Merchant recognition method and apparatus, electronic device and computer readable medium

HK40074007BActive Publication Date: 2026-09-11TENCENT TECHNOLOGY (SHENZHEN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
HK42022063106
Authority / Receiving Office
HK · HK
Patent Type
Patents
Current Assignee / Owner
Filing Date
2022-11-02
Publication Date
2026-09-11

AI Technical Summary

Technical Problem

Existing technologies struggle to efficiently identify abnormal merchants registering in bulk, making it difficult for risk control to detect them in a timely manner. Furthermore, online computations are highly complex, resulting in significant resource waste.

Method used

By representing the target merchant name as a vector, calculating the similarity between the merchant vector and the center vector of the historical set, creating a new set, and determining the target type of merchant based on the similarity and transaction information, an incremental clustering method is adopted to reduce computational complexity and maintain the stability of set identifiers.

Benefits of technology

It improves the accuracy of merchant identification, reduces computational complexity, enables continuous online discovery of new target merchant types, avoids resource waste, and meets online real-time requirements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Embodiments of the present disclosure provide a merchant identification method and device, electronic equipment and computer readable medium. The method comprises: obtaining a target merchant name registered by a target merchant; processing the target merchant name to obtain a target merchant vector representation of the target merchant name; calculating the similarity between the target merchant vector representation and a set center vector representation of a historical set; creating a new set as a target set corresponding to the target merchant name according to the similarity; and determining the target merchant as a target type merchant according to target set information of the target set, wherein the target type merchant refers to a merchant that uses a first number of merchant names registered within a preset time range for abnormal transactions, and the similarity between any two of the first number of merchant names is less than a preset threshold. The technical solution provided by the embodiments of the present disclosure can improve the accuracy of target type merchant identification, and continuously mine new target type merchant sets online.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of computer application, in particular, to a merchant identification method and device, electronic equipment and computer readable medium. BACKGROUND

[0002] In the online payment ecology, there are many phenomena of batch registration of merchants for abnormal transaction behavior in a short time, which causes great economic losses. Although the names of these batch-registered merchants are not completely the same, they are highly similar, and this batch of merchants will be discarded after a few transactions, and if relying on transaction behavior to detect them, it will be difficult for risk control to timely discover them and thus be in a passive situation. Therefore, how to efficiently identify this batch of registered abnormal merchants is a problem worth studying.

[0003] Therefore, a new merchant identification method, device, electronic equipment and computer readable medium are needed.

[0004] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present disclosure, and therefore can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0005] The embodiments of the present disclosure provide a merchant identification method, device, electronic equipment and computer readable medium, thereby at least to some extent improving the accuracy of target type merchant identification, and continuously mining to obtain new target type merchant set.

[0006] Other characteristics and advantages of the present disclosure will become apparent from the following detailed description, or will be learned by practice of the present disclosure.

[0007] The embodiments of the present disclosure provide a merchant identification method, device, electronic equipment and computer readable medium, thereby at least to some extent improving the accuracy of target type merchant identification, and continuously mining to obtain new target type merchant set.

[0008] This disclosure provides a merchant identification device, comprising: a merchant name acquisition module configured to acquire the registered target merchant name of a target merchant; a vector representation module configured to process the target merchant name to obtain a target merchant vector representation of the target merchant name; a similarity calculation module configured to calculate the similarity between the target merchant vector representation and the set center vector representation of a historical set; a target set determination module configured to create a new set based on the similarity as the target set corresponding to the target merchant name; and a target type merchant identification module configured to determine the target merchant as a target type merchant based on the target set information of the target set, wherein the target type merchant refers to a merchant that uses a first number of merchant names registered within a preset time range to conduct abnormal transactions, and the similarity between any two of the first number of merchant names is less than a preset threshold.

[0009] In an exemplary embodiment of this disclosure, the target set determination module includes: a first target set determination unit, configured to create a new set as the target set corresponding to the target merchant name if the maximum value of the similarity is less than a similarity threshold; and a second target set determination unit, configured to determine the historical set with the highest similarity as the target set if the maximum value of the similarity is greater than or equal to the similarity threshold.

[0010] In one exemplary embodiment of this disclosure, the target set information is an identifier indicating that the target set is a target type merchant set. The target type merchant identification module includes: a first target type merchant identification unit, configured to, if the maximum value in the similarity is less than a similarity threshold, after creating a new set as the target set corresponding to the target merchant name, determine the new merchant name corresponding to the new set within the time range from the creation time to the current time of the new set; and when the new set is determined to be a target type merchant set based on the transaction information of the new merchant name and the target merchant name, determine the target merchant as a target type merchant; and a second target type merchant identification unit, configured to, if the maximum value in the similarity is greater than or equal to the similarity threshold, and when the target set is determined to be a target type merchant set based on the target set information of the target set, determine the target merchant as a target type merchant.

