CN115689563A

CN115689563A - Method, device, computer equipment and storage medium for transaction secret-free payment

Info

Publication number: CN115689563A
Application number: CN202211180857.7A
Authority: CN
Inventors: 陈宇新; 吴延生; 吴少忠; 刘强
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2022-09-27
Filing date: 2022-09-27
Publication date: 2023-02-03

Abstract

The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for transaction privacy-free payment, a computer device, and a storage medium. Can be used in the field of financial science and technology or other related fields. The method comprises the following steps: extracting a target triple in the current transaction data, wherein the target triple comprises: a first entity element, a second entity element, and a relationship element between the first entity element and the second entity element; determining whether the target triple meets a triple rule or not according to the atlas vector representation of the atlas elements recorded in the knowledge atlas; and if so, determining that the current transaction is a secret-free payment transaction. Thereby achieving the effect of simplifying the payment transaction process.

Description

Method, device, computer equipment and storage medium for transaction secret-free payment

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for transaction privacy-free payment, a computer device, and a storage medium.

Background

With the development of financial technology, online and offline payment scenarios are becoming more and more abundant, and payment methods for specifying a secret-free quota and a secret-free scenario by a customer are very popular, for example, whether a secret-free payment operation can be performed can be confirmed by identifying a biological characteristic input by the customer.

However, in a transaction scenario in which the customer pays frequently, the biometric characteristics of the customer need to be repeatedly identified and confirmed, which complicates the transaction flow. Therefore, how to simplify the secret-free payment process in the transaction scenario is a problem that needs to be solved urgently at present.

Disclosure of Invention

In view of the above, it is desirable to provide a method, an apparatus, a computer device, a computer readable storage medium, and a computer program product for secure payment of a transaction, which can simplify a process of the secure payment in a transaction scenario.

In a first aspect, the present application provides a method for transaction privacy-free payment. The method comprises the following steps:

extracting a target triple in the current transaction data, wherein the target triple comprises: a first entity element, a second entity element, and a relationship element between the first entity element and the second entity element;

determining whether the target triple meets a triple rule or not according to the atlas vector representation of the atlas elements recorded in the knowledge atlas;

and if so, determining that the current transaction is a secret-free payment transaction.

In one embodiment, determining whether the target triplet satisfies the triplet rule based on a atlas vector representation of an atlas element recorded in the knowledge-atlas includes:

determining a first entity vector representation corresponding to a first entity element, a second entity vector representation corresponding to a second entity element and a relation vector representation corresponding to a relation element in a target triple according to the atlas vector representation of the atlas element recorded in the knowledge atlas;

and determining whether the target triple meets the triple rule according to the relation among the first entity vector representation, the second entity vector representation and the relation vector representation.

In one embodiment, determining whether the target triplet satisfies the triplet rule based on a relationship between the first entity vector representation, the second entity vector representation, and the relationship vector representation includes:

determining a vector sum of the first entity vector representation and the relationship vector representation;

and determining whether the target triple meets the triple rule or not according to the size relation between the vector sum and the second entity vector representation.

In one embodiment, the method further includes:

constructing a knowledge graph according to historical triples in historical transaction data, and setting graph vector representation for graph elements in the knowledge graph;

extracting positive and negative ternary group pairs from the knowledge graph;

determining vector loss values of the positive and negative triad pairs based on the triad rules and the atlas vector representation of the atlas elements;

judging whether the vector loss value meets the threshold requirement or not;

if the vector loss value does not meet the threshold requirement, updating the map vector representation of the map elements based on the vector loss value, and returning to re-execute the operation of extracting the positive and negative triad pairs from the knowledge map until the vector loss value meets the threshold requirement.

In one embodiment, setting a graph vector representation for graph elements in a knowledge graph includes:

and setting map vector representation for map elements in the knowledge map according to a preset sampling interval and vector dimensions.

In one embodiment, determining vector penalty values for the positive and negative triplet pairs based on the triplet rules and the atlas vector representation of the atlas element comprises:

determining a first vector representation of a positive triplet in the positive and negative triplet pairs and a second vector representation of a negative triplet in the positive and negative triplet pairs according to the atlas vector representations of the atlas elements;

a vector penalty value for the positive and negative triplet pair is determined based on the triplet rule, the first vector representation, and the second vector representation.

In one embodiment, updating the atlas vector representation of the atlas element based on the vector penalty value includes:

and updating the map vector representation of the map elements by adopting a gradient descending mode based on the vector loss value.

In one embodiment, the method further includes:

determining a first type of triple according to entity attribute relationship information in historical transaction data;

determining a second type of triple according to the relationship between the entities of different first type of triples;

and determining historical triples in the historical transaction data according to the first type triples and the second type triples.

