Information processing method, device and storage medium

Abstract

The embodiments of the present application disclose an information processing method, device, and storage medium. The information processing method includes: when a transaction to be identified is obtained, obtaining transaction information to be identified corresponding to the transaction to be identified; determining target class community topology structure information matching the transaction information to be identified from multiple types of community topology structure information corresponding to multiple communities; obtaining target features matching the target class community topology structure information in the transaction information to be identified; determining the falsity of the transaction to be identified based on the target features, so as to adjust the exposure of the product corresponding to the transaction to be identified based on the falsity.

Description

Technical Field

[0001] The present application relates to the field of information processing technology, and in particular to an information processing method and device, and a storage medium. Background Art

[0002] With the continuous development of Internet technology, there is more and more information on the Internet, and transaction information is also gradually increasing.

[0003] In the prior art, false transactions are manually screened out from various transaction links such as browsing, ordering, warehousing, logistics, and evaluation. False transaction features are obtained from these false transactions, and the false transactions are identified to obtain identification information. The false transaction features and identification information are used to train a binary classification model, so as to use the binary classification model to judge the falsity of other transactions. Since these false transactions are manually determined, features are usually extracted from the perspectives of stores, commodities, categories, etc. The granularity of feature division is coarse, and the accuracy of the binary classification model obtained in this way is low, thereby reducing the accuracy in identifying false transactions. Summary of the invention

[0004] In order to solve the above technical problems, the embodiments of the present application hope to provide an information processing method and device, and a storage medium, which can improve the accuracy of identifying fraudulent transactions.

[0005] The technical solution of this application is implemented as follows:

[0006] The present invention provides an information processing method, which includes:

[0007] When a transaction to be identified is obtained, obtaining transaction information to be identified corresponding to the transaction to be identified;

[0008] Determining target class community topology structure information matching the transaction information to be identified from multiple types of community topology structure information corresponding to multiple communities;

[0009] Acquire a target feature matching the target class community topology structure information in the transaction information to be identified;

[0010] The falsity of the transaction to be identified is determined according to the target feature, so as to adjust the exposure of the commodity corresponding to the transaction to be identified based on the falsity.

[0011] The present application provides an information processing device, the device comprising:

[0012] An acquisition unit is used to acquire, when acquiring a transaction to be identified, transaction information to be identified corresponding to the transaction to be identified; and to acquire a target feature matching the target class community topology structure information from the transaction information to be identified;

[0013] A determination unit is used to determine target class community topology structure information that matches the transaction information to be identified from multiple types of community topology structure information corresponding to multiple communities; determine the falsity of the transaction to be identified based on the target characteristics, so as to adjust the exposure of the product corresponding to the transaction to be identified based on the falsity.

[0014] The present application provides an information processing device, the device comprising:

[0015] A memory, a processor and a communication bus, wherein the memory communicates with the processor via the communication bus, and the memory stores an information processing program executable by the processor. When the information processing program is executed, the information processing method described above is executed by the processor.

[0016] An embodiment of the present application provides a storage medium having a computer program stored thereon, which is applied to an information processing device, and is characterized in that the computer program implements the above-mentioned information processing method when executed by a processor.

[0017] The embodiment of the present application provides an information processing method, device, and storage medium. The information processing method includes: when a transaction to be identified is obtained, the transaction information to be identified corresponding to the transaction to be identified is obtained; from the multi-class community topology structure information corresponding to multiple communities, the target class community topology structure information matching the transaction information to be identified is determined; the target feature matching the target class community topology structure information is obtained in the transaction information to be identified; the falsity of the transaction to be identified is determined according to the target feature, so as to adjust the exposure of the commodity corresponding to the transaction to be identified based on the falsity. By adopting the above method implementation scheme, the information processing device determines the target class community topology structure information matching the transaction information to be identified from the multi-class community topology structure information corresponding to multiple communities; and obtains the target feature matching the target class community topology structure information in the transaction information to be identified. The information processing device can use the target feature to determine the falsity of the transaction to be identified. Since the topology of a false transaction is different from the topology of a non-false transaction, the information processing device can accurately determine the falsity of the transaction to be identified according to the multi-class community topology structure information corresponding to multiple communities, thereby improving the accuracy of identifying false transactions. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flow chart of an information processing method provided in an embodiment of the present application;

[0019] Figure 2 An exemplary community discovery schematic diagram based on label propagation provided in an embodiment of the present application;

[0020] Figure 3A schematic diagram of an exemplary information processing structure provided in an embodiment of the present application;

[0021] Figure 4 A schematic diagram of the structure of an information processing device provided in an embodiment of the present application Figure 1 ;

[0022] Figure 5 A schematic diagram of the structure of an information processing device provided in an embodiment of the present application Figure 2 . DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. It should be understood that the specific embodiments described here are only used to explain the present application and are not used to limit the present application.