[0011] In one exemplary embodiment of this disclosure, the merchant identification device further includes a set center vector update module, configured to update the set center vector representation of the target set according to the target merchant vector representation.

[0012] In an exemplary embodiment of this disclosure, the set center vector update module includes a first set center vector update unit configured to determine the target merchant vector representation as the set center vector representation of the newly added set if the maximum value in the similarity is less than a similarity threshold; and a second set center vector update unit configured to determine the number of target merchants corresponding to the historical set with the highest similarity if the maximum value in the similarity is greater than or equal to the similarity threshold; determine a first weight and a second weight based on the number of target merchants; and perform a weighted summation calculation on the set center vector representation of the historical set and the target merchant vector representation based on the first weight and the second weight, so as to update the set center vector representation of the target set based on the weighted summation calculation result.

[0013] In one exemplary embodiment of this disclosure, the merchant identification device further includes: a first set identifier determination module, configured to: obtain the set identifier of the historical set if the maximum value in the similarity is less than a similarity threshold; sort the set identifiers of the historical set to obtain the largest set identifier; and determine the set identifier of the newly added set based on the increment of the largest set identifier. A second set identifier determination module is configured to: determine the set identifier of the historical set with the largest similarity as the set identifier of the target set if the maximum value in the similarity is greater than or equal to a similarity threshold.

[0014] In one exemplary embodiment of this disclosure, the vector representation module includes: a stop word removal unit configured to remove stop words from the target merchant name; a word segmentation unit configured to segment the target merchant name after stop word removal to obtain word segments of the target merchant name and part-of-speech information of the word segments; a first word segmentation weighting unit configured to set a first word segmentation weight for word segments whose part-of-speech information is a place name; a second word segmentation weighting unit configured to set a second word segmentation weight for word segments whose word length is a preset length, wherein the first word segmentation weight is greater than the second word segmentation weight; a third word segmentation weighting unit configured to set a third word segmentation weight for word segments whose part-of-speech information is not a place name and whose word length is greater than the preset length; and a vector representation unit configured to perform weighted calculation on the word segments according to the first word segmentation weight, the second word segmentation weight, and the third word segmentation weight to obtain a vector representation of the target merchant name.

[0015] This disclosure provides an electronic device, including: at least one processor; and a storage device for storing at least one program, which, when executed by the at least one processor, causes the at least one processor to implement the merchant identification method as described in the above embodiments.

[0016] This disclosure provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the merchant identification method as described in the above embodiments.

[0017] In some embodiments of the present disclosure, when a target merchant registers using a target merchant name, the target merchant name is represented by a vector, and the similarity between the target merchant vector representation and the set center vector representation of each historical set is calculated. A new set is then created based on the similarity as the target set corresponding to the target merchant name. This incrementally clusters the continuously increasing target merchant names, improving the accuracy of set partitioning without affecting existing historical sets and avoiding resource waste caused by repeated offline calculations of existing historical sets. Furthermore, based on the newly created target set, the target merchant can be identified as a target type merchant according to the target set information, enabling continuous online mining to obtain new target type merchant sets.

[0018] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0020] Figure 1 A schematic diagram of an exemplary system architecture for which the merchant identification method or apparatus of the present disclosure embodiments can be applied is shown.

[0021] Figure 2 A flowchart illustrating a merchant identification method according to an embodiment of the present disclosure is shown.

[0022] Figure 3 A flowchart illustrating a merchant identification method according to an embodiment of the present disclosure is shown.

[0023] Figure 4 A flowchart illustrating a merchant identification method according to an embodiment of the present disclosure is shown.

[0024] Figure 5 This diagram illustrates the processing flow of the target merchant name after stop words have been removed.

[0025] Figure 6 This diagram illustrates the processing flow of the target merchant name after stop words have been removed.

[0026] Figure 7 A flowchart illustrating a merchant identification method according to an embodiment of the present disclosure is shown.

[0027] Figure 8 A block diagram of a merchant identification device according to an embodiment of the present disclosure is shown schematically.

[0028] Figure 9 A schematic diagram of the structure of an electronic device suitable for implementing embodiments of the present disclosure is shown. Detailed Implementation

[0029] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0030] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0031] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in at least one hardware module or integrated circuit, or in different network and / or processor devices and / or microcontroller devices.

[0032] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0033] In traditional business scenarios, traditional clustering methods such as K-means clustering and density-based spatial clustering of applications with noise (DBSCAN) are typically used. These methods obtain set information through offline clustering and then use the offline set information online to classify newly registered merchants into existing sets according to their registered merchant names.

[0034] However, the above method has the following shortcomings.

[0035] (1) The above scheme requires a large amount of data for offline training, and the computational complexity is generally O(K*N^2). Training is very time-consuming. Even when applied to online inference, the computational complexity for a single sample is still O(K*N), which cannot meet the real-time requirements of online computing.

