Method and apparatus for supporting terminal transaction based on risk prediction

Abstract

The application discloses a kind of based on risk prediction support terminal transaction method and device, it is related to financial field, this method includes: obtaining the historical transaction data of bank user, according to the historical transaction data, bank user is classified, obtains multiple user sub-set;For each user sub-set, determine the risk probability of each transaction scene corresponding to the user sub-set and the probability lower bound value corresponding to the risk probability;According to the risk probability corresponding to each transaction scene and the probability lower bound value corresponding to the risk probability, determine safe user sub-set;For each safe user sub-set, determine the low-risk transaction element combination corresponding to the safe user sub-set;When user transacts using mobile terminal, if it is determined that the user sub-set to which the user belongs is safe user sub-set and the transaction element of the transaction satisfies the corresponding low-risk transaction element combination, then do not make real-time risk control to the transaction.The application can reduce the waste of resources and transaction waiting time.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of financial technology, and in particular to a method and device for supporting terminal transaction based on risk prediction. BACKGROUND

[0002] This section is intended to provide background or context to the embodiments of the application recited in the claims. The description herein is not admitted to be prior art merely by inclusion in this section.

[0003] Currently, banks need to perform real-time risk control on transactions of users in the bank, so as to ensure the security of transactions of the users, and the real-time risk control needs a large amount of resources. Compared with normal transactions, the number of risk transactions is relatively small, that is, real-time risk control is performed on all transactions, which causes waste of resources and even significantly increases the waiting time of the users. SUMMARY

[0004] Embodiments of the present application provide a method for supporting terminal transaction based on risk prediction, which comprises:

[0005] obtaining historical transaction data of bank users, classifying the bank users according to the historical transaction data, and obtaining a plurality of user subsets;

[0006] for each user subset, determining a risk probability corresponding to each transaction scenario of the user subset, and a lower bound value of probability corresponding to the risk probability;

[0007] determining a safe user subset according to the risk probability corresponding to each transaction scenario and the lower bound value of probability corresponding to the risk probability;

[0008] for each safe user subset, determining a low-risk transaction element combination corresponding to the safe user subset;

[0009] when a user uses a mobile terminal to transact, if it is determined that the user subset to which the user belongs is a safe user subset and the transaction element of the transaction satisfies the low-risk transaction element combination corresponding to the safe user subset, then no real-time risk control is performed on the transaction of the user.

[0010] Embodiments of the present application also provide a device for supporting terminal transaction based on risk prediction, which comprises:

[0011] a user classification module, configured to obtain historical transaction data of bank users, classify the bank users according to the historical transaction data, and obtain a plurality of user subsets;

[0012] a risk probability determination module, configured to, for each user subset, determine a risk probability corresponding to each transaction scenario of the user subset, and a lower bound value of probability corresponding to the risk probability;

[0013] a safe user subset determination module configured to determine a safe user subset according to the risk probability corresponding to each transaction scenario and the probability lower bound value corresponding to the risk probability;

[0014] a low-risk transaction element combination determination module configured to determine, for each safe user subset, a low-risk transaction element combination corresponding to the safe user subset;

[0015] a risk control module configured to, when a user transacts by using a mobile terminal, if it is determined that the user subset to which the user belongs is a safe user subset and the transaction element of the transaction satisfies the low-risk transaction element combination corresponding to the safe user subset, not perform real-time risk control on the transaction of the user.

[0016] The embodiment of the present application also provides a computer device, which comprises a memory, a processor and a computer program stored in the memory and capable of being run on the processor, and the processor implements the above-mentioned terminal transaction supporting method based on risk prediction when implementing the computer program.

[0017] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and the computer program implements the above-mentioned terminal transaction supporting method based on risk prediction when being executed by a processor.

[0018] The embodiment of the present application also provides a computer program product, which comprises a computer program, and the computer program implements the above-mentioned terminal transaction supporting method based on risk prediction when being executed by a processor.