[0024] Embodiment 1

[0025] The present application embodiment provides an information processing method, Figure 1 A flow chart of an information processing method provided in an embodiment of the present application, such as Figure 1 As shown, the information processing method may include:

[0026] S101. When a transaction to be identified is obtained, obtain transaction information to be identified corresponding to the transaction to be identified.

[0027] An information processing method provided in an embodiment of the present application is suitable for determining whether a transaction to be identified is a false transaction.

[0028] In the embodiments of the present application, the information processing device can be implemented in various forms. For example, the information processing device described in the present application may include devices such as mobile phones, cameras, tablet computers, laptop computers, PDAs, portable media players (PMPs), navigation devices, wearable devices, smart bracelets, pedometers, and devices such as digital TVs, desktop computers, and servers.

[0029] In the embodiment of the present application, the transaction to be identified may be a transaction obtained by the information processing device from other devices; the transaction to be identified may also be a transaction obtained by the information processing device according to certain conditions; the transaction to be identified may also be a transaction obtained by the information processing device through other means, and the specific method may be determined according to actual conditions, and the embodiment of the present application does not limit this.

[0030] In the embodiment of the present application, the number of transactions to be identified may be one, the number of transactions to be identified may be two, or the number of transactions to be identified may be multiple. The specific number of transactions to be identified may be determined according to actual conditions, and the embodiment of the present application does not limit this.

[0031] In the embodiment of the present application, the transaction information to be identified is all information associated with the transaction to be identified, such as the user corresponding to the transaction to be identified, the product corresponding to the transaction to be identified, the evaluation information of the transaction to be identified, the logistics information of the transaction to be identified, etc.

[0032] In an embodiment of the present application, the information processing device may obtain the transaction to be identified upon receiving a false transaction detection instruction; the information processing device may also obtain the transaction to be identified once at a preset time period; the specific manner in which the information processing device obtains the transaction to be identified may be determined based on actual conditions, and the embodiment of the present application does not limit this.

[0033] It should be noted that the preset time period can be a time period configured in the information processing device, or it can be a time period obtained by the information processing device in other ways. The specific way in which the information processing device obtains the preset time period can be determined according to actual conditions, and the embodiments of the present application do not limit this.

[0034] S102: Determine target class community topology structure information matching the transaction information to be identified from multiple types of community topology structure information corresponding to multiple communities.

[0035] In an embodiment of the present application, after the information processing device obtains the transaction information to be identified corresponding to the transaction to be identified, the information processing device can determine the target class community topology structure information matching the transaction information to be identified from multiple types of community topology structure information corresponding to multiple communities.

[0036] In the embodiment of the present application, multiple communities correspond one-to-one to multiple types of community topology structure information, that is, one community corresponds to one type of community topology structure information.

[0037] In the embodiment of the present application, the multiple types of community topology structure information may be information used to describe multiple topology structures. Specifically, the community topology structure information may be structure information of an asynchronous bipartite graph corresponding to the community.

[0038] In an embodiment of the present application, before the information processing device determines the target class community topology structure information that matches the transaction information to be identified from the multiple types of community topology structure information corresponding to the multiple communities, the information processing device also obtains multiple sample transaction information corresponding to the multiple sample transactions; the information processing device constructs an asynchronous bipartite graph based on the multiple sample transaction information; the information processing device determines the multiple types of community topology structure information based on the asynchronous bipartite graph.

[0039] In an embodiment of the present application, multiple sample transaction information refers to all information associated with multiple sample transactions, such as multiple sample users corresponding to multiple sample transactions, multiple sample commodities corresponding to multiple sample transactions, multiple evaluation information of multiple sample transactions, multiple logistics information of multiple sample transactions, etc.