[0036] (2) When the above scheme is applied online, it can only classify the newly added merchant names into the set obtained by offline training. It cannot effectively mine the newly added set among the newly registered merchants. This greatly reduces the accuracy of classifying the registered merchant names into the set and limits the ability to discover new target type merchant sets.

[0037] (3) The above scheme has the problem of unstable set identifiers. If the set information is obtained through offline retraining, the set identifiers obtained after retraining will not correspond to those obtained in the previous training, thus making it incompatible with historical results. If the maliciousness of the set is reassessed based on historical data of the total amount, it will lead to a large waste of resources.

[0038] Therefore, there is a need for a new merchant identification method, device, electronic device, and computer-readable medium.

[0039] Figure 1 A schematic diagram of an exemplary system architecture 100 to which the merchant identification method or apparatus of the present disclosure embodiments can be applied is shown.

[0040] like Figure 1 As shown, system architecture 100 may include one or more of terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0041] It should be understood that Figure 1The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, there can be any number of terminal devices, networks, and servers. For example, server 105 could be a server cluster composed of multiple servers.

[0042] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Terminal devices 101, 102, and 103 can be various electronic devices with displays and web browsing capabilities, including but not limited to smartphones, tablets, laptops, desktop computers, wearable devices, virtual reality devices, smart home devices, etc.

[0043] Server 105 can be a server providing various services. For example, terminal device 103 (or terminal device 101 or 102) uploads the registered target merchant name of the target merchant to server 105. Server 105 can obtain the registered target merchant name; process the target merchant name to obtain the target merchant vector representation; calculate the similarity between the target merchant vector representation and the set center vector representation of the historical set; create a new set based on the similarity as the target set corresponding to the target merchant name; and determine the target merchant as a target type merchant based on the target set information of the target set. The target type merchant refers to a merchant that uses a first number of registered merchant names within a preset time range to conduct abnormal transactions, where the similarity between any two of the first number of merchant names is less than a preset threshold. The server then feeds back the identification result of the target merchant as a target type merchant to terminal device 103.

[0044] Figure 2 A flowchart illustrating a merchant identification method according to an embodiment of the present disclosure is shown schematically. The method provided in this disclosure can be processed by any electronic device with computing power, such as the one described above. Figure 1 In the embodiments, server 105 and / or terminal devices 102 and 103 are used as examples in the following embodiments, but this disclosure is not limited thereto.

[0045] like Figure 2 As shown, the merchant identification method provided in this embodiment may include the following steps.

[0046] In step S202, the target merchant name registered by the target merchant is obtained.

[0047] In this embodiment of the disclosure, the target merchant may be a user using terminal device 102 or 103. The user generates a registered target merchant name by operating the terminal device and sends it to server 105 through the terminal device so that the server can obtain the target merchant name registered by the target merchant.

[0048] In step S204, the target merchant name is processed to obtain the target merchant vector representation of the target merchant name.

[0049] In this embodiment of the disclosure, the target merchant name can be represented as a multi-dimensional vector containing semantic information according to processes such as stop word removal, word segmentation, and vector representation.

[0050] In step S206, the similarity between the target merchant vector representation and the set center vector representation of the historical set is calculated.

[0051] In this embodiment, the historical set can be a set of multiple merchant names obtained by pre-training offline based on historically registered merchant names to cluster them. The set center vector representation of each historical set can be determined based on the merchant vector representations of the historically registered merchant names included in each historical set. Specifically, in step S206, the similarity can be obtained by calculating the cosine distance between the target merchant name and the set center vector representations of each historical set. The formula for calculating the similarity is as follows.

[0052]

[0053] Among them, V sim Let V be a K-dimensional vector, representing the similarity between the target merchant and the K historical sets. mch V is the vector representation of the target merchant. centers V is the set center vector of all (K in this example) historical sets. center_k The set center vector of the kth historical set (e.g. Figure 5 The K historical set center vectors shown are 501. Let i be the i-th element in the set center vector of the k-th historical set. Let be the i-th element in the target merchant vector representation, and n be the dimension of the target merchant vector representation. In this embodiment, n is 200, see... Figure 5 and Figure 6 Example (Example).

[0054] In step S208, a new set is created based on similarity as the target set corresponding to the target merchant name.

[0055] In this embodiment of the disclosure, the maximum value of the similarity can be used. Create a new set as the target set corresponding to the target merchant names. The maximum similarity value. It can be represented as Each merchant name can belong to a different merchant set. The similarity between any two merchant names in a given merchant set is less than a preset set threshold.

[0056] If the maximum value of the similarity score is less than the similarity threshold, a new set can be created as the target set corresponding to the target merchant name. In this embodiment, when the maximum value of the similarity score is less than the similarity threshold, it can be considered that the target merchant name is dissimilar to the historical set. Therefore, by creating a new set directly as the target set corresponding to the target merchant name, the continuously added target merchant names can be clustered incrementally, which helps to directly mine the target type of merchants online in the future.

