Method and device for identifying virtual currency exchange names based on counterparty matching

Abstract

The present invention discloses a method and device for identifying the name of a virtual currency exchange based on counterparty matching, the method comprising: constructing a virtual currency address cluster data set; constructing a virtual currency address label data set; performing label address expansion; performing label category recognition model training to obtain a supervised multi-classification label recognition model; performing category prediction on address clusters with unknown labels and transaction times exceeding a threshold, that is, obtaining a suspected exchange address cluster; extracting an incoming counterparty address set of the suspected exchange address cluster; and identifying the label name of the suspected exchange address cluster. The present invention uses an incoming counterparty wallet address matching method based on a known virtual currency exchange wallet address to identify a virtual currency exchange entity in a virtual currency scenario, and is suitable for related evidence collection in the supervision process of related institutions in virtual currency application scenarios to ensure the security of cyberspace.

Description

Technical Field

[0001] The present invention belongs to the technical field of blockchain application security, and in particular relates to a method and device for identifying the name of a virtual currency exchange based on counterparty matching. Background Art

[0002] Compared with the traditional financial system, the unique characteristics of virtual currency (address anonymization, decentralized transactions, etc.) make virtual currency transactions have strong anti-traceability, but also make the identity recognition mechanism of virtual currency transaction addresses face more challenges. Faced with the ever-expanding virtual currency market in the blockchain industry, its supervision is still very difficult. The corresponding technical characteristics of virtual currency are mainly reflected in the following aspects:

[0003] (1) Address anonymity. A virtual currency address is an account used by a user to participate in virtual currency transactions. However, this address is created by the user himself and is not associated with his identity information. In addition, no third party is required to participate in the creation process.

[0004] (2) Transaction complexity. The virtual currency payment system allows users to generate different addresses after each transaction. Therefore, the user’s transaction information is scattered in different addresses. At the same time, these addresses involve a large number of virtual currency transactions. Therefore, it is difficult to identify the identity of a virtual currency address by analyzing the transaction behavior data of a single virtual currency address.

[0005] In summary, the anonymity of virtual currency addresses and the divergence of transactions have become obstacles to the supervision of illegal use of virtual currency for illegal activities. In order to solve the problems of anonymity and complexity of virtual currency addresses, the principle of most existing solutions is to collect virtual currency exchange wallet addresses through different open source channels, and then identify virtual currency identities based on labels. Although such methods can be expanded by artificial intelligence based on known virtual currency label categories (exchanges, coin mixers, service providers, etc.), they cannot identify the specific ownership of the address (such as the specific names of wallet addresses of Huobi Exchange, Helixmixer coin mixers, etc.). Such methods result in us only being able to conduct illegal fund investigation from known specific wallet addresses, which is too limited because open source channels are limited in collecting exchange label addresses. Summary of the invention

[0006] In order to solve the problem that there are limitations in the existing information collection of virtual currency exchange wallet addresses through open source channels, the present invention provides a virtual currency exchange name identification method based on counterparty matching. The present invention uses the method of matching the incoming counterparty wallet address based on the known virtual currency exchange wallet address to identify the virtual currency exchange entity in the virtual currency scenario, which is suitable for relevant evidence collection in the supervision process of relevant institutions in the virtual currency application scenario to ensure the security of cyberspace.

[0007] The present invention is achieved through the following technical solutions:

[0008] A method for identifying a virtual currency exchange name based on counterparty matching includes:

[0009] Construct a virtual currency address cluster dataset based on the historical transaction details of virtual currency;

[0010] Obtain all labels in the existing virtual currency transaction-related network, perform label processing on them, and construct a virtual currency address label dataset;

[0011] Using the virtual currency address label data set to expand the label address of the virtual currency address cluster data set;

[0012] Extracting transaction behavior feature data based on the historical transaction details of virtual currency, and training a label category recognition model based on the extracted feature data to obtain a supervised multi-classification label recognition model;

[0013] The supervised multi-classification label recognition model obtained through training is used to predict the category of address clusters with unknown labels and transaction times exceeding a threshold, that is, to obtain a suspected exchange address cluster;

[0014] Extracting the incoming counterparty address set of the suspected exchange address cluster;

[0015] According to the extracted inbound counterparty address set of the suspected exchange address cluster, the label name of the suspected exchange address cluster is identified.

[0016] Preferably, the step of constructing a virtual currency address cluster data set based on virtual currency historical transaction details of the present invention specifically includes:

[0017] Obtaining virtual currency historical transaction data and parsing and processing it to obtain virtual currency historical transaction details data;

[0018] Based on the virtual currency transaction details data obtained by analysis, the "multi-input" address clustering method is used to cluster virtual currency addresses to form a virtual currency address cluster data set;

[0019] The "multi-input" address clustering method is to cluster all input addresses in a multi-input transaction into an address cluster data set.