[0040] In an embodiment of the present application, the process of an information processing device determining multiple types of community topology structure information based on an asynchronous bipartite graph includes: the information processing device iteratively updates the asynchronous bipartite graph to obtain an iteratively updated bipartite graph; the information processing device determines the community modularity of the iteratively updated bipartite graph based on the iteratively updated bipartite graph; when the difference value between the community modularity and the previous community modularity is less than or equal to a preset difference threshold, the information processing device determines multiple community labels corresponding to multiple sample transactions based on multiple iteratively updated user labels and multiple iteratively updated product labels; the information processing device divides the multiple sample transactions into communities based on the multiple community labels to obtain multiple communities; and obtains multiple types of community topology structure information from the multiple communities.

[0041] It should be noted that the iteratively updated bipartite graph includes a plurality of iteratively updated user labels and a plurality of iteratively updated product labels.

[0042] In an embodiment of the present application, when the difference between the community modularity and the previous community modularity is greater than a preset difference threshold, the information processing device continues to update the iteratively updated bipartite graph until the difference between the community modularity of the new bipartite graph and the previous community modularity is less than or equal to the preset difference threshold.

[0043] In an embodiment of the present application, the preset difference threshold may be a threshold configured in the information processing device; the preset difference threshold may also be a threshold received by the information processing device before the information processing device compares the community modularity with the previous community modularity; the specific manner in which the information processing device obtains the preset difference threshold may be determined based on actual conditions, and the embodiment of the present application does not limit this.

[0044] It should be noted that if the community modularity is the modularity calculated after the information processing device iteratively updates the asynchronous bipartite graph for the first time, then the previous community modularity may be the initial community modularity, and the value of the initial community modularity is 0; if the community modularity is not the modularity calculated after the information processing device iteratively updates the asynchronous bipartite graph for the first time, then the previous community modularity may be the modularity obtained before the information processing device obtains the community modularity.

[0045] In the embodiment of the present application, the calculation method of community modularity Q is shown in formula (1):

[0046]

[0047] Where m represents the total number of transactions in the asynchronous bipartite graph, k i represents the number of transactions of node i, A ij Represents the actual number of transactions between node i and node j, Represents the expected value of the number of edges between node i and node j when edges (transactions) are randomly placed. If δ(i,j) = 0, it means that node i and node j do not belong to the same community; if δ(i,j) = 1, it means that node i and node j belong to the same community.

[0048] It should be noted that if node i is a sample user node in the asynchronous bipartite graph, then node j is a sample product node in the asynchronous bipartite graph; if node i is a sample product node in the asynchronous bipartite graph, then node j is a sample user node in the asynchronous bipartite graph.

[0049] In an embodiment of the present application, multiple sample transaction information includes multiple sample transaction relations, multiple initial user labels and multiple initial product labels; the process of the information processing device constructing an asynchronous bipartite graph based on multiple sample transaction information includes the information processing device constructing an initial asynchronous bipartite graph based on multiple sample transaction relations; and determining at least one maximum clique structure and at least one maximum clique label corresponding to the at least one maximum clique structure from the initial asynchronous bipartite graph; the information processing device determines multiple initial user labels based on the at least one maximum clique label; the information processing device propagates the multiple initial user labels to multiple product nodes corresponding to the multiple user nodes in the initial asynchronous bipartite graph to obtain multiple initial product labels; the information processing device obtains an asynchronous bipartite graph based on the multiple initial user labels, the multiple initial product labels and the initial asynchronous bipartite graph.

[0050] It should be noted that the at least one maximum clique label is a label set when at least one maximum clique structure is determined; multiple user nodes correspond to multiple initial user labels one by one, that is, one user node corresponds to one initial user label.

[0051] In an embodiment of the present application, the multiple initial user tags are specifically initial user tags corresponding to multiple sample users in multiple sample transaction information; the multiple initial product tags are specifically initial product tags of multiple sample products in multiple sample transaction information.

[0052] It should be noted that the initial labels corresponding to the multiple commodity nodes are empty until the information processing device propagates the multiple initial user labels to the multiple commodity nodes according to the multiple sample transaction relationships.

[0053] In an embodiment of the present application, the process of an information processing device determining multiple initial user labels based on at least one maximum group label includes: when a first user node among multiple user nodes is in at least one maximum group structure, the information processing device determines a target maximum group structure with the most transactions and a target maximum group label of the target maximum group structure in the at least one maximum group structure; the information processing device uses the target maximum group label as the first user label of the first user node until multiple initial user labels corresponding to multiple user nodes are determined from at least one maximum group.