[0019] Compared with the technical solution of performing risk control on all transactions in the prior art, in the embodiment of the present application, the historical transaction data of bank users is acquired, the bank users are classified according to the historical transaction data, and a plurality of user subsets are obtained; for each user subset, the risk probability corresponding to each transaction scenario of the user subset is determined, and the probability lower bound value corresponding to the risk probability is determined; a safe user subset is determined according to the risk probability corresponding to each transaction scenario and the probability lower bound value corresponding to the risk probability; for each safe user subset, a low-risk transaction element combination corresponding to the safe user subset is determined; when a user transacts by using a mobile terminal, if it is determined that the user subset to which the user belongs is a safe user subset and the transaction element of the transaction satisfies the low-risk transaction element combination corresponding to the safe user subset, real-time risk control is not performed on the transaction of the user, the risk of the huge transaction data of bank users can be controlled by using limited real-time risk control resources, and therefore the waste of resources and the waiting time of users in transactions can be reduced. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to make the technical solutions in the embodiments of the present application or the prior art clearer, the accompanying drawings needed in the embodiments or prior art description will be briefly introduced below. Obviously, the accompanying drawings in the following description only need to be some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort. In the drawings:

[0021] Figure 1 The method flow of supporting terminal transaction based on risk prediction in the embodiment of the present application Figure One ;

[0022] Figure 2 The method flow of supporting terminal transaction based on risk prediction in the embodiment of the present application Figure Two ;

[0023] Figure 3 The method flow of supporting terminal transaction based on risk prediction in the embodiment of the present application Figure Three ;

[0024] Figure 4 The method flow of supporting terminal transaction based on risk prediction in the embodiment of the present application Figure Four ;

[0025] Figure 5 The method flow of supporting terminal transaction based on risk prediction in the embodiment of the present application Figure Five ;

[0026] Figure 6 The method flow of supporting terminal transaction based on risk prediction in the embodiment of the present application Figure Six ;

[0027] Figure 7 The method flow of supporting terminal transaction based on risk prediction in the embodiment of the present application Figure Seven ;

[0028] Figure 8 The method flow of supporting terminal transaction based on risk prediction in the embodiment of the present application Figure Eight ;

[0029] Figure 9 The method flow of supporting terminal transaction based on risk prediction in the embodiment of the present application Figure Nine ;

[0030] Figure 10 The structure block of the device for supporting terminal transaction based on risk prediction in the embodiment of the present application Figure One ;

[0031] Figure 11 The structure block of the device for supporting terminal transaction based on risk prediction in the embodiment of the present application Figure Two . DETAILED DESCRIPTION

[0032] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, further detailed description of the embodiments of the present application will be given below with reference to the drawings. Herein, the illustrative embodiments of the present application and their descriptions are used to explain the present application but not to limit the present application.

[0033] Based on the problems in the prior art, the present application proposes a method for supporting terminal transaction based on risk prediction, which is implemented at the bank end. The idea of the method is as follows: some users have relatively small transaction risks, and it can be determined that the users are low-risk users. Then, based on transaction data, a low-risk transaction element combination of the low-risk users is determined. If the transaction of a user satisfies the low-risk transaction element combination, no risk control (including identity verification, such as face recognition) is performed on the transaction of the user. If it does not satisfy, risk control is performed on the transaction of the user according to the main risk type corresponding to the current transaction.

[0034] Based on the above method idea, as shown in Figure 1 , the method can specifically include the following processes:

[0035] Step 101: Obtain historical transaction data of bank users, classify the bank users according to the historical transaction data, and obtain a plurality of user sub-sets;

[0036] Step 102: For each user sub-set, determine the risk probability of each transaction scene corresponding to the user sub-set, and the probability lower bound value corresponding to the risk probability;

[0037] Step 103: Determine the safe user sub-set according to the risk probability of each transaction scene and the probability lower bound value corresponding to the risk probability;

[0038] Step 104: For each safe user sub-set, determine the low-risk transaction element combination corresponding to the safe user sub-set;

[0039] Step 105: When a user uses a mobile terminal to transact, if it is determined that the user sub-set to which the user belongs is a safe user sub-set and the transaction element of the transaction satisfies the low-risk transaction element combination corresponding to the safe user sub-set, no real-time risk control is performed on the transaction of the user.

[0040] Specifically, the historical transaction data of the user can include transaction time, transaction location, transaction amount, transaction channel, transaction counterparty, identification of whether the transaction has a risk, and risk type, etc.

[0041] In an embodiment, obtaining historical transaction data of bank users, classifying the bank users according to the historical transaction data, and obtaining a plurality of user sub-sets include:

[0042] Obtaining user dimension data of the bank users, wherein the user dimensions can include risk levels, main transaction categories, income levels, main transaction channels and payment transaction amounts;

[0043] For the obtained user dimension classification, a plurality of user dimension categories are obtained.

