past, and the second transaction knowledge graphs are multiple; The comparing the first transaction knowledge graph and the second transaction knowledge graph to calculate the degree of attribute difference includes: Aggregate the plurality of the second transaction knowledge graphs to obtain an average transaction knowledge graph of the target customer, where the entity of the average transaction knowledge graph is the target transaction and the edge is an aggregation result of the second attribute information; The first transaction knowledge graph and the average transaction knowledge graph are compared to calculate the degree of difference of the first attribute.

3. The method according to claim 2, characterized in that The step of comparing the first transaction knowledge graph and the second transaction knowledge graph to calculate the degree of attribute difference also includes: Filtering a target transaction knowledge graph that meets a preset condition from the plurality of second transaction knowledge graphs; The first transaction knowledge graph and the target transaction knowledge graph are compared to calculate the degree of difference of the second attribute.

4. The method according to claim 1, characterized in that: The identifying the fraudulent nature of the target transaction currently initiated by the target customer according to the attribute difference degree and the profile information of the target customer includes: Retrieving a fraud prediction model, wherein the fraud prediction model is obtained by training a basic model in advance using the historical attribute difference degree when the target transaction is initiated by the designated customer and the profile information of the designated customer as samples, with the goal of making the fraud prediction result of the sample close to the fraud labeling result; Inputting the attribute difference degree and the target customer's profile information into the fraud prediction model to obtain the fraud probability output by the fraud prediction model; Determining whether the fraud probability is greater than a preset probability threshold; If so, fraud is determined to have occurred; If not, it is determined that no fraud has occurred.

5. The method according to claim 4, characterized in that The method further comprises: In the event of fraud, transaction confirmation is performed to the target customer.

6. A transaction identification device, characterized in that: The device comprises: An information acquisition module, used to respond to a target transaction currently initiated by a target customer, and acquire the portrait information of the target customer and the current first attribute information of the target transaction; an attribute analysis module, used to retrieve the second attribute information of the target customer when initiating the target transaction in history, and analyze the degree of attribute difference between the first attribute information and the second attribute information; A fraud identification module, used for identifying the fraudulent nature of the target transaction currently initiated by the target customer based on the degree of attribute difference and the profile information of the target customer; The attribute analysis module is specifically used for: A first transaction knowledge graph of the target customer is constructed with the target transaction as an entity and the first attribute information as an edge; a second transaction knowledge graph of the target customer is constructed with the target transaction as an entity and the second attribute information as an edge; and the first transaction knowledge graph and the second transaction knowledge graph are compared to calculate the degree of attribute difference.

7. An electronic device, characterized in that: include: at least one memory and at least one processor; The memory stores a program, and the processor calls the program stored in the memory, and the program is used to implement the transaction identification method according to any one of claims 1 to 5.

8. A storage medium, characterized in that: The storage medium stores computer executable instructions, and the computer executable instructions are used to execute the transaction identification method according to any one of claims 1 to 5.

Images