In a second aspect, the application also provides a transaction privacy-free payment device. The device includes:

the extracting module is used for extracting a target triple in the current transaction data, wherein the target triple comprises: a first entity element, a second entity element, and a relationship element between the first entity element and the second entity element;

the first determining module is used for determining whether the target triple meets a triple rule or not according to the atlas vector representation of the atlas element recorded in the knowledge atlas;

and the second determining module is used for determining that the current transaction is the secret-free payment transaction if the current transaction is satisfied.

In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the following steps when executing the computer program:

if yes, determining the current transaction as a secret payment free transaction.

In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

In a fifth aspect, the present application further provides a computer program product. The computer program product comprising a computer program which when executed by a processor performs the steps of:

According to the method, the device, the computer equipment, the storage medium and the computer program product for the secret payment exempting transaction, the vector representations of the first entity element, the second entity element and the relation element in the target triple in the extracted current transaction data are assigned according to the atlas vector representations of the atlas elements recorded in the knowledge atlas, whether the target triple in the current transaction data meets the triple rule or not is judged according to the assigned vector representations of the first entity element, the second entity element and the relation element, and if yes, the current transaction is determined to be the secret payment exempting transaction. Because the target triple in the current transaction data is assigned, and whether the current transaction meets the whole process of the secret-free payment is judged according to the target triple, the customer does not need to input the biological characteristics, the secret-free payment process of the current transaction of the user can be completed under the condition that the user does not sense the secret-free payment process, and the effect of simplifying the payment transaction process is achieved.

Drawings

FIG. 1 is a diagram of an application environment of a method for secure payment for transactions in one embodiment;

FIG. 2 is a schematic flow chart diagram of a method for transaction privacy-free payment under one embodiment;

FIG. 3 is a flow diagram illustrating a process for determining whether a target triple satisfies a triple rule in one embodiment;

FIG. 4 is a schematic flow diagram illustrating the construction of a knowledge-graph and the determination of vector representations for graph elements contained therein, according to one embodiment;

FIG. 5 is a flow diagram illustrating the determination of history triples, in one embodiment;

FIG. 6 is a flow chart illustrating a method for secure payment for a transaction according to another embodiment;

FIG. 7 is a block diagram of a transaction privacy-free payment device in one embodiment;

FIG. 8 is a block diagram of a first determining module in the transaction privacy-free payment apparatus according to another embodiment;

FIG. 9 is a block diagram of a transaction privacy-free payment device according to another embodiment;

fig. 10 is a block diagram of a loss value determination module in the transaction privacy-exempt payment device according to another embodiment;

FIG. 11 is a block diagram of a transaction privacy-free payment device in another embodiment;

FIG. 12 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The transaction secret-free payment method provided by the embodiment of the application can be applied to the application environment shown in fig. 1. In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 1. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing relevant data for data processing. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of transaction privacy free payment.

In one embodiment, as shown in fig. 2, there is provided a method for transaction-based payment without secret, which is illustrated by applying the method to the computer device in fig. 1, and includes the following steps:

s201, extracting the target triple in the current transaction data.

Wherein, the target triple includes: a first entity element, a second entity element, and a relationship element between the first entity element and the second entity element.

In this embodiment, the first entity element and the second entity element in the target triple are vocabularies that can characterize an entity or an entity attribute value in the current transaction data. For example, when the first entity element is a vocabulary of entities, it may be a customer, a merchant, a transaction, and so on in the current transaction data. When the first entity element is a vocabulary of entity attribute values, it may be the name of the customer, the name of the merchant, restaurant, the amount of the transaction 100 dollars, and so on. The relationship elements in the target triplet may be a vocabulary for characterizing a relationship between the first entity element and the second entity element. When the first entity element is an entity and the second entity element is an entity attribute value, the relationship element may be an attribute relationship. For example, when the first entity element is a customer and the second entity element is Zhang III, the relationship element may be a "name". When the first entity element and the second entity element are both entities, the relationship element may be an entity relationship between the two entities. For example, when the first entity element is a transaction and the second entity element is a merchant, the relationship element may be a "place of transaction".

It should be noted that the target triple may be an "entity-relationship-entity" triple, or an "entity-attribute value" triple, and at least one of the target triples is included.

In this embodiment, there are many ways to extract the target triple, which is not limited to this. One way that can be achieved is: taking all transaction information of the current transaction as current transaction data, extracting a group of words of 'entity-relationship-entity' or a group of words of 'entity-attribute value' from the current transaction data according to a preset entity and relationship extraction rule, and taking three words in each group of words as a first entity element, a relationship element and a third entity element in a target triple respectively. Another way to implement this is: and taking all transaction information of the current transaction as current transaction data, inputting the current transaction data into a preset model for extracting the triple, and extracting the target triple in the current transaction data.