[0054] In an embodiment of the present application, the process of an information processing device determining multiple initial user labels based on at least one maximum group label includes: when a first user node among multiple user nodes is not in at least one maximum group structure, the information processing device obtains a first number of the first user node and uses the first number as the first user label of the first user node until multiple initial user labels corresponding to multiple users are determined from at least one maximum group.

[0055] It should be noted that the first number is a number determined when the first user node is created.

[0056] It should be noted that the information processing device aggregates at least one maximum group structure g into the set The process includes: when the information processing device determines at least one maximum group structure, the information processing device aggregates each node (user node and product node) in the maximum group structure into a set middle.

[0057] In the embodiment of the present application, after the information processing device constructs the asynchronous bipartite graph, the information processing device searches for at least one maximum clique structure g in the asynchronous bipartite graph and summarizes at least one maximum clique structure g into a set In, according to the collection Assign an initial label to each user node and obtain the initial user label corresponding to the user node, that is, as shown in formula (2), if user node i (Label i ) exists in the set In the set, the number of the largest group with the most transactions is selected as the initial user label of user node i. If user node i does not exist in the set In , the number of user node i is used as the initial user label. The initial label of the product node is empty. After initializing multiple initial user labels, some user nodes are divided into the same community in advance, thereby accelerating the iteration of label propagation.

[0058]

[0059] S103: Acquire target features that match the target class community topology structure information in the transaction information to be identified.

[0060] In an embodiment of the present application, after the information processing device determines the target class community topology structure information that matches the transaction information to be identified from multiple types of community topology structure information corresponding to multiple communities, the information processing device obtains the target features that match the target class community topology structure information in the transaction information to be identified.

[0061] It should be noted that the number of target features can be one, or the number of target features can be two; the number of target features can also be multiple; the specific number of target features can be determined according to actual conditions, and the embodiments of the present application do not limit this.

[0062] In the embodiment of the present application, the target feature can be the result of community division (multiple communities) and the features extracted from different dimensions such as community, community and product, community and user, community and order. For example, the community and product dimension refers to treating all orders under each product in each community as a group and calculating the various features of the orders in the group. In addition to general statistical features such as mean, median, standard deviation, etc., the information entropy H(X) is also calculated to reflect the impact of different indicators on each community. As shown in formula (3), P(x i ) is the amount of information x i For example, to calculate the information entropy of the coupon denomination in a group, X represents the denomination of the coupon used by the user in each order. i When P(x i ) represents the probability of using a 10-yuan coupon in all orders in the group, and the final information entropy value H(X) reflects the measure of uncertainty in the denomination of the coupons in the group.

[0063] H(X)=-∑ i P(x i )logP(x i ) (3)

[0064] S104: Determine the falsity of the transaction to be identified according to the target feature, so as to adjust the exposure of the commodity corresponding to the transaction to be identified based on the falsity.

[0065] In an embodiment of the present application, after the information processing device obtains the target feature that matches the target class community topology structure information in the transaction information to be identified, the information processing device can determine the falsity of the transaction to be identified based on the target feature.

[0066] In an embodiment of the present application, the process of an information processing device determining the falsity of a transaction to be identified based on a target feature includes: the information processing device inputs the target feature into a false transaction detection model to obtain a false label of the transaction to be identified; and the information processing device determines the falsity of the transaction to be identified based on the false label.

[0067] It should be noted that the fake transaction detection model can be a binary classification model. The specific fake transaction detection model can be an xgboost binary classification model. The fake transaction detection model can also be other models used to determine the falsity of a transaction. The specific model can be determined according to actual conditions, and the embodiments of the present application do not limit this.

[0068] In an embodiment of the present application, the information processing device inputs the target feature into the false transaction detection model. Before obtaining the false label of the transaction to be identified, the information processing device will also obtain multiple sample features that match multiple categories of community topological structure information from multiple sample transaction information; the information processing device identifies multiple sample false transaction identifiers corresponding to the multiple sample features; the information processing device uses the multiple sample features and the multiple sample false transaction identifiers to train the initial false transaction detection model to obtain a false transaction detection model.