[0020] Preferably, the steps of obtaining all labels in the existing virtual currency transaction-related network, labeling them, and constructing a virtual currency address label data set specifically include:

[0021] By crawling all the tags in the WalletExplorer website, get the small tags of each tag address;

[0022] The obtained small labels are classified and labeled to obtain a virtual currency address label dataset.

[0023] Preferably, the step of using the virtual currency address label data set to expand the label address of the virtual currency address cluster data set in the present invention specifically includes:

[0024] The "multi-input" address clustering method is used to expand the virtual currency address cluster dataset with virtual currency label addresses to obtain the exchange address cluster dataset;

[0025] The exchange address cluster dataset is aggregated according to small tags to obtain the exchange platform address cluster dataset.

[0026] Preferably, the present invention extracts transaction behavior feature data based on virtual currency historical transaction details data, and trains a label category recognition model based on the extracted feature data to obtain a supervised multi-classification label recognition model, and the steps specifically include:

[0027] Obtain and integrate features from the virtual currency transaction details data from three aspects: address transaction behavior, address cluster transaction behavior, and address cluster network structure;

[0028] The integrated feature data is input into the preset supervised multi-classification algorithm to train the label category recognition model, obtain the supervised multi-classification label recognition model, and set corresponding probability thresholds for different categories.

[0029] Preferably, the step of identifying the label name of the suspected exchange address cluster according to the extracted inbound counterparty address set of the suspected exchange address cluster specifically includes:

[0030] Intersect the incoming counterparty address set of the suspected exchange address cluster with each known exchange platform address cluster to obtain the length of the address intersection;

[0031] Based on the length of the address intersection, the intersection proportion score is calculated;

[0032] Based on the intersection ratio score and the ratio score threshold, the label name of the suspected exchange address cluster can be identified.

[0033] Preferably, the intersection proportion score calculation formula of the present invention is as follows:

[0034] Score=Length(C) / Min(Length(A),Length(B))

[0035] Among them, Length(C) represents the intersection length, Length(A) represents the length of the incoming counterparty address set of the suspected exchange address cluster, Length(B) represents the length of the known exchange platform address cluster, and Min() represents the minimum value.

[0036] In a second aspect, the present invention proposes a virtual currency exchange name recognition device based on counterparty matching, comprising: a parsing processing unit, a labeling unit, an address expansion unit, a model training unit, a category prediction unit, an extraction unit and a recognition unit;

[0037] The analysis processing unit constructs a virtual currency address cluster data set based on the virtual currency historical transaction details data;

[0038] The labeling unit is used to obtain all labels in the existing virtual currency transaction related network, and perform labeling processing on them to construct a virtual currency address label data set;

[0039] The address expansion unit uses the virtual currency address label data set to perform label address expansion on the virtual currency address cluster data set;

[0040] The model training unit extracts features based on the historical transaction details of virtual currency, and trains a label category recognition model based on the extracted features to obtain a supervised multi-classification label recognition model;

[0041] The category prediction unit uses the trained supervised multi-classification label recognition model to predict the category of the address cluster with unknown labels and the number of transactions exceeding the threshold, that is, to obtain the suspected exchange address cluster;

[0042] The extraction unit is used to extract the set of incoming counterparty addresses of the suspected exchange address cluster;

[0043] The identification unit identifies the label name of the suspected exchange address cluster according to the extracted inbound counterparty address set of the suspected exchange address cluster.

[0044] In a third aspect, the present invention proposes a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method of the present invention when executing the computer program.

[0045] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the method of the present invention when executed by a processor.

[0046] The present invention has the following advantages and beneficial effects:

[0047] 1. The present invention can quickly expand the exchange label address based on the existing transaction label library through the virtual currency address clustering method. Compared with the manual label expansion method, the label expansion method based on the address clustering method can quickly, efficiently and accurately expand the exchange address label data set.

[0048] 2. The present invention expands the labels based on the method of collecting and clustering virtual currency addresses through open source channels, and uses a supervised multi-classification model to train the label recognition model to identify more suspected exchange address clusters.

[0049] 3. The present invention adopts a virtual currency exchange name identification method based on incoming counterparty matching to match and identify suspected exchange address clusters, and determine the name of the exchange platform to which the suspected exchange address clusters belong. The mode of combining exchange business rules with machine learning greatly improves the accuracy of the exchange label address data set. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0051] Figure 1 It is a schematic diagram of the identification method flow of the present invention.