[0057] In an exemplary embodiment, if the maximum value of the similarity is greater than or equal to the similarity threshold, then the historical set with the highest similarity is determined as the target set.

[0058] In step S210, the target merchant is determined as a target type merchant based on the target set information of the target set. The target type merchant refers to a merchant that uses a first number of merchant names registered within a preset time range to conduct abnormal transactions, and the similarity between any two of the first number of merchant names is less than a preset threshold.

[0059] In this embodiment of the disclosure, the target set information is an identifier indicating that the target set is a target type merchant set. Specifically, when the target set is a target type merchant set, the target merchants can be confirmed as target type merchants. For a specified target set, its target set information can be determined based on the historical transaction behavior of the merchant names (including the target merchant names) included in the target set. Target type merchants can be malicious merchant types. The types that each target merchant set can belong to may include: benevolent merchant types and malicious merchant types.

[0060] The merchant identification method provided in this disclosure, when a target merchant registers using a target merchant name, represents the target merchant name as a vector and calculates the similarity between the target merchant vector representation and the set center vector representations of each historical set. A new set is created based on the similarity and used as the target set corresponding to the target merchant name. This incrementally clusters newly added target merchant names, improving the accuracy of set partitioning without affecting existing historical sets and avoiding resource waste caused by repeated offline calculations of existing historical sets. Furthermore, based on the newly created target set, the target merchant can be identified as a target type merchant according to the target set information, enabling continuous online mining to obtain new target type merchant sets.

[0061] In an exemplary embodiment, the process of processing the target merchant name to obtain a vector representation of the target merchant name may include the following steps.

[0062] Remove stop words from the target merchant name; perform word segmentation on the target merchant name after stop word removal to obtain the word segmentation of the target merchant name and the词性 information of the word segmentation; set the first word segmentation weight for the word segmentation with the词性 information being a place name; set the second word segmentation weight for the word segmentation with the word length being a preset length, where the first word segmentation weight is greater than the second word segmentation weight; set the third word segmentation weight for the word segmentation with the词性 information not being a place name and the word length being greater than the preset length; perform weighted calculation on the word segmentation according to the first word segmentation weight, the second word segmentation weight, and the third word segmentation weight to obtain the vector representation of the target merchant name.

[0063] When removing stop words from the target merchant name, first remove meaningless special symbols, etc. in the target merchant name to avoid their impact on subsequent semantic representation. Secondly, count the words that appear frequently in historical merchant names to obtain a merchant name stop word list such as "Co., Ltd.", which is included in most merchant names. Removing such words can reduce the useless information in the target merchant name and retain more discriminative words.

[0064] Figure 5 and Figure 6 The above processing flow of the target merchant name after stop word removal is schematically shown by examples. For example Figure 5 and Figure 6 As shown, the word segmentation and its词性 information of the target merchant name "Fanbabi Internet Retail Store on Ke'erqin Street, Ke'erqin District, Tongliao City" after stop word removal can be expressed as: "Tongliao City, ns", "Ke'erqin District, ns", "Ke'erqin, ns", "Street, ns", "Fan, d", "Ba, y", "Bi, n", "Internet, n", "Retail Store, n". Among them, ns indicates that the词性 information is a place name, d indicates an adverb, y indicates a modal particle, and n indicates a noun. Among them, each word segmentation can be mapped to a word vector to obtain the word vector representation of each word segmentation. For example Figure 5 and Figure 6 is represented as a 200-dimensional vector. And different weights are set according to the词性 of each word segmentation. For example Figure 5 in, the first weight set for the word segmentation of place names (for example, the word segmentations: "Tongliao City", "Ke'erqin District", "Ke'erqin", "Street") is 1.2, and the second weight set for the word segmentation with the word length being the preset length ( Figure 5 In the example, the preset length is 1) of the word segmentation (for example, the word segmentations: "Fan", "Ba", "Bi") is 0.5. For the word segmentation with the词性 information not being a place name and the word length being greater than the preset length (for example, the word segmentations: "Internet", "Retail Store") is set with the third weight of 1. The vector representation of the obtained target merchant name can be, for example Figure 5 shown as 504 in Figure 6 or mch shown as 610 in The calculation method of the vector representation V of the target merchant name is as follows:

[0065]

[0066] Among them, V word_l Let L be the word segmentation vector representation obtained by mapping the l-th word segmentation, where L is the total number of words included in the target merchant name.

[0067] In this embodiment, since location names in the target merchant categories registered in batches contain more semantic information, different weights are assigned to word segmentation based on part-of-speech information and word length. Higher weights are assigned to nouns related to place names, while lower weights are assigned to single-character segments. This increases the weight of place name segmentation, allowing for more accurate classification of target merchant names into the correct set. Simultaneously, some randomly generated words may appear in the batch-registered merchant names. These words are segmented into multiple single-character words. By reducing the weight of single-character words, the interference of meaningless words in the target merchant names on semantic representation can be avoided.