[0069] In an embodiment of the present application, the information processing device can splice multiple sample features into a feature vector, and then the information processing device can use the feature vector and multiple sample false transaction identifiers to train an initial false transaction detection model to obtain a false transaction detection model.

[0070] For example, Figure 2 As shown, in the asynchronous bipartite graph constructed based on the purchase relationship between users and products, there are two types of nodes (user nodes and product nodes). The edges (connections) between user nodes and product nodes represent the user's behavior of purchasing products, that is, transactions. Each node (user node or product node) has a label list (the label list of the user node is shown in the square brackets on the left side of the user node, and the label list of the product node is shown in the square brackets on the right side of the product node) to record the process of label propagation. Step A (label initialization): Find at least one maximum group structure g in the constructed asynchronous bipartite graph, and store the user nodes and product nodes corresponding to at least one maximum group structure g into the set In the Assign an initial label to each user node. That is, if user node i appears in the set , the unique number of the largest group with the most edges (transactions) is selected as the initial label of node i. If user node i does not appear in the set In the example, the unique number of node i is used as the initial label (for example, as shown in the asynchronous bipartite graph on the right side of step A: the initial user label of the first user is a, the initial user label of the second user is a, the initial user label of the third user is b, the initial user label of the fourth user is c, the initial user label of the fifth user is d, and the initial user label of the sixth user is d, while the initial label of the commodity node is empty). After initialization, some user nodes are divided into the same community in advance, thereby accelerating the iteration of label propagation. Steps B to F (label propagation and modularity evaluation): Label propagation is carried out between user nodes and commodity nodes in sequence. Step B is to propagate from the user label of the user node to the commodity node to obtain the initial commodity label, that is, each user node first randomly selects a label from its own user label list as the label to be transmitted, and then transmits it to the commodity node according to the edge between the two types of nodes. Each commodity node receives and summarizes the labels and selects the label with the highest frequency to update its own label list. Therefore, after the propagation is completed, only one label is added to the label list of each commodity node, that is, the iteratively updated commodity label is obtained. Step C is to propagate the iteratively updated product label from the product node to the user node. The propagation process is similar to step B. After the propagation is completed, a new label is added to the label list of each user node to obtain the iteratively updated user label. Similarly, steps D, E, F, etc. repeat the above propagation process. Among them, each two propagations, such as steps B and C, are called an iteration. Each iteration is completed to evaluate the modularity, and the value Q of the modularity is obtained, and it is compared with the modularity Q' of the previous iteration. When the change ΔQ=QQ' of the modularity is less than or equal to the preset difference threshold, the continued propagation has little impact on the community structure, and the propagation can be stopped. If the change ΔQ of the modularity is greater than the preset difference threshold, the propagation process continues. Steps G~H (summarize community results). The scheme in the embodiment of the present application pays more attention to the community belonging to the edge in the figure, so step G calculates the belonging community for each edge. The method is to merge the label lists of the two nodes connected by each edge, and then select the label with the highest frequency of occurrence as the belonging community of the edge. If there are multiple labels with the same frequency of occurrence, one is randomly selected. In this way, each edge has a unique belonging community, and a user or product node can belong to multiple communities to form overlapping communities, such as Figure 2 As shown in the last asynchronous bipartite graph, a user node can belong to two communities at the same time (one circle corresponds to one community).

[0071] For example, Figure 3As shown, after the information processing device obtains multiple sample transaction information corresponding to the multiple sample transactions, the information processing device iteratively updates (label propagation) the asynchronous bipartite graph (label initialization) to obtain the iteratively updated bipartite graph, determines the community modularity of the iteratively updated bipartite graph according to the iteratively updated bipartite graph, and when the difference between the community modularity and the previous community modularity is less than or equal to the preset difference threshold, determines the multiple community labels corresponding to the multiple sample transactions according to the multiple iteratively updated user labels and the multiple iteratively updated commodity labels; divides the multiple sample transactions into communities according to the multiple community labels to obtain multiple communities (community division results), and obtains multiple types of community topological structure information from the multiple communities. The information processing device obtains the multiple types of community topological structure information from the multiple sample transaction information. multiple sample features (features of community dimension, features of community and commodity dimension, features of community and order dimension, features of community and target object dimension, ...) that match the information; identify multiple sample false transaction identifiers corresponding to the multiple sample features, use the multiple sample features and the multiple sample false transaction identifiers to train an initial false transaction detection model to obtain a false transaction detection model (xgboost classification model), so as to obtain the transaction information to be identified corresponding to the transaction to be identified when the transaction to be identified is obtained; determine the target class community topology structure information that matches the transaction information to be identified from the multiple classes of community topology structure information corresponding to the multiple communities; obtain the target features that match the target class community topology structure information in the transaction information to be identified; determine the falsity of the transaction to be identified according to the target features, and obtain the false transaction detection result.