[0044] According to the historical transaction data of the bank users, the number of transactions of each bank corresponding to each bank user is determined, and then a transaction attribute vector corresponding to each bank user is determined, wherein the components of the transaction attribute vector correspond to the transactions of the banks one by one, and the value of each component is equal to the number of transactions of the bank corresponding to the component corresponding to the bank user;

[0045] A distance function of the bank users is determined, wherein the independent variable of the function is two bank users, and the corresponding function value is the distance between the transaction attribute vectors corresponding to the two bank users;

[0046] For each user dimension category, a plurality of users corresponding to the user dimension data belonging to the user dimension category are selected from all the users of the bank; according to the distance function of the bank users, the plurality of selected users are subjected to cluster analysis to obtain a plurality of user sub-sets corresponding to the user dimension category;

[0047] For each user sub-set obtained above, the maximum value of the proportion of the number of users corresponding to each discrete index value of the discrete index (such as risk level, age level, occupation, etc., user attribute with discrete value) in the user sub-set is determined (the ratio of the number of users corresponding to each discrete index value to the number of users in the user sub-set);

[0048] The following steps are repeatedly executed until the maximum value of the proportion of the number of users corresponding to each discrete index value of the discrete index in each user sub-set is greater than or equal to a clustering threshold value:

[0049] The user sub-set k is selected, wherein the maximum value of the proportion of the number of users corresponding to each discrete index value of the discrete index in the user sub-set k is less than the clustering threshold value; according to the distance function of the bank users, all the users contained in the user sub-set k are subjected to cluster analysis, and the plurality of new user sub-sets obtained are used to replace the user sub-set k.

[0050] The transaction elements can include transaction time, transaction location, transaction amount, transaction counterparty category, etc.

[0051] In the embodiment of the present application, as shown in Figure 2 Step 102, for each user sub-set, determines the risk probability of each transaction scenario corresponding to the user sub-set, and the probability lower bound value corresponding to the risk probability, including:

[0052] Step 201: obtaining transaction data corresponding to each transaction scenario of the user sub-set;

[0053] Step 202: for each transaction scenario, dividing the transaction data corresponding to the transaction scenario of the user sub-set into a plurality of sub-data in chronological order, so that the transaction quantity of each sub-data is greater than a transaction quantity threshold;

[0054] Step 203: for each sub-data, determining the proportion of risk data in the sub-data, and determining the proportion as a risk probability corresponding to the sub-data;

[0055] Step 204: determining the risk probability corresponding to the transaction scenario of the user sub-set as the mean of the risk probabilities corresponding to the plurality of sub-data;

[0056] Step 205: determining the variance σ of the risk probability corresponding to the transaction scenario of the user sub-set based on the risk probabilities corresponding to the plurality of sub-data;

[0057] Step 206: selecting an acceptable probability error threshold ε;

[0058] Step 207: determining the lower bound of probability corresponding to the risk probability corresponding to the transaction scenario of the user sub-set as: wherein n is the number of the plurality of sub-data.

[0059] In the embodiment of the present application, as shown in Figure 3 Step 103 comprises:

[0060] Step 301: determining the partial order of the user sub-sets according to the risk probability corresponding to each transaction scenario and the lower bound of probability corresponding to the risk probability, wherein for any two user sub-sets, the partial order can be used to determine whether the first user sub-set is safer than the second user sub-set;

[0061] Step 302: determining a plurality of maximal user sub-sets of the partial order according to the partial order of the user sub-sets, wherein the maximal user sub-set is a maximal element of the partial order;

[0062] Step 303: determining the plurality of maximal user sub-sets of the partial order as the safe user sub-sets.

[0063] It should be noted that the definition of the maximal element of the partial order is that there is no other element superior to the maximal element in the set corresponding to the partial order.

[0064] In the embodiment of the present application, as shown in Figure 4As shown in FIG. 1, step 301 determines the partial order of the user subsets according to the risk probabilities corresponding to respective transaction scenarios and the probability lower bound values corresponding to the risk probabilities, and includes:

[0065] Step 401: For any two user subsets, if for any transaction scenario, the risk probability of a first user subset of the two user subsets corresponding to the transaction scenario is less than or equal to the risk probability of a second user subset of the two user subsets corresponding to the transaction scenario, and the probability lower bound values corresponding to the risk probabilities of the first user subset corresponding to respective transaction scenarios are all less than the acceptable probability lower bound value, it is determined that the first user subset is safer than the second user subset.

[0066] It should be noted that the acceptable probability lower bound value refers to a probability that the error of the acceptable risk probability is greater than the acceptable probability error threshold ε.