[0068] Figure 3 A flowchart illustrating a merchant identification method according to an embodiment of the present disclosure is shown.

[0069] like Figure 3 As shown, the merchant identification method provided in this embodiment may include the following steps.

[0070] In step S302, the target merchant name registered by the target merchant is obtained.

[0071] In this embodiment of the disclosure, this step can be similar to step S202, and will not be described again here.

[0072] In step S304, the target merchant name is processed to obtain the target merchant vector representation of the target merchant name.

[0073] In this embodiment of the disclosure, this step can be similar to step S204, and will not be described again here.

[0074] In step S306, the similarity between the target merchant vector representation and the set center vector representation of the historical set is calculated.

[0075] In this embodiment of the disclosure, this step can be similar to step S206, and will not be described again here.

[0076] In step S308, if the maximum value of the similarity is less than the similarity threshold, a new set is created as the target set corresponding to the target merchant name.

[0077] In this embodiment of the disclosure, the newly added set may be set with a corresponding set identifier.

[0078] In step S310, if the maximum value of the similarity is less than the similarity threshold, then after creating a new set as the target set corresponding to the target merchant name, the new merchant name corresponding to the new set is determined within the time range from the creation time to the current time.

[0079] In this embodiment of the disclosure, for example, when at 14:00 on February 10, 2021, according to Figure 5 When a new set A is created as the target set corresponding to the target merchant name "Tongliao City Keerqin District Keerqin Street Fanbabi Internet Retail Store" (hereinafter referred to as a1), the merchant names (a2 to a1) that can be assigned to set A within the time range from the creation time (i.e., creation time) of the new set A at 14:00 to the current time can be included. m (where m is an integer greater than 2) is determined as the new merchant name.

[0080] In step S312, when the newly added set is determined to be the target type merchant set based on the transaction information of the newly added merchant name and the target merchant name, the target merchant is determined to be the target type merchant.

[0081] In this embodiment of the disclosure, it can be based on a1, a2, ..., a m The transaction information determines the target set information for the newly added set (i.e., the target set). Transaction information is an identifier indicating that the merchant corresponding to the merchant name has engaged in abnormal transaction behavior. Specifically, based on the transaction information, the values ​​a1, a2, ..., a... within a preset time range can be obtained. m When the ratio of the number of merchants exhibiting abnormal transaction behavior to m is greater than the target type merchant proportion threshold, the target set information of the newly added set is considered to be: a target type merchant set. The target type merchant proportion threshold can be set to a value between 0 and 1 according to actual conditions, and this embodiment does not impose any special limitation on it. Specifically, when the target set information is "a target type merchant set", the newly added set can be confirmed as a target type merchant set, and the target merchants can be confirmed as target type merchants, thereby achieving online identification of target type merchants.

[0082] In an exemplary embodiment, the method may further include updating the set center vector representation of the target set based on the target merchant vector representation. For example, the target merchant vector representation may be determined as the set center vector representation of the newly added set.

[0083] In an exemplary embodiment, the method may further include: obtaining the set identifier of the historical set; sorting the set identifiers of the historical set to obtain the largest set identifier; and determining the set identifier of the newly added set based on the increment of the largest set identifier. In this embodiment, the set identifier of the historical set can be obtained by numbering the historical sets during offline training. For example, when 100 historical sets are obtained through offline training, these 100 historical sets can be numbered sequentially as 1, 2, 3, ..., 100. The largest set identifier is 100. The increment calculation can be, for example, an addition operation of 1, so the set identifier of the newly added set is 101.

[0084] For example, these 100 historical sets are numbered sequentially as 2, 4, 6, ..., 200. The largest set identifier is 200. The increment calculation is an addition of 2, so the set identifier of the newly added set is 202.

[0085] Figure 4 A flowchart illustrating a merchant identification method according to an embodiment of the present disclosure is shown.

[0086] like Figure 4 As shown, the merchant identification method provided in this embodiment may include the following steps.

[0087] In step S402, the target merchant name registered by the target merchant is obtained.

[0088] In this embodiment of the disclosure, this step can be similar to step S202, and will not be described again here.

[0089] In step S404, the target merchant name is processed to obtain the target merchant vector representation of the target merchant name.

[0090] In this embodiment of the disclosure, this step can be similar to step S204, and will not be described again here.

[0091] In step S406, the similarity between the target merchant vector representation and the set center vector representation of the historical set is calculated.

[0092] In this embodiment of the disclosure, this step can be similar to step S206, and will not be described again here.

[0093] In step S408, if the maximum value of the similarity is greater than or equal to the similarity threshold, then the historical set with the highest similarity is determined as the target set.

[0094] In step S410, when the target set is determined to be a target type merchant set based on the target set information of the target set, the target merchant is determined to be a target type merchant.

[0095] In this embodiment of the disclosure, the method for determining the target type of merchant set can be the same as described above. Figure 3 The relevant description of step S312 is not repeated here.