[0072] It can be understood that the information processing device determines the target class community topology structure information that matches the transaction information to be identified from the multiple classes of community topology structure information corresponding to multiple communities; and obtains the target features that match the target class community topology structure information in the transaction information to be identified. The information processing device can use the target features to determine the falsity of the transaction to be identified. Since the topology structure of a false transaction is different from the topology structure of a non-false transaction, the information processing device can accurately determine the falsity of the transaction to be identified based on the multiple classes of community topology structure information corresponding to multiple communities, thereby improving the accuracy of identifying false transactions.

[0073] Embodiment 2

[0074] Based on the same inventive concept of the first embodiment, the embodiment of the present application provides an information processing device 1, corresponding to an information processing method; Figure 4 A schematic diagram of the structure of an information processing device provided in an embodiment of the present application Figure 1 , the information processing device 1 may include:

[0075] The acquisition unit 11 is used to acquire the transaction information to be identified corresponding to the transaction to be identified when the transaction to be identified is acquired; and acquire the target feature matching the target class community topology structure information from the transaction information to be identified;

[0076] The determination unit 12 is used to determine the target class community topology structure information that matches the transaction information to be identified from multiple types of community topology structure information corresponding to multiple communities; determine the falsity of the transaction to be identified based on the target characteristics, so as to adjust the exposure of the product corresponding to the transaction to be identified based on the falsity.

[0077] In some embodiments of the present application, the device further comprises a building block;

[0078] The acquisition unit 11 is used to acquire multiple sample transaction information corresponding to multiple sample transactions; and determine the multiple types of community topology structure information according to the asynchronous bipartite graph.

[0079] The construction unit is used to construct an asynchronous bipartite graph according to the multiple sample transaction information.

[0080] In some embodiments of the present application, the apparatus further includes an updating unit and a dividing unit;

[0081] The updating unit is used to iteratively update the asynchronous bipartite graph to obtain an iteratively updated bipartite graph; the iteratively updated bipartite graph includes a plurality of iteratively updated user tags and a plurality of iteratively updated product tags;

[0082] The determining unit 12 is used to determine the community modularity of the iteratively updated bipartite graph according to the iteratively updated bipartite graph; when the difference between the community modularity and the previous community modularity is less than or equal to a preset difference threshold, determine the multiple community tags corresponding to the multiple sample transactions according to the multiple iteratively updated user tags and the multiple iteratively updated commodity tags;

[0083] The division unit is used to divide the multiple sample transactions into communities according to the multiple community tags to obtain multiple communities; and to obtain the multiple types of community topology structure information from the multiple communities.

[0084] In some embodiments of the present application, the plurality of sample transaction information includes a plurality of sample transaction relationships, a plurality of initial user tags, and a plurality of initial commodity tags; the device further includes a propagation unit;

[0085] The construction unit is used to construct an initial asynchronous bipartite graph based on the multiple sample transaction relationships;

[0086] The determining unit 12 is used to determine at least one maximum clique structure and at least one maximum clique label corresponding to the at least one maximum clique structure from the initial asynchronous bipartite graph; the at least one maximum clique label is a label set when the at least one maximum clique structure is determined; and the multiple initial user labels are determined according to the at least one maximum clique label;

[0087] The propagation unit is used to propagate the multiple initial user tags to multiple commodity nodes corresponding to the multiple user nodes in the initial asynchronous bipartite graph to obtain the multiple initial commodity tags; the multiple user nodes correspond to the multiple initial user tags one by one; and the asynchronous bipartite graph is obtained according to the multiple initial user tags, the multiple initial commodity tags and the initial asynchronous bipartite graph.