[0067] In the embodiment of the present application, as shown in FIG. 1, step 104, for each safe user subset, determines the low-risk transaction element combination corresponding to the safe user subset, and includes: Figure 5

[0068] Step 501: For each transaction element combination, obtain the transaction data of the safe user subset;

[0069] Step 502: For each transaction element combination, determine the proportion of risk data in the transaction data of the transaction element combination corresponding to respective transaction scenarios, and determine the proportion as the risk probability of the transaction element combination corresponding to respective transaction scenarios;

[0070] Step 503: Determine the partial order of the transaction element combinations, wherein for any two transaction element combinations, if for any transaction scenario, the risk probability of a first transaction element combination of the two transaction element combinations corresponding to the transaction scenario is less than or equal to the risk probability of a second transaction element combination of the two transaction element combinations corresponding to the transaction scenario, it is determined that the first transaction element combination is safer than the second transaction element combination;

[0071] Step 504: Determine, according to the partial order of the transaction element combinations, a plurality of maximal transaction element combinations of the partial order, wherein the maximal transaction element combination is a maximal element of the partial order;

[0072] Step 505: Determine the plurality of maximal transaction element combinations of the partial order as the low-risk transaction element combination corresponding to the safe user subset.

[0073] The number of transaction element combinations is in exponential form of the number of transaction elements, that is, the maximal transaction element combination is determined according to the definition of the maximal element, and there is a large amount of redundant calculation, and the complexity of the calculation is also very high. ​

[0074] In the embodiment of the present application, step 504 determines, according to the partial order of the transaction element combination, a plurality of maximal transaction element combinations of the partial order, including:

[0075] 1. initializing the set of element combinations to be determined and the set of element combinations to be compared to all transaction element combinations, and initializing the set of partial order element combinations to be empty;

[0076] 2. cyclically performing the following three steps until the set of element combinations to be determined is empty:

[0077] taking one transaction element combination a from the set of element combinations to be determined, and deleting the transaction element combination a from the set of element combinations to be determined, and comparing the transaction element combination a with each transaction element combination b in the set of element combinations to be compared except the transaction element combination a;

[0078] if the transaction element combination b is safer than the transaction element combination a, deleting the transaction element combination a from the set of element combinations to be determined; if the transaction element combination a is safer than the transaction element combination b, deleting the transaction element combination b from the set of element combinations to be determined, and determining the transaction element combination b as a next transaction element combination of the transaction element combination a;

[0079] if it is confirmed that each transaction element combination in the set of element combinations to be compared except the transaction element combination a is not safer than the transaction element combination a, adding the transaction element combination a to the set of partial order element combinations, and deleting all next transaction element combinations of the transaction element combination a from the set of element combinations to be compared;

[0080] 3. taking the transaction element combinations in the set of partial order element combinations as maximal transaction element combinations.

[0081] Compared with the definition of the maximal element according to the partial order, the calculation of the above embodiment of determining the maximal transaction element combination omits a lot of redundant calculation, and the complexity is greatly reduced, especially when the number of transactions of the transaction element combination is very large.

[0082] In the embodiment of the present application, as shown in Figure 6 step 501 obtains, for each transaction element combination, the transaction data of the safe user sub-set, including:

[0083] Step 601: discretizing the corresponding value of the continuous value transaction element;

[0084] Step 602: for each transaction element combination, obtaining the transaction data of the safe user sub-set according to the corresponding value of the discretized transaction element.

[0085] Specifically, for the transaction elements being continuous values, the transaction elements are discretized. For example, for the time element, it is determined whether the transaction element is a common value in the historical transaction data of the corresponding user, and thus the value of the transaction element in the transaction data is set as a common value identifier (yes or no). For example, if a user often shops from 5 pm to 7 pm, the common values of the time element include 5 pm to 7 pm, and if the transaction time of the user is 6 pm, the common value identifier corresponding to the transaction time element is yes. For example, the payment amount of the user is usually between 1 and 200 yuan, and the common value of the transaction amount element is 1 to 200 yuan, and if the consumption amount of the user is 1000, the common value identifier corresponding to the transaction amount element is no.

[0086] In the embodiment of the present application, as shown in Figure 7 , the method further comprises:

[0087] Step 701: For various transaction element combinations, if the element combination is not a low-risk transaction element combination of the user, based on the transaction element combination, the most similar low-risk transaction element combination is determined, and based on the transaction element combination and the most similar low-risk transaction element combination, the risk transaction scenario corresponding to the transaction element combination is determined.

[0088] Step 702: When the user transacts with the mobile terminal, if it is determined that the user sub-set to which the user belongs is a safe user sub-set and the transaction elements of the transaction do not satisfy the low-risk transaction element combination corresponding to the safe user sub-set, it is determined whether the risk transaction scenario corresponding to all transaction element combinations of the transaction contains the transaction corresponding transaction scenario, and if not, no real-time risk control is performed on the transaction of the user.