[0096] In an exemplary embodiment, the method may further include: updating the set center vector representation of the target set based on the target merchant vector representation. For example, the number of target merchants corresponding to the historical set with the highest similarity may be determined; a first weight and a second weight may be determined based on the number of target merchants; and a weighted summation calculation may be performed on the set center vector representation of the historical set and the target merchant vector representation based on the first weight and the second weight, so as to update the set center vector representation of the target set based on the weighted summation calculation result.

[0097] When the number of target merchants is m, the first weight can be expressed as: The second weight can be expressed as: The weighted summation result can be expressed as:

[0098] Among them, V center_k1 V is the set center vector representation of the updated target set. center_k Let be the set center vector representation of the target set. The updated set center vector 502 can be illustrated as follows: Figure 5 502 shown.

[0099] In an exemplary embodiment, the method may further include: determining the set identifier of the historical set with the highest similarity as the set identifier of the target set. For example... Figure 5 The set identifier shown is 503.

[0100] Figure 7 A flowchart illustrating a merchant identification method according to an embodiment of the present disclosure is shown.

[0101] like Figure 7 As shown, the merchant identification method provided in this embodiment may include the following steps.

[0102] In step S702, the cosine similarity between the target merchant vector representation and the set center vector representation of the historical set is calculated.

[0103] In step S704, it is determined whether the maximum cosine similarity is greater than the similarity threshold.

[0104] In step S706, if the maximum cosine similarity is greater than the similarity threshold, the historical set with the maximum cosine similarity is determined as the target set corresponding to the target merchant name, and the set center vector representation of the target set is updated according to the target merchant vector representation.

[0105] In step S708, if the maximum cosine similarity is less than or equal to the similarity threshold, a new set is added, and the target merchant vector representation is determined as the set center vector representation of the new set. An example diagram of this embodiment is shown below. Figure 5 , Figure 6 As shown. In Figure 5 The input consists of the target merchant name and the set centroid vector representation of the historical set. The output consists of the set identifier of the set to which the target merchant belongs and the updated set centroid vector representation.

[0106] This application effectively solves the problem that existing clustering techniques cannot detect new sets in real time, while avoiding the work of repeatedly reviewing bad merchant sets caused by unstable set identifiers. This application also uses merchant vectors containing semantic information to represent merchant names, which greatly improves the accuracy of clustering while reducing computational complexity.

[0107] From a practical application perspective, this solution has been applied to the detection of malicious merchant groups (such as malicious merchants). Currently, more than 200 malicious groups have been discovered, and more than 5,000 new registered merchants of the target type have been found. Compared with manual review and labeling, the increase in the amount of data has reached 100%. In August, the daily cumulative transaction volume of the target type merchants discovered reached 20 million.

[0108] In this embodiment, the incremental clustering scheme reduces the computational complexity of a single sample to O(K), a reduction of several orders of magnitude compared to existing technologies, thus meeting the real-time requirements of low computational complexity in online applications. The incremental clustering-based scheme effectively discovers new clusters online. When a new merchant does not belong to an existing historical set, it is set as a new set, effectively solving the problem that existing technologies cannot effectively discover newly added merchant name sets. Furthermore, this application ensures that the set identifier is auto-incrementing, guaranteeing its uniqueness and stability, avoiding the shortcomings of existing technologies that cannot guarantee the stability and uniqueness of set identifiers due to multiple training iterations.

[0109] The merchant identification method proposed in this application not only has low computational complexity, meeting the needs of real-time online computing, but also has the ability to discover new sets in real time. This not only improves the accuracy of the model in dividing sets, but also effectively mines new sets of target type merchants. In terms of scalability, this solution can not only be used to mine target type merchant groups, but also to detect abnormal personal accounts, abnormal WeChat groups, and other information.

[0110] The following describes an embodiment of the apparatus disclosed herein, which can be used to execute the merchant identification method described above. For details not disclosed in the apparatus embodiments of this disclosure, please refer to the embodiments of the merchant identification method described above.

[0111] Figure 8A block diagram of a merchant identification device according to an embodiment of the present disclosure is shown schematically.

[0112] Reference Figure 8 As shown, a merchant identification device 800 according to an embodiment of the present disclosure may include: a merchant name acquisition module 810, a vector representation module 820, a similarity calculation module 830, a target set determination module 840, and a target type merchant identification module 850.

[0113] The merchant name acquisition module 810 can be configured to acquire the target merchant name registered by the target merchant.

[0114] The vector representation module 820 can be configured to process the target merchant name to obtain the target merchant vector representation of the target merchant name.

[0115] The similarity calculation module 830 can be configured to calculate the similarity between the target merchant vector representation and the set center vector representation of the historical set.

[0116] The target set determination module 840 can be configured to create a new set based on the similarity as the target set corresponding to the target merchant name.

[0117] The target type merchant identification module 850 can be configured to identify the target merchant as a target type merchant based on the target set information of the target set. The target type merchant refers to a merchant that uses a first number of merchant names registered within a preset time range to conduct abnormal transactions, and the similarity between any two of the first number of merchant names is less than a preset threshold.