S202, determining whether the target triple meets the triple rule or not according to the atlas vector representation of the atlas element recorded in the knowledge atlas.

The graph elements may be elements corresponding to each node and edge relation in the knowledge graph, and include entity elements and relationship elements. Specifically, in this embodiment, the nodes in the knowledge graph correspond to entity elements, and the edge relationship connecting the two nodes corresponds to relationship elements. The atlas vector representation refers to representing atlas elements by using low-dimensional vectors. Optionally, in this embodiment, the atlas vector representation of the atlas element may be an accurate atlas vector representation of the atlas element, which is obtained by constructing a knowledge atlas based on triples (i.e., historical triples) extracted from a large amount of historical transaction data, and continuously updating an inaccurate initial vector representation of each atlas element in the knowledge atlas according to a transform Embedding (transform) model algorithm.

The triplet rule may be a correct triplet in which the vector representation of the first entity element plus the vector representation of the relationship element is a vector addition rule approximately equal to the vector representation of the second entity element. Illustratively, if the vector defining a correct triplet in the knowledgegraph is represented as (h, r, t), then the vector addition rule of h + r ≈ t is satisfied between the first entity vector representation h of the first entity element of the correct triplet, the second entity vector representation t of the second entity element, and the vector representation r of the relationship element in the knowledgegraph. That is, the present embodiment may determine whether the target triplet satisfies the triplet rule by determining whether the target triplet satisfies the vector addition rule of h + r ≈ t.

In addition, the triple rule may also be a scoring value of the target triple obtained by combining the vector representations of the first entity element, the second entity element and the relationship element in the target triple through a preset scoring function. And if the scoring value of the target triple meets the preset requirement, the target triple meets the triple rule. For example, if the number of the target triple is one, it is determined whether the target triple meets the triple rule by determining whether the score of the triple meets a preset threshold. If the number of the target triples is multiple, the target triples can be compared with a preset threshold value according to the average value of the scores of all the target triples, and whether the target triples meet the triplet rule or not is judged according to the comparison result.

Optionally, in this embodiment, the vector representations corresponding to the first entity element, the second entity element, and the relationship element in the target triplet are searched according to the atlas vector representations of the atlas elements recorded in the knowledge atlas, and then, according to the vector representations corresponding to the first entity element, the second entity element, and the relationship element, the method for determining the target triplet according to the triplet rule described above is combined to determine whether the target triplet satisfies the triplet rule.

S203, if yes, determining the current transaction as a secret-free payment transaction.

The secret-free payment transaction in this embodiment is a payment transaction that can be completed without a password or a biometric feature of a consumer. Specifically, in this embodiment, if it is determined that the target triple in the current transaction data satisfies the preset triple rule, it is determined that the current transaction is a secret-free payment transaction. Correspondingly, if the target triple in the current transaction data does not meet the preset triple rule, determining that the current transaction cannot be subjected to secret-free payment.

In the method for the transaction secret payment exemption, the vector representations of the first entity element, the second entity element and the relation element in the target triple in the extracted current transaction data are assigned according to the atlas vector representation of the atlas element recorded in the knowledge atlas, so that whether the target triple in the current transaction data meets the triple rule is judged, and if yes, the current transaction is determined to be the secret payment exemption transaction. Because the target triple in the current transaction data is assigned, and whether the current transaction meets the whole process of the secret-free payment is judged according to the target triple, the customer does not need to input the biological characteristics, the secret-free payment process of the current transaction of the user can be completed under the condition that the user does not sense the secret-free payment process, and the effect of simplifying the payment transaction process is achieved.

Optionally, after extracting the target triple in the current transaction data, it is necessary to determine whether the target triple satisfies the triple rule. In one embodiment, when the triplet rule is a vector addition rule with h + r ≈ t, as shown in fig. 3, determining whether the target triplet satisfies the triplet rule according to the atlas vector representation of the atlas element recorded in the knowledge-atlas includes:

s301, according to the atlas vector representation of the atlas elements recorded in the knowledge atlas, determining a first entity vector representation corresponding to a first entity element, a second entity vector representation corresponding to a second entity element, and a relation vector representation corresponding to a relation element in the target triple.

The atlas vector representation of the atlas elements recorded in the knowledge atlas is standard vector representation of each atlas element determined in advance, and comprises atlas elements serving as nodes, namely standard vector representation of entity elements, and atlas elements serving as node-edge relations, namely standard vector representation of relation elements.