[0088] In some embodiments of the present application, the determining unit 12 is used to determine, when the first user node among the multiple user nodes is in the at least one maximum group structure, a target maximum group structure with the most transactions and a target maximum group label of the target maximum group structure in the at least one maximum group structure; and use the target maximum group label as the first user label of the first user node until multiple initial user labels corresponding to multiple user nodes are determined from the at least one maximum group. .

[0089] In some embodiments of the present application,

[0090] The acquisition unit 11 is used to acquire a first number of a first user node among the multiple user nodes when the first user node is not in the at least one maximum group structure, and use the first number as a first user tag of the first user node until multiple initial user tags corresponding to multiple users are determined from the at least one maximum group, and the first number is a number determined when the first user node is created.

[0091] In some embodiments of the present application, the device further comprises an input unit;

[0092] The input unit is used to input the target feature into a false transaction detection model to obtain a false label of the transaction to be identified;

[0093] The determining unit 12 is used to determine the falsity of the transaction to be identified according to the false label.

[0094] In some embodiments of the present application, the apparatus further comprises an identification unit and a training unit;

[0095] The acquisition unit 11 is used to acquire multiple sample features matching multiple types of community topological structure information from multiple sample transaction information;

[0096] The identification unit is used to identify multiple sample false transaction identifications corresponding to the multiple sample features;

[0097] The training unit is used to train an initial false transaction detection model using the multiple sample features and the multiple sample false transaction identifiers to obtain the false transaction detection model.

[0098] It should be noted that, in actual applications, the above-mentioned acquisition unit 11 and determination unit 12 can be implemented by a processor 13 on the information processing device 1, specifically a CPU (Central Processing Unit), an MPU (Microprocessor Unit), a DSP (Digital Signal Processing) or a field programmable gate array (FPGA); the above-mentioned data storage can be implemented by a memory 14 on the information processing device 1.

[0099] The present application also provides an information processing device 1, such as Figure 5 As shown, the information processing device 1 includes: a processor 13, a memory 14 and a communication bus 15. The memory 14 communicates with the processor 13 via the communication bus 15. The memory 14 stores a program executable by the processor 13. When the program is executed, the information processing method described above is executed by the processor 13.

[0100] In practical applications, the memory 14 may be a volatile memory, such as a random access memory (RAM); or a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk (HDD) or a solid-state drive (SSD); or a combination of the above types of memory, and provide instructions and data to the processor 13.

[0101] An embodiment of the present application provides a computer-readable storage medium having a computer program thereon, and when the program is executed by the processor 13, the information processing method as described above is implemented.

[0102] It can be understood that the information processing device determines the target class community topology structure information that matches the transaction information to be identified from the multiple classes of community topology structure information corresponding to multiple communities; and obtains the target features that match the target class community topology structure information in the transaction information to be identified. The information processing device can use the target features to determine the falsity of the transaction to be identified. Since the topology structure of a false transaction is different from the topology structure of a non-false transaction, the information processing device can accurately determine the falsity of the transaction to be identified based on the multiple classes of community topology structure information corresponding to multiple communities, thereby improving the accuracy of identifying false transactions.

[0103] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of hardware embodiments, software embodiments, or embodiments in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) that contain computer-usable program code.

[0104] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, 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 device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0105] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0106] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1The steps for the functions specified in one or more boxes.

[0107] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application.

Claims

1. An information processing method, characterized in that: The method comprises: When a transaction to be identified is obtained, obtaining transaction information to be identified corresponding to the transaction to be identified; Determining target class community topology structure information matching the transaction information to be identified from multiple types of community topology structure information corresponding to multiple communities; Acquire a target feature matching the target class community topology structure information in the transaction information to be identified; Determining the falsity of the transaction to be identified according to the target feature, so as to adjust the exposure of the commodity corresponding to the transaction to be identified based on the falsity; Before determining the target class community topology structure information matching the transaction information to be identified from the multiple classes of community topology structure information corresponding to the multiple communities, the method further includes: Obtaining multiple sample transaction information corresponding to multiple sample transactions; Constructing an asynchronous bipartite graph according to the plurality of sample transaction information; According to the asynchronous bipartite graph, the multi-class community topology structure information is determined.