[0118] The merchant identification method provided in this disclosure, when a target merchant registers using a target merchant name, represents the target merchant name as a vector and calculates the similarity between the target merchant vector representation and the set center vector representations of each historical set. A new set is created based on the similarity and used as the target set corresponding to the target merchant name. This incrementally clusters newly added target merchant names, improving the accuracy of set partitioning without affecting existing historical sets and avoiding resource waste caused by repeated offline calculations of existing historical sets. Furthermore, based on the newly created target set, the target merchant can be identified as a target type merchant according to the target set information, enabling continuous online mining to obtain new target type merchant sets.

[0119] In an exemplary embodiment, the target set determination module 840 may include: a first target set determination unit, which may be configured to create a new set as the target set corresponding to the target merchant name if the maximum value in the similarity is less than the similarity threshold; and a second target set determination unit, which may be configured to determine the historical set with the highest similarity as the target set if the maximum value in the similarity is greater than or equal to the similarity threshold.

[0120] In an exemplary embodiment, the target set information is an identifier indicating that the target set is a target type merchant set. The target type merchant identification module 850 may include: a first target type merchant identification unit, configurable to: if the maximum value in the similarity is less than a similarity threshold, then after creating a new set as the target set corresponding to the target merchant name, determine the new merchant name corresponding to the new set within the time range from the creation time to the current time; and when the new set is determined to be a target type merchant set based on the transaction information of the new merchant name and the target merchant name, determine the target merchant as a target type merchant; and a second target type merchant identification unit, configurable to: if the maximum value in the similarity is greater than or equal to a similarity threshold, and when the target set is determined to be a target type merchant set based on the target set information of the target set, determine the target merchant as a target type merchant.

[0121] In an exemplary embodiment, the merchant identification device 800 may further include a set center vector update module, which can be configured to update the set center vector representation of the target set according to the target merchant vector representation.

[0122] In an exemplary embodiment, the set center vector update module may include a first set center vector update unit, which may be configured to determine the target merchant vector representation as the set center vector representation of the newly added set if the maximum value in the similarity is less than a similarity threshold; and a second set center vector update unit, which may be configured to determine the number of target merchants corresponding to the historical set with the highest similarity if the maximum value in the similarity is greater than or equal to the similarity threshold; determine a first weight and a second weight based on the number of target merchants; and perform a weighted summation calculation on the set center vector representation of the historical set and the target merchant vector representation based on the first weight and the second weight, so as to update the set center vector representation of the target set based on the weighted summation calculation result.

[0123] In an exemplary embodiment, the merchant identification device 800 may further include: a first set identifier determination module, configured to: obtain the set identifier of the historical set if the maximum value in the similarity is less than a similarity threshold; sort the set identifiers of the historical set to obtain the largest set identifier; and determine the set identifier of the newly added set based on the increment of the largest set identifier. A second set identifier determination module, configured to: determine the set identifier of the historical set with the largest similarity as the set identifier of the target set if the maximum value in the similarity is greater than or equal to a similarity threshold.

[0124] In an exemplary embodiment, the vector representation module 820 may include: a stop word removal unit, configurable to remove stop words from the target merchant name; a word segmentation unit, configurable to segment the target merchant name after stop word removal to obtain word segments of the target merchant name and part-of-speech information of the word segments; a first word segmentation weight unit, configurable to set a first word segmentation weight for word segments whose part-of-speech information is a place name; a second word segmentation weight unit, configurable to set a second word segmentation weight for word segments whose word length is a preset length, wherein the first word segmentation weight is greater than the second word segmentation weight; a third word segmentation weight unit, configurable to set a third word segmentation weight for word segments whose part-of-speech information is not a place name and whose word length is greater than the preset length; and a vector representation unit, configurable to perform a weighted calculation on the word segments according to the first word segmentation weight, the second word segmentation weight, and the third word segmentation weight to obtain a vector representation of the target merchant name.

[0125] Figure 9 A schematic diagram of an electronic device suitable for implementing embodiments of the present disclosure is shown. It should be noted that... Figure 9 The electronic device 900 shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments disclosed herein.

[0126] like Figure 9 As shown, the electronic device 900 includes a central processing unit (CPU) 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage section 908 into a random access memory (RAM) 903. The RAM 903 also stores various programs and data required for system operation. The CPU 901, ROM 902, and RAM 903 are interconnected via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0127] The following components are connected to I / O interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to I / O interface 905 as needed. A removable medium 911, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 910 as needed so that computer programs read from it can be installed into storage section 908 as needed.

[0128] In particular, according to embodiments of this disclosure, the processes described below with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 909, and / or installed from removable medium 911. When the computer program is executed by central processing unit (CPU) 901, it performs various functions defined in the system of this application.