In this embodiment, it may be determined whether elements (i.e., the first entity element, the second entity element, and the relationship element) in the target triple in the current transaction data exist in the knowledge graph, that is, belong to graph elements, and if yes, assign values to the elements in the target triple according to graph vector representations of the graph elements recorded in the knowledge graph. For example, if a relationship element in the current transaction data exists in the knowledge graph, assigning a value to the relationship element in the target triple by using the graph vector representation r of the relationship element recorded in the knowledge graph. It can be understood that, if none of the elements in the target triplet in the current transaction data exists in the knowledge graph, or if only part of the elements in the target triplet in the current transaction data exists in the knowledge graph, an accurate vector representation cannot be determined for the elements in the target triplet at this time, and then the target triplet cannot be subsequently determined whether the triplet rule is satisfied, so that it can be directly determined that the current transaction cannot be paid secret-free payment.

S302, determining whether the target triple meets the triple rule or not according to the relation among the first entity vector representation, the second entity vector representation and the relation vector representation.

In this embodiment, the manner of determining whether the target triplet satisfies the triplet rule may be to determine a vector sum represented by the first entity vector and the relationship vector, and determine whether the target triplet satisfies the triplet rule according to a magnitude relationship between the vector sum and the second entity vector, so as to determine whether the target triplet satisfies the preset triplet rule.

Specifically, it may be determined whether a vector sum of the first entity vector representation and the relationship vector representation is equal to the second entity vector representation, if so, it indicates that the target triplet satisfies the preset triplet rule, otherwise, it does not satisfy the preset triplet rule.

Optionally, in view that the vector sum of the first entity vector representation and the relationship vector representation is not necessarily exactly equal to the second entity vector representation, in this embodiment, it may also be determined whether the target triplet satisfies the triplet rule by determining a size relationship between a difference between the vector sum of the first entity vector representation and the relationship vector representation and the second entity vector representation and a preset threshold by presetting a threshold. If the difference between the vector sum of the first entity vector representation and the relationship vector representation and the second entity vector representation is less than the threshold, it may be determined that the target triplet satisfies the non-cryptographically paid transaction. And if the difference between the vector sum of the first entity vector representation and the relation vector representation and the second entity vector representation is larger than the threshold value, determining that the target triple does not satisfy the secret-free payment transaction.

In the embodiment, the magnitude relation between the sum of the first entity vector representation and the relation vector representation and the second entity vector representation is judged by presetting a threshold, so that whether the target triple meets the triple rule can be accurately determined, and the effect of simplifying the judgment process can be achieved.

In the above embodiment, assignment is performed on the vector representations of the first entity element and the relationship element in the target triple and the vector representation of the second entity element according to the vector representations of the map elements recorded in the knowledge map, so that the accuracy of determining the vector representations of the entity elements and the relationship elements in the target triple is ensured, and thus the judgment result can be more accurate. And then whether the current transaction belongs to the secret-free payment transaction can be judged more accurately.

In addition, in a scenario of determining whether a target triplet satisfies a triplet rule, it is necessary to first determine a map vector representation of a map element recorded in a knowledge map. In one embodiment, as shown in FIG. 4, the method includes:

s401, constructing a knowledge graph according to historical triples in historical transaction data, and setting graph vector representation for graph elements in the knowledge graph.

The historical transaction data may be information related to all payment transactions completed by the user, and may include information of a transaction client, information of a transaction merchant, transaction information of a bank card, and the like. A knowledge graph is a technical method for describing the association between knowledge and modeling world everything by using a graph model. Alternatively, the historical transaction data may be structured data including customer information, merchant information, and the like, obtained from a relational database in the financial institution system.

Optionally, the method for extracting the historical triple according to the historical transaction data in this embodiment is similar to the method for extracting the target triple from the current transaction data described in the above embodiment, and details are not repeated here. And forming a knowledge graph by taking the triple of the first entity-relation-second entity as a basic unit according to the acquired entity elements and relation elements. Specifically, the first entity and the second entity may be used as two nodes of the knowledge graph, and the relationship between the first entity and the second entity may be used as an edge relationship between the two nodes.

There are many ways to set the atlas vector representation for the atlas elements in the knowledge-atlas, and this is not a limitation. For example, an image vector representation of a preset vector dimension may be encoded for each map element according to a certain encoding rule; it is also possible to encode a map vector representation in which a preset vector dimension is randomly set for each map element.

Optionally, in an optional embodiment, a map vector representation may be set for map elements in the knowledge map according to a preset sampling interval and a vector dimension.

Wherein, the sampling interval and the vector dimension are preset, for example, the sampling interval may be

K may be a preset vector dimension, and the setting of the vector dimension may be randomly set according to the number of entity elements in the historical transaction data, and is generally between 50 and 200. It will be appreciated that the more entity elements in the historical transaction data, the greater the value of k may be. The smaller the number of entity elements in the historical transaction data, the smaller the value of k may be.