2. The method according to claim 1, characterized in that Determining the multi-class community topology structure information according to the asynchronous bipartite graph includes: Iteratively updating the asynchronous bipartite graph to obtain an iteratively updated bipartite graph; the iteratively updated bipartite graph includes a plurality of iteratively updated user labels and a plurality of iteratively updated product labels; Determine the community modularity of the iteratively updated bipartite graph according to the iteratively updated bipartite graph; When the difference between the community modularity and the previous community modularity is less than or equal to a preset difference threshold, determining a plurality of community tags corresponding to the plurality of sample transactions according to the plurality of iteratively updated user tags and the plurality of iteratively updated commodity tags; The plurality of sample transactions are divided into communities according to the plurality of community labels to obtain a plurality of communities; and the plurality of types of community topological structure information are obtained from the plurality of communities.

3. The method according to claim 1, characterized in that The plurality of sample transaction information includes a plurality of sample transaction relations, a plurality of initial user tags, and a plurality of initial commodity tags. The step of constructing an asynchronous bipartite graph according to the plurality of sample transaction information includes: Constructing an initial asynchronous bipartite graph based on the multiple sample transaction relationships; determining at least one maximum clique structure and at least one maximum clique label corresponding to the at least one maximum clique structure from the initial asynchronous bipartite graph; the at least one maximum clique label is a label set when determining the at least one maximum clique structure; Determining the plurality of initial user labels according to the at least one maximum group label; In the initial asynchronous bipartite graph, the multiple initial user labels are propagated to multiple commodity nodes corresponding to the multiple user nodes to obtain the multiple initial commodity labels; the multiple user nodes correspond to the multiple initial user labels one by one; The asynchronous bipartite graph is obtained according to the multiple initial user tags, the multiple initial product tags and the initial asynchronous bipartite graph.

4. The method according to claim 3, characterized in that The step of determining the plurality of initial user tags according to the at least one maximum group tag comprises: When a first user node among the plurality of user nodes is in the at least one maximum clique structure, determining a target maximum clique structure with the most transactions and a target maximum clique label of the target maximum clique structure in the at least one maximum clique structure; The target maximum clique label is used as the first user label of the first user node until a plurality of initial user labels corresponding to a plurality of user nodes are determined from the at least one maximum clique.

5. The method according to claim 3, characterized in that: The step of determining the plurality of initial user tags according to the at least one maximum group tag comprises: When a first user node among the multiple user nodes is not in the at least one maximum clique structure, a first number of the first user node is obtained, and the first number is used as a first user tag of the first user node until multiple initial user tags corresponding to multiple users are determined from the at least one maximum clique, and the first number is a number determined when the first user node is created.

6. The method according to claim 1, characterized in that The determining the falsity of the transaction to be identified according to the target feature includes: Inputting the target feature into a fake transaction detection model to obtain a fake label of the transaction to be identified; The falsity of the transaction to be identified is determined according to the false label.

7. The method according to claim 6, characterized in that Before inputting the target feature into the false transaction detection model to obtain the false label of the transaction to be identified, the method further includes: In the multiple sample transaction information, multiple sample features matching the multiple types of community topological structure information are obtained; Identify multiple sample false transaction identifiers corresponding to the multiple sample features; An initial false transaction detection model is trained using the multiple sample features and the multiple sample false transaction identifiers to obtain the false transaction detection model.

8. An information processing device, characterized in that: The device comprises: An acquisition unit is used to acquire, when acquiring a transaction to be identified, transaction information to be identified corresponding to the transaction to be identified; and to acquire a target feature matching the target class community topology structure information from the transaction information to be identified; A determination unit, configured to determine target class community topology structure information matching the transaction information to be identified from multiple classes of community topology structure information corresponding to multiple communities; determine the falsity of the transaction to be identified according to the target feature, so as to adjust the exposure of the commodity corresponding to the transaction to be identified based on the falsity; Wherein, the device further comprises a construction unit; The acquisition unit is used to acquire multiple sample transaction information corresponding to the multiple sample transactions; The construction unit is used to construct an asynchronous bipartite graph according to the plurality of sample transaction information; The determining unit is used to determine the multi-class community topology structure information according to the asynchronous bipartite graph.

9. An information processing device, characterized in that: The device comprises: A memory, a processor and a communication bus, wherein the memory communicates with the processor via the communication bus, the memory stores an information processing program executable by the processor, and when the information processing program is executed, the method according to any one of claims 1 to 7 is performed by the processor.

10. A storage medium having a computer program stored thereon, applied to an information processing device, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images