[0129] It should be noted that the computer-readable medium disclosed herein may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having at least one wire, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0130] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing at least one executable instruction for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0131] The modules and / or units described in the embodiments of this disclosure can be implemented in software or hardware, and can also be located in a processor. The names of these modules and / or units do not necessarily constitute a limitation on the module and / or unit itself.

[0132] On the other hand, this application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to perform the methods described in the following embodiments. For example, the electronic device may perform... Figure 2 or Figure 3 or Figure 4 or Figure 7 The steps shown.

[0133] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0134] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of this disclosure.

[0135] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0136] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A merchant identification method, characterized in that, include: Obtain the target merchant's registered name; The target merchant name is processed to obtain the target merchant vector representation of the target merchant name; Calculate the similarity between the target merchant's vector representation and the set center vector representation of the historical set; If the maximum value of the similarity is less than the similarity threshold, a new set is created as the target set corresponding to the target merchant name; the new merchant names corresponding to the new set are determined within the time range from the creation time to the current time. The process of identifying the target merchant as a target type merchant based on the target set information of the target set includes: determining the new set as a target type merchant set based on the transaction information of the newly added merchant name and the target merchant name; wherein, the target type merchant refers to a merchant that uses a first number of merchant names registered within a preset time range to conduct abnormal transactions, and the similarity between any two of the first number of merchant names is less than a preset threshold.

2. The method as described in claim 1, characterized in that, Also includes: If the maximum value of the similarity is greater than or equal to the similarity threshold, then the historical set with the highest similarity is determined as the target set.

3. The method as described in claim 2, characterized in that, The target set information is an identifier indicating that the target set is a target type merchant set; wherein, determining the target merchant as a target type merchant based on the target set information of the target set includes: If the maximum value of the similarity is greater than or equal to the similarity threshold, and the target set is determined to be a target type merchant set based on the target set information, then the target merchant is determined to be a target type merchant.

4. The method as described in claim 2, characterized in that, Also includes: Update the set center vector representation of the target set according to the target merchant vector representation.

5. The method as described in claim 4, characterized in that, Updating the set center vector representation of the target set based on the target merchant vector representation includes: If the maximum value of the similarity is less than the similarity threshold, then the target merchant vector representation is determined as the set center vector representation of the newly added set; If the maximum value of the similarity is greater than or equal to the similarity threshold, then the number of target merchants corresponding to the historical set with the highest similarity is determined; a first weight and a second weight are determined based on the number of target merchants; and a weighted summation is performed on the set center vector representation of the historical set and the target merchant vector representation based on the first weight and the second weight, so as to update the set center vector representation of the target set based on the weighted summation result.

6. The method as described in claim 2, characterized in that, Also includes: If the maximum value in the similarity is less than the similarity threshold, the set identifier of the historical set is obtained; the set identifiers of the historical set are sorted to obtain the largest set identifier; and the set identifier of the new set is determined by calculating the increment of the largest set identifier. If the maximum value of the similarity is greater than or equal to the similarity threshold, the set identifier of the historical set with the highest similarity is determined as the set identifier of the target set.

7. The method as described in claim 1, characterized in that, Processing the target merchant name to obtain its vector representation includes: Remove stop words from the target merchant name; The target merchant name after removing stop words is segmented to obtain the segmented words and part-of-speech information of the segmented words; Assign the first weight to the word segment whose part-of-speech information is a place name; A second segmentation weight is assigned to the word segment with a preset length, and the first segmentation weight is greater than the second segmentation weight; The third segmentation weight is assigned to segmentation words whose part-of-speech information is not a place name and whose word length is greater than the preset length. The word segments are weighted according to the first word segmentation weight, the second word segmentation weight, and the third word segmentation weight to obtain a vector representation of the target merchant name.

8. A merchant identification device, characterized in that, include: The merchant name acquisition module is configured to retrieve the registered merchant name of the target merchant. The vector representation module is configured to process the target merchant name to obtain a target merchant vector representation of the target merchant name; The similarity calculation module is configured to calculate the similarity between the target merchant's vector representation and the set center vector representation of the historical set; The target set determination module is configured to, if the maximum value in the similarity is less than the similarity threshold, create a new set as the target set corresponding to the target merchant name; and determine the new merchant names corresponding to the new set within the time range from the creation time to the current time. The target type merchant identification module is configured to identify the target merchant as a target type merchant based on the target set information of the target set. This includes: determining the newly added set as a target type merchant set based on the transaction information of the newly added merchant name and the target merchant name; wherein, the target type merchant refers to a merchant that uses a first number of merchant names registered within a preset time range to conduct abnormal transactions, and the similarity between any two of the first number of merchant names is less than a preset threshold.

9. An electronic device, characterized in that, include: At least one processor; Storage device for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method as described in any one of claims 1-7.

10. A computer-readable medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-7.

11. A computer program product comprising a computer program, characterized in that, When the computer program is executed, it implements the method as described in any one of claims 1-7.