In this example, three map elements in the knowledge map are used, namely "For example, the vector representation of the first entity is defined as h, the vector representation of the relationship is defined as r, the vector representation of the second entity is defined as t, random sampling is performed in a preset sampling interval according to a preset vector dimension k, and h, r and t are assigned, for example, the vector representation of the first entity is h = [ h ] ", and the vector representation of the second entity is defined as t ₁ ,h ₂ ,...,h _k ]Vector representation of relationship r = [ r ] ₁ ,r ₂ ,...,r _k ]Vector representation of the second entity t = [ t ] ₁ ,t ₂ ,...,t _k ]。

S402, extracting positive and negative triad pairs from the knowledge graph.

Wherein, a positive and negative triplet group can be composed of a positive sample triplet and a negative sample triplet in the knowledge-graph. This embodiment may be to extract multiple sets of positive and negative triplet pairs from the knowledge-graph.

Specifically, in this embodiment, two nodes having an edge relationship are extracted, and the edge relationship between the two nodes forms a positive sample triple.

Randomly extracting elements (which can be entity elements or relationship elements) in a knowledge graph, and replacing one or more elements in the positive sample triples to obtain negative sample triples. For example, if the second entity element in the positive sample triplet (h, r, t) is replaced, the negative sample triplet is (h, r, t').

In this embodiment, the positive sample triples satisfy the vector addition rule h + r ≈ t, and the negative sample triples do not satisfy the vector addition rule, i.e., h + r ≠ t'.

And S403, determining vector loss values of the positive and negative triads based on the triad rules and the atlas vector representation of the atlas elements.

Wherein the vector loss value is used to evaluate the quality of the atlas vector representation.

There are many ways to determine the vector penalty value for a positive and negative triplet pair, and this is not a limitation. For example, in this embodiment, a first vector representation of a positive triplet in the positive and negative triplet pairs and a second vector representation of a negative triplet in the positive and negative triplet pairs may be determined from the atlas vector representations of the atlas elements, and a vector penalty value for the positive and negative triplet pairs may be determined based on the triplet rule, the first vector representation and the second vector representation.

The triple rule may be a scoring function, which may be shown in the following formula (1), to score the triple.

In the formula, D _(h,r,t) Representing the score value, L, of a triplet (h, r, t) ₂ (h + r, t) represents the Euclidean distance between vectors h + r and t, euclidean distance being a commonly used distance definition that is the true distance between two points in k-dimensional space, h represents the vector representation of the first entity of the triplet, r represents the vector representation of the relationship of the triplet, t represents the vector representation of the second entity of the triplet, k represents the dimension of the vector representation in the triplet, h represents the distance between t and h _i The ith element, r, in the vector representation representing the first entity in the triplet _i The ith element, t, in the vector representation representing the relationship in the triplet _i The ith element in the vector representation representing the second entity in the triplet. It is understood that i is all integers from 1 to k.

Based on the scoring value for each positive and negative triplet pair, a loss function is defined, which can be expressed as the following equation (2).

In the formula, loss represents the vector Loss value of the positive and negative ternary group pair, gamma is a hyperparameter and is a positive number greater than 0, and the convergence can be 1,D during calculation _(h,r,t) Representing a scored value, D, of a positive sample triplet _(h,r,t') Representing the scored value of a negative sample triplet. [] ₊ Is expressed as]When the result in (1) is less than or equal to 0, taking 0; when the result is greater than 0, the original value is selected.

Another way may be to calculate one positive vector penalty value for each triplet of positive samples and, correspondingly, one negative vector penalty value for each triplet of negative samples.

In this embodiment, the vector loss value of the positive-negative ternary pair is determined according to the loss function, and a guarantee is provided for subsequently determining whether the vector loss value of the positive-negative ternary pair meets the requirement.

S404, judging whether the vector loss value meets the threshold value requirement.

In this embodiment, a threshold may be preset to determine whether a vector loss value of the positive and negative sample triplet pair meets a requirement, and if the vector loss value is one, it may be determined that the vector loss value meets the requirement by determining whether the vector loss value is smaller than the threshold.

If the vector loss value is two, the preset threshold is also two, specifically, one threshold is preset for each of the positive sample triplet and the negative sample triplet, such as the positive sample threshold and the negative sample threshold. A comparison may be made based on the vector penalty value for the positive sample triplet with a positive sample threshold. And comparing the vector loss value of the negative sample triplet with a negative sample threshold value to obtain a comparison result. Whether the vector loss value of the positive sample triplet and the vector loss value of the negative sample triplet meet the corresponding threshold requirement or not can be judged by judging whether the vector loss values meet the threshold requirement or not. For example, if the vector loss value of a positive sample triplet is less than the corresponding positive sample threshold, and the vector loss value of a negative sample triplet is greater than the corresponding negative sample threshold, then the vector loss value is determined to satisfy the threshold requirement.