[0089] In the embodiment of the present application, as shown in Figure 8 , step 701 determines the risk transaction scenario corresponding to the transaction element combination based on the transaction element combination and the most similar low-risk transaction element combination, comprising:

[0090] Step 801: Obtain the difference element combination of the transaction element combination and the most similar low-risk transaction element combination, and determine the risk transaction scenario corresponding to the transaction element combination according to the risk probability of each transaction scenario corresponding to the difference element combination.

[0091] Explanation: For example, the transaction scenario corresponding to the risk probability exceeding the specified threshold in the risk probability of each transaction scenario corresponding to the difference element combination is taken as the risk transaction scenario corresponding to the transaction element combination.

[0092] In the embodiment of the present application, as shown in Figure 9 , the method further comprises:

[0093] Step 901: sending the low-risk transaction element combination to the smart terminal of the corresponding user, and when the network signal of the user mobile terminal is not good, judging whether to perform real-time risk control on the transaction of the user according to the low-risk transaction element combination stored on the smart terminal of the user.

[0094] The acquisition, storage, use, processing and the like of data in the technical solution of the present application comply with relevant provisions of national laws and regulations.

[0095] The present application also provides a terminal transaction supporting device based on risk prediction, as described in the following embodiments. Since the principle of solving problems of the device is similar to that of the terminal transaction supporting method based on risk prediction, the implementation of the device can be referred to the implementation of the terminal transaction supporting method based on risk prediction, and the repeated parts will not be described here.

[0096] Figure 10 The terminal transaction supporting device based on risk prediction in the embodiments of the present application has the structure as shown in the figure Figure One As shown in the figure Figure 10 The device comprises:

[0097] A user classification module 02 is configured to acquire historical transaction data of a bank user, classify the bank user according to the historical transaction data, and obtain a plurality of user subsets;

[0098] A risk probability determination module 04 is configured to determine, for each user subset, a risk probability of the user subset corresponding to each transaction scenario and a probability lower bound value corresponding to the risk probability.

[0099] A safe user subset determination module 06 is configured to determine a safe user subset according to the risk probability corresponding to each transaction scenario and the probability lower bound value corresponding to the risk probability.

[0100] A low-risk transaction element combination determination module 08 is configured to determine, for each safe user subset, a low-risk transaction element combination corresponding to the safe user subset.

[0101] A risk control module 10 is configured to, when a user uses a mobile terminal to transact, if it is determined that the user subset to which the user belongs is a safe user subset and the transaction element of the transaction satisfies the low-risk transaction element combination corresponding to the safe user subset, not performing real-time risk control on the transaction of the user.

[0102] In the embodiments of the present application, the risk probability determination module 04 is specifically configured to:

[0103] acquire transaction data corresponding to each transaction scenario of the user subset;

[0104] For each transaction scenario, transaction data corresponding to the transaction scenario of the user subset is divided into a plurality of sub-data in chronological order, so that the number of transactions of each sub-data is greater than the transaction volume threshold;

[0105] For each sub-data, a proportion of risk data in the sub-data is determined, and the proportion is determined as a risk probability corresponding to the sub-data;

[0106] The risk probability corresponding to the transaction scenario of the user subset is determined as the mean of the risk probabilities corresponding to the plurality of sub-data;

[0107] Based on the risk probabilities corresponding to the plurality of sub-data, a variance σ of the risk probability corresponding to the transaction scenario of the user subset is determined;

[0108] An acceptable probability error threshold ε is selected;

[0109] It is determined that the probability lower bound value corresponding to the risk probability corresponding to the transaction scenario of the user subset is: Where n is the number of the plurality of sub-data.

[0110] In the embodiment of the application, the secure user subset determination module 06 is specifically configured to:

[0111] According to the risk probability corresponding to each transaction scenario and the probability lower bound value corresponding to the risk probability, a partial order of the user subset is determined, wherein for any two user subsets, the partial order can be used to determine whether the first user subset in the two user subsets is safer than the second user subset;

[0112] According to the partial order of the user subset, a plurality of maximal user subsets of the partial order are determined, wherein the maximal user subset is a maximal element of the partial order;

[0113] The plurality of maximal user subsets of the partial order are determined as secure user subsets.