It should be noted that the smaller the loss value of the positive sample triplet, the better the representation of the representative atlas vector; accordingly, the larger the loss value of a negative sample triplet, the better the representation of the atlas vector.

And S405, if the vector loss value does not meet the requirement, updating the map vector representation of the map elements based on the vector loss value, and returning to re-execute the operation of extracting the positive and negative triad pairs from the knowledge map until the vector loss value meets the requirement of a threshold value.

There are many ways to update the atlas vector representation of the atlas element based on the vector penalty value, and this is not a limitation. One way that can be achieved is: and inputting the vector loss value, the positive sample triple and the negative sample triple into a preset updating model, and automatically outputting the updated positive sample triple and the updated negative sample triple. Another way to implement this is: and updating the map vector representation of the map elements by adopting a gradient descending mode based on the vector loss value.

In this embodiment, if the vector loss value is greater than the preset threshold, it may be determined that the vector loss value does not satisfy the requirement. Based on the vector loss value, the atlas vector representation of the atlas element is updated in a gradient descent manner. The manner in which the gradient descent is performed can be as shown in the following equation (3).

In the formula (I), the compound is shown in the specification,

to the learning rate, h _a A vector representation, r, representing the updated first entity _a Vector representation representing updated relationships, t _a Vector representation, t ', representing the updated second entity' _a A vector representation representing the second entity element in the updated negative sample triplet. h represents the vector representation of the first entity before update, r represents off before updateAnd (3) vector representation of the system, t represents the vector representation of the second entity before updating, t' represents the vector representation of the second entity element in the negative sample triplet before updating, and Loss represents the vector Loss value of the positive and negative triplet pair.

And S406, if yes, stopping updating the atlas vector representation of the positive and negative sample triples.

In this embodiment, for the positive and negative triplet pairs whose loss values do not meet the requirement, after the atlas vector representation of the atlas element is updated based on the vector loss value, the operation of extracting the positive and negative triplet pairs from the knowledge atlas is returned to be executed again until the vector loss value meets the requirement of the threshold. And updating the vector representation of the entities and the relations in the knowledge graph based on the vector loss values by adopting a gradient descending mode, so that the vector representation of the entities and the relations in the knowledge graph can be more accurate.

In the above embodiment, the positive and negative triplet sets are extracted from the knowledge graph, the loss values of the positive and negative triplets are determined by using the loss function based on the triplet rules and the vector representations of the entities and the relationships in the knowledge graph, and the vector representations of the triplets whose loss values do not meet the threshold requirement are updated until the threshold requirement is met. The accuracy of atlas vector representation of the triples of historical transaction data can be improved, so that the accuracy of assignment of the target triples of the current transaction is ensured, and the effect of enabling the judgment result to be more accurate can be achieved.

In addition, in the scene of constructing the knowledge graph according to the historical triples in the historical transaction data, the historical triples include two types. In one embodiment, as shown in fig. 5, the method further comprises:

s501, determining a first type of triple according to entity attribute relationship information in historical transaction data.

The entity attribute relationship information may be a relationship for representing a relationship between an entity and an attribute value. The first type of triplet may be the record "entity-attribute value" triplet.

In this embodiment, according to entity attribute relationship information in historical transaction data, relevant words representing entities, attributes, and attribute values are extracted to form a first type of triple.

For example, a first type of triple may be "customer-name-Zhang three," where "customer" is an entity, "Zhang three" is an attribute value, and "name" is a relationship between "customer" and "Zhang three.

S502, determining a second type of triple according to the relation between the entities of different first type of triples.

Wherein the second type of triple may be a triple that records "first entity-relationship-second entity".

In this embodiment, the second type of triple is determined according to a relationship between different entities in different first type of triples. Specifically, an entity in one first-type triple may be defined as one entity in a second-type triple, and an entity element in another first-type triple may be defined as another entity in the second-type triple, and if a certain relationship exists between the two entities, a second-type triple may be determined according to the two entities and the corresponding relationship thereof.

And S503, determining the historical triples in the historical transaction data according to the first triples and the second triples.

In this embodiment, all the determined first-type triples and all the determined second-type triples may be superimposed, and the repeated triples are removed at the same time, so as to determine the historical triples in the historical transaction data.

In the embodiment, the triples in the historical transaction data are divided into the first-class triples and the second-class triples, wherein the first-class triples are entity-attribute values, the second-class triples are first entity-relationship-second entities, and the historical triples in the historical transaction data are determined according to the first-class triples and the second-class triples, so that the historical triples can be richer, and therefore, the graph elements determined according to the historical triples can be more comprehensive, the corresponding vector representation can be found in a knowledge graph aiming at the target triples extracted by the current transaction, and guarantee is provided for accurately determining whether the current transaction supports secret-free payment.

To facilitate understanding by those skilled in the art, the transaction privacy-free payment method provided by the present disclosure is described in detail, and as shown in fig. 6, the method may include:

s601, extracting the target triple in the current transaction data.

Wherein, the target triple comprises: a first entity element, a second entity element, and a relationship element between the first entity element and the second entity element.

S602, according to the atlas vector representation of the atlas elements recorded in the knowledge atlas, determining a first entity vector representation corresponding to a first entity element, a second entity vector representation corresponding to a second entity element, and a relation vector representation corresponding to a relation element in the target triple.

S603, determining the vector sum of the first entity vector representation and the relation vector representation.

And S604, judging whether the target triple meets the triple rule or not according to the magnitude relation between the vector sum and the second entity vector representation. If yes, go to S605, otherwise go to S606.

And S605, determining the current transaction as a secret-free payment transaction.

And S606, determining the current transaction as the non-secure payment transaction.

It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.

Based on the same inventive concept, the embodiment of the application also provides a transaction secret-free payment device for realizing the transaction secret-free payment method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the method, so specific limitations in one or more embodiments of the transaction privacy-free payment device provided below can be referred to the limitations of the transaction privacy-free payment method in the above, and details are not described here.

In one embodiment, as shown in fig. 7, there is provided a transaction privacy-exempt payment device 1, comprising: an extraction module 10, a first judgment module 11 and a determination module 12, wherein:

and the extraction module 10 is used for extracting the target triple in the current transaction data.

The first judging module 11 is configured to determine whether the target triplet satisfies a triplet rule according to the atlas vector representation of the atlas element recorded in the knowledge atlas.

A first determining module 12, configured to determine that the current transaction is a crypto payment exempt transaction if the current transaction is satisfied.

In one embodiment, as shown in fig. 8, the judging module 11 includes a first determining unit 111 and a judging unit 112. Wherein:

the first determining unit 111 is configured to determine, according to a atlas vector representation of atlas elements recorded in the knowledge atlas, a first entity vector representation corresponding to a first entity element, a second entity vector representation corresponding to a second entity element, and a relationship vector representation corresponding to a relationship element in the target triple.

The determining unit 112 is configured to determine whether the target triple satisfies the triple rule according to a relationship between the first entity vector representation, the second entity vector representation, and the relationship vector representation.

In one embodiment, the determining unit 112 is further configured to determine a vector sum of the first entity vector representation and the relationship vector representation; and determining whether the target triple meets the triple rule or not according to the magnitude relation between the vector sum and the second entity vector representation.

In one embodiment, as shown in fig. 9, the above-mentioned transaction privacy-exempt payment apparatus 1 shown in fig. 7 further includes: a building module 13, an extraction module 14, a loss value determination module 15, a second decision module 16 and an update module 17. Wherein:

and the construction module 13 is configured to construct a knowledge graph according to the historical triples in the historical transaction data, and set graph vector representations for graph elements in the knowledge graph.

And the extraction module 14 is used for extracting the positive and negative ternary group pairs from the knowledge graph.

And the loss value determining module 15 is configured to determine a vector loss value of the positive and negative triplet pairs based on the triplet rules and the atlas vector representation of the atlas elements.

And a second judging module 16, configured to judge whether the vector loss value meets a threshold requirement.

An update module 17 for, if not, updating the atlas vector representation of the atlas element based on the vector penalty value

Specifically, after the update module 17 updates the map vector representation of the map element based on the vector loss value, the operation of extracting the positive and negative triplet pairs from the knowledge-map is executed again by the extraction module 14 until the vector loss value meets the threshold requirement.

In one embodiment, the building module 13 is further configured to set a map vector representation for map elements in the knowledge map according to a preset sampling interval and a vector dimension.

In one embodiment, as shown in fig. 10, the loss value determining module 15 includes: a second determination unit 151 and a third determination unit 152. Wherein:

a second determining unit 151 for determining a first vector representation of a positive triplet of the pair of positive and negative triples and a second vector representation of a negative triplet of the pair of positive and negative triples on basis of the atlas vector representation of the atlas element.

A third determining unit 152 for determining a vector penalty value for the positive and negative triplet pair based on the triplet rule, the first vector representation and the second vector representation.

In one embodiment, the update module 17 is further configured to update the atlas vector representation of the atlas element in a gradient descent manner based on the vector penalty value.

In one embodiment, as shown in fig. 11, the above-mentioned transaction privacy-exempt payment apparatus 1 shown in fig. 7 further includes: a second determination module 18, a third determination module 19 and a fourth determination module 20.

And the second determining module 18 is configured to determine the first type of triple according to the entity attribute relationship information in the historical transaction data.

A third determining module 19, configured to determine a second type of triple according to a relationship between entities of different first type of triples.

And a fourth determining module 20, configured to determine a historical triple in the historical transaction data according to the first type triple and the second type triple.