[0114] In the embodiment of the application, the secure user subset determination module 06 is specifically configured to:

[0115] For any two user subsets, if for any transaction scenario, the risk probability corresponding to the transaction scenario of the first user subset of the two user subsets is less than or equal to the risk probability corresponding to the transaction scenario of the second user subset of the two user subsets, and the probability lower bound value corresponding to the risk probability corresponding to each transaction scenario of the first user subset is less than the acceptable probability lower bound value, it is determined that the first user subset is safer than the second user subset.

[0116] In the embodiment of the application, the low-risk transaction element combination determination module 08 is specifically configured to:

[0117] For each transaction element combination, obtain transaction data of the secure user sub-set;

[0118] For each transaction element combination, determine a proportion of risk data in transaction data of each transaction scenario corresponding to the transaction element combination, and determine the proportion as a risk probability of each transaction scenario corresponding to the transaction element combination;

[0119] Determine a partial order of transaction element combinations, wherein, for any two transaction element combinations, if, for any transaction scenario, a risk probability of a first transaction element combination of the two transaction element combinations corresponding to the transaction scenario is less than or equal to a risk probability of a second transaction element combination of the two transaction element combinations corresponding to the transaction scenario, it is determined that the first transaction element combination is safer than the second transaction element combination;

[0120] According to the partial order of transaction element combinations, determine a plurality of maximal transaction element combinations of the partial order, wherein the maximal transaction element combination is a maximal element of the partial order;

[0121] Determine the plurality of maximal transaction element combinations of the partial order as low-risk transaction element combinations corresponding to the secure user sub-set.

[0122] In the embodiment of the present application, the low-risk transaction element combination determining module 08 is specifically configured to:

[0123] For a continuous value transaction element, discretize a corresponding value of the transaction element;

[0124] For each transaction element combination, according to the discretized corresponding value of the transaction element, obtain transaction data of the secure user sub-set.

[0125] In the embodiment of the present application, further comprising:

[0126] For each transaction element combination, if the element combination is not a low-risk transaction element combination of the user, determine a most similar low-risk transaction element combination based on the transaction element combination, and determine a risk transaction scenario corresponding to the transaction element combination based on the transaction element combination and the most similar low-risk transaction element combination;

[0127] When a user uses a mobile terminal to transact, if it is determined that a user sub-set to which the user belongs is a secure user sub-set and transaction elements of the transaction do not satisfy a low-risk transaction element combination corresponding to the secure user sub-set, it is determined whether a risk transaction scenario corresponding to all transaction element combinations of the transaction contains a transaction scenario corresponding to the transaction, and if not, no real-time risk control is performed on the transaction of the user.

[0128] In the embodiment of the present application, based on the transaction element combination and the most similar low-risk transaction element combination, the risk transaction scenario corresponding to the transaction element combination is determined, including:

[0129] The difference element combination of the transaction element combination and the most similar low-risk transaction element combination is obtained, and based on the risk probability of each transaction scenario corresponding to the difference element combination, the risk transaction scenario corresponding to the transaction element combination is determined.

[0130] Description: For example, the transaction scenario corresponding to the risk probability exceeding the specified threshold in the risk probability of each transaction scenario corresponding to the difference element combination is taken as the risk transaction scenario corresponding to the transaction element combination.

[0131] In the embodiment of the present application, as shown in Figure 11 The device further includes:

[0132] The low-risk transaction element combination issuing module 12 is configured to send the low-risk transaction element combination to the smart terminal of the corresponding user, and when the network signal of the user terminal is not good, determine whether to perform real-time risk control on the transaction of the user according to the low-risk transaction element combination stored on the smart terminal of the user.

[0133] The embodiment of the present application further provides a computer device, including a memory, a processor and a computer program stored on the memory and executable on the processor, and the processor executes the computer program to realize the above-mentioned terminal transaction method based on risk prediction support.

[0134] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the above-mentioned terminal transaction method based on risk prediction support.

[0135] The embodiment of the present application further provides a computer program product, which includes a computer program, and the computer program is executed by a processor to realize the above-mentioned terminal transaction method based on risk prediction support.

[0136] Compared with the prior art technical solution of performing risk control on all transactions, in the embodiment of the present application, the historical transaction data of the bank user is acquired, the bank user is classified according to the historical transaction data, and a plurality of user subsets are obtained; for each user subset, the risk probability of each transaction scene corresponding to the user subset is determined, and the probability lower bound value corresponding to the risk probability is determined; according to the risk probability of each transaction scene and the probability lower bound value corresponding to the risk probability, a safe user subset is determined; for each safe user subset, the low-risk transaction element combination corresponding to the safe user subset is determined; when the user transacts by using the mobile terminal, if it is determined that the user subset to which the user belongs is the safe user subset and the transaction element of the transaction satisfies the low-risk transaction element combination corresponding to the safe user subset, real-time risk control is not performed on the transaction of the user, the risk of the huge user transaction data of the bank can be controlled by using limited real-time risk control resources, and therefore, the waste of resources and the transaction waiting time of the user are reduced.