The modules in the transaction privacy-free payment device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 12. The computer device comprises a processor, a memory, a communication interface, a display screen and an input device which are connected through a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a method of transaction privacy free payment. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that the architecture shown in fig. 12 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory having a computer program stored therein and a processor that when executing the computer program performs the steps of:

In one embodiment, the processor, when executing the computer program, further performs the steps of:

and determining whether the target triple meets the triple rule or not according to the magnitude relation between the vector sum and the second entity vector representation.

extracting positive and negative ternary group pairs from the knowledge graph;

judging whether the vector loss value meets the threshold requirement or not;

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

determining whether the target triple meets a triple rule or not according to the atlas vector representation of the atlas element recorded in the knowledge atlas;

In one embodiment, the computer program when executed by the processor further performs the steps of:

extracting positive and negative ternary group pairs from the knowledge graph;

judging whether the vector loss value meets the threshold requirement or not;

In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, performs the steps of:

extracting positive and negative ternary group pairs from the knowledge graph;

judging whether the vector loss value meets the threshold requirement or not;

and setting map vector representation for map elements in the knowledge map according to a preset sampling interval and a vector dimension.

It should be noted that, the historical transaction data (including but not limited to customer personal information, customer consumption information, etc.) and the current transaction data (including but not limited to merchant transaction information, etc.) referred to in this application are both information and data authorized by the user or sufficiently authorized by each party.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include a Read-Only Memory (ROM), a magnetic tape, a floppy disk, a flash Memory, an optical Memory, a high-density embedded nonvolatile Memory, a resistive Random Access Memory (ReRAM), a Magnetic Random Access Memory (MRAM), a Ferroelectric Random Access Memory (FRAM), a Phase Change Memory (PCM), a graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic, quantum computing based transaction privacy free payment logic, etc., without limitation.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A method of transacting a privacy-exempt payment, the method comprising:

determining whether the target triple meets a triple rule according to the atlas vector representation of the atlas element recorded in the knowledge atlas;

2. The method of claim 1, wherein determining whether the target triplet satisfies a triplet rule based on a atlas vector representation of atlas elements recorded in an knowledge-atlas comprises:

determining a first entity vector representation corresponding to the first entity element, a second entity vector representation corresponding to a second entity element, and a relation vector representation corresponding to the relation element in the target triple according to a map vector representation of a map element recorded in a knowledge map;

determining whether the target triplet satisfies a triplet rule according to a relationship between the first entity vector representation, the second entity vector representation, and the relationship vector representation.

3. The method of claim 2, wherein determining whether the target triplet satisfies a triplet rule based on a relationship between the first entity vector representation, the second entity vector representation, and the relationship vector representation comprises:

and determining whether the target triple meets a triple rule or not according to the magnitude relation between the vector sum and the second entity vector representation.

4. The method according to any one of claims 1 to 3, further comprising:

extracting positive and negative ternary group pairs from the knowledge graph;

determining vector loss values for the positive and negative triad pairs based on the triad rules and the atlas vector representations of the atlas elements;

judging whether the vector loss value meets the threshold requirement or not;

if not, updating the atlas vector representation of the atlas element based on the vector loss value, and returning to re-execute the operation of extracting the positive and negative triad pairs from the knowledge atlas until the vector loss value meets the threshold requirement.

5. The method of claim 4, wherein setting a graph vector representation for graph elements in the knowledge-graph comprises:

6. The method of claim 4, wherein determining vector penalty values for the positive and negative triplet pairs based on the triplet rules and the atlas vector representation of the atlas element comprises:

determining, from the atlas vector representations of the atlas elements, a first vector representation of a positive triplet in the positive and negative triplet pair, and a second vector representation of a negative triplet in the positive and negative triplet pair;

determining a vector penalty value for the positive-negative triplet pair based on a triplet rule, the first vector representation, and the second vector representation.

7. The method of claim 4, wherein updating the atlas vector representation of the atlas element based on the vector penalty value comprises:

and updating the atlas vector representation of the atlas element in a gradient descending manner based on the vector loss value.

8. The method of claim 4, further comprising:

9. A transaction privacy-free payment device, the device comprising:

an extracting module, configured to extract a target triple in current transaction data, where the target triple includes: a first entity element, a second entity element, and a relationship element between the first entity element and the second entity element;

the judging module is used for determining whether the target triple meets the triple rule or not according to the atlas vector representation of the atlas element recorded in the knowledge atlas;

and the determining module is used for determining that the current transaction is the secret-free payment transaction if the current transaction is satisfied.

10. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 8.

11. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 8.

12. A computer program product comprising a computer program, characterized in that the computer program realizes the steps of the method of any one of claims 1 to 8 when executed by a processor.