[0137] Those skilled in the art will appreciate that embodiments of the present application can be supplied as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage media, etc.) having computer-usable program code embodied therein.

[0138] The present application is described in reference to flowcharts and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus create means for implementing the functions specified in the flowcharts and / or block diagrams. Figure One one or more flows and / or blocks Figure One means for carrying out the functions specified in the flowchart

[0139] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction means, which implements the functions specified in the flowcharts and / or block diagrams. Figure One one or more flows and / or blocks Figure One means for carrying out the functions specified in the flowchart

[0140] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, thus the instructions executed on the computer or other programmable data processing devices provide processes for implementing the flows Figure One The flow or flows and / or blocks Figure One The steps of the functions specified in the flow or flows and / or blocks

[0141] The above-described specific embodiments, the purpose, technical solutions and beneficial effects of the present application are further described in detail, it should be understood that the above-described is only a specific embodiment of the present application, and is not used to limit the protection scope of the present application, any modification, equivalent replacement, improvement, 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 supporting a terminal transaction based on a risk prediction, the method comprising the steps of: The method comprises the following steps: ​ obtaining historical transaction data of bank users, classifying the bank users according to the historical transaction data, and obtaining a plurality of user subsets; for each user subset, determining the risk probability of each transaction scenario corresponding to the user subset and the lower bound value of the risk probability; determining a safe user subset according to the risk probability of each transaction scenario and the lower bound value of the risk probability; for each safe user subset, determining a low-risk transaction element combination corresponding to the safe user subset; when a user uses a mobile terminal to transact, if it is determined that the user subset to which the user belongs is a safe user subset and the transaction elements of the transaction satisfy the low-risk transaction element combination corresponding to the safe user subset, no real-time risk control is performed on the transaction of the user; for each safe user subset, determining a low-risk transaction element combination corresponding to the safe user subset comprises: for each transaction element combination, obtaining transaction data of the safe user subset; for each transaction element combination, determining the proportion of risk data in the transaction data of each transaction scenario corresponding to the transaction element combination, and determining the proportion as the risk probability of each transaction scenario corresponding to the transaction element combination; determining a partial order of the transaction element combinations, wherein for any two transaction element combinations, if the risk probability of the first transaction element combination of the two transaction element combinations for any transaction scenario is less than or equal to the risk probability of the second transaction element combination of the two transaction element combinations for the transaction scenario, it is determined that the first transaction element combination is safer than the second transaction element combination; determining a plurality of maximal transaction element combinations of the partial order according to the partial order of the transaction element combinations, wherein the maximal transaction element combination is a maximal element of the partial order; determining the plurality of maximal transaction element combinations of the partial order as the low-risk transaction element combination corresponding to the safe user subset.

2. The method of claim 1, wherein, for each user subset, determining the risk probability of each transaction scenario corresponding to the user subset and the lower bound value of the risk probability comprises: obtaining transaction data of each transaction scenario corresponding to the user subset; for each transaction scenario, dividing the transaction data of the transaction scenario corresponding to the user subset into a plurality of sub-data in chronological order, so that the number of transactions of each sub-data is greater than a transaction volume threshold; for each sub-data, determining the proportion of risk data in the sub-data, and determining the proportion as the risk probability corresponding to the sub-data; determining the risk probability of the transaction scenario corresponding to the user subset as the mean of the risk probabilities corresponding to the plurality of sub-data; based on the risk probabilities corresponding to the plurality of sub-data, determining the variance σ of the risk probability of the transaction scenario corresponding to the user subset; selecting an acceptable probability error threshold ε; Determine the probability lower bound value corresponding to the risk probability of the user sub-set corresponding to the transaction scenario as: Wherein n is the number of the plurality of sub-data.

3. The method of claim 1, wherein, determining a safe user subset according to the risk probability of each transaction scenario and the lower bound value of the risk probability comprises: determining a partial order of the user subsets according to the risk probabilities corresponding to respective transaction scenarios and the probability lower bound values corresponding to the risk probabilities, wherein for any two user subsets, the partial order is used to determine whether a first user subset of the two user subsets is safer than a second user subset of the two user subsets; determining a plurality of maximal user subsets of the partial order according to the partial order of the user subsets, wherein the maximal user subset is a maximal element of the partial order; determining the plurality of maximal user subsets of the partial order as safe user subsets.

4. The method of claim 3, wherein, determining a partial order of the user subsets according to the risk probabilities corresponding to respective transaction scenarios and the probability lower bound values corresponding to the risk probabilities, comprises: for any two user subsets, if for any transaction scenario, a risk probability of a first user subset of the two user subsets corresponding to the transaction scenario is less than or equal to a risk probability of a second user subset of the two user subsets corresponding to the transaction scenario, and probability lower bound values corresponding to the risk probabilities of the first user subset corresponding to respective transaction scenarios are all less than an acceptable probability lower bound value, it is determined that the first user subset is safer than the second user subset.

5. The method of claim 1, wherein, for each transaction element combination, obtaining transaction data of the safe user subsets, comprises: for a continuous value transaction element, discretizing a corresponding value of the transaction element; for each transaction element combination, obtaining transaction data of the safe user subsets according to the discretized corresponding values of the transaction elements.

6. The method of claim 1, wherein, further comprising: for various transaction element combinations, if the element combination is not a low-risk transaction element combination of the user, determining a most similar low-risk transaction element combination based on the transaction element combination, and determining a risk transaction scenario corresponding to the transaction element combination based on the transaction element combination and the most similar low-risk transaction element combination; when the user transacts by using a mobile terminal, if it is determined that a user subset to which the user belongs is a safe user subset and transaction elements of the transaction do not satisfy a low-risk transaction element combination corresponding to the safe user subset, it is determined whether a risk transaction scenario corresponding to all transaction element combinations of the transaction contains a transaction scenario corresponding to the transaction, and if not, no real-time risk control is performed on the transaction of the user.

7. The method of claim 1, wherein, determining a risk transaction scenario corresponding to the transaction element combination based on the transaction element combination and the most similar low-risk transaction element combination, comprises: obtaining a difference element combination of the transaction element combination and the most similar low-risk transaction element combination, and determining the risk transaction scenario corresponding to the transaction element combination according to risk probabilities of respective transaction scenarios corresponding to the difference element combination.

8. The method of claim 1, wherein, further comprising: sending the low-risk transaction element combination to a smart terminal corresponding to the user, and when a network signal of the user mobile terminal is not good, determining whether to perform real-time risk control on the transaction of the user according to the low-risk transaction element combination stored on the smart terminal of the user.

9. A risk prediction based support terminal transaction device, comprising: comprising: a user classification module, configured to obtain historical transaction data of a bank user, and classify the bank user according to the historical transaction data to obtain a plurality of user subsets; The risk probability determining module is configured to determine, for each user subset, a risk probability corresponding to each transaction scenario for the user subset, and a lower bound value corresponding to the risk probability; The secure user subset determining module is configured to determine a secure user subset according to the risk probability corresponding to each transaction scenario and the lower bound value corresponding to the risk probability; The low-risk transaction element combination determining module is configured to determine, for each secure user subset, a low-risk transaction element combination corresponding to the secure user subset; The risk control module is configured to, when a user uses a mobile terminal to conduct a transaction, if it is determined that a user subset to which the user belongs is a secure user subset and a transaction element of the transaction satisfies a low-risk transaction element combination corresponding to the secure user subset, not performing real-time risk control on the transaction of the user. The low-risk transaction element combination determining module is specifically configured to: For each transaction element combination, obtain transaction data of the secure user subset; For each transaction element combination, determine a proportion of risk data in transaction data corresponding to each transaction scenario for the transaction element combination, and determine the proportion as a risk probability corresponding to each transaction scenario for the transaction element combination; Determine a partial order of transaction element combinations, wherein, for any two transaction element combinations, if, for any transaction scenario, a risk probability corresponding to the transaction scenario for a first transaction element combination of the two transaction element combinations is less than or equal to a risk probability corresponding to the transaction scenario for a second transaction element combination of the two transaction element combinations, it is determined that the first transaction element combination is safer than the second transaction element combination; Determine, according to the partial order of transaction element combinations, a plurality of maximal transaction element combinations of the partial order, wherein the maximal transaction element combination is a maximal element of the partial order; Determine the plurality of maximal transaction element combinations of the partial order as the low-risk transaction element combination corresponding to the secure user subset.

10. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the method of any one of claims 1 to 8.

11. A computer readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the method of any one of claims 1 to 8.

12. A computer program product, characterised in that, The computer program product includes a computer program, and the computer program is executed by the processor to implement the method of any one of claims 1 to 8.

Images