[0052] Figure 2 Schematic diagram of the deposit and withdrawal process for a virtual currency exchange.

[0053] Figure 3 It is a schematic diagram of the computer device structure of the present invention.

[0054] Figure 4 This is a principle block diagram of the identification device of the present invention. DETAILED DESCRIPTION

[0055] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0056] Example 1

[0057] The existing information on virtual currency exchange wallet addresses collected through open source channels is too limited and cannot provide more accurate and reliable virtual currency exchange information to relevant institutions, resulting in the inability of relevant institutions to efficiently and comprehensively combat the scope of illegal activities using virtual currency. Therefore, this embodiment proposes a virtual currency exchange name identification method based on counterparty matching. This embodiment uses the method of matching incoming counterparties of virtual currency exchanges based on known virtual currency exchange information to quickly and effectively identify the exchange platform ownership of unknown virtual currency entities in specific virtual currency scenarios, thereby providing more reliable and accurate virtual currency transaction information to relevant institutions to ensure the security of cyberspace.

[0058] Specific as Figure 1 As shown, the identification method of this embodiment specifically includes the following steps:

[0059] Step 101, constructing an address cluster data set based on virtual currency historical transaction details data.

[0060] Step 101 of this embodiment specifically includes the following sub-steps:

[0061] Step 201, parse and process the historical transaction data of virtual currency to obtain transaction details such as transaction ID, transaction input address, transaction output address, transaction amount, transaction time, etc.; the parsing process of this embodiment is to build a virtual currency full node, extract and process the historical detailed data of virtual currency, so as to obtain transaction details such as transaction ID, transaction input address, transaction output address, transaction amount, transaction time, etc.

[0062] Step 202, based on the virtual currency transaction details data obtained through the analysis process, a "multi-input" address clustering method is used to cluster virtual currency addresses to form an address cluster (entity) data set.

[0063] The principle of the "multi-input" address clustering method of this embodiment is:

[0064] In the virtual currency transaction scenario, all input addresses in a virtual currency transaction are owned by the same user (entity); when the user's payment amount exceeds the amount of virtual currency in each available address in the user's wallet, in order to avoid the loss of transaction fees caused by executing multiple transactions to complete the payment, the user will select multiple virtual currency addresses from the wallet and aggregate them together for matching payment to achieve multi-input transactions. In addition, the funds in each address used in the virtual currency transaction need to be signed separately. We can conversely assume that all input addresses in a multi-input transaction come from the same user.

[0065] Step 102, obtain all tags in existing virtual currency transaction related websites, perform labeling processing on them, and construct a virtual currency address tag data set.

[0066] Step 102 of this embodiment is specifically as follows:

[0067] By crawling all the tags in the WalletExplorer website, get the small tags of each tag address, such as Huobi Exchange, Oken Exchange, Imtoken Wallet and other small tags;

[0068] The obtained small labels are classified and labeled, and finally a virtual currency address label dataset including major categories such as exchanges, service providers, coin mixers, dark webs, and blacklists is obtained.

[0069] Step 103: Use the virtual currency address label data set to expand the label address of the virtual currency address cluster data set.

[0070] Step 103 of this embodiment specifically includes the following sub-steps:

[0071] Step 301, using the principle of the "multi-input" address clustering method described in step 202 above, if a single address belongs to an exchange tag, then the address cluster to which the address belongs also belongs to the exchange tag. Based on this, the virtual currency address tag data set is used to expand the virtual currency address cluster data set with the virtual currency tag address to obtain the exchange address cluster data set;

[0072] Step 302 , aggregate the exchange address cluster data set according to the small tags to obtain the exchange platform address cluster data set.

[0073] Among them, according to the virtual currency exchange deposit and withdrawal process, such as Figure 2 As shown in the figure, in the process of deposit wallets converging to hot wallets, multiple deposit addresses often occur as inputs, so the deposit addresses of exchanges are aggregated into an address cluster; in the process of withdrawing coins, exchanges usually use hot wallet addresses as input addresses and withdrawal addresses as output addresses. However, due to the increase in virtual currency transaction fees, exchanges will also put deposit addresses and hot wallet addresses in the input direction to pay virtual currency to withdrawal addresses. Therefore, in the exchange address cluster, deposit addresses and hot wallets are usually included, among which the number of deposit addresses in the address cluster is much higher than that of hot wallets (deposit addresses are wallet addresses assigned to different users by exchanges, and private keys are controlled by exchanges).

[0074] Step 104 , extract features, and train a label category recognition model based on the extracted features to obtain a supervised multi-classification label recognition model.

[0075] Step 104 of this embodiment specifically includes:

[0076] According to the virtual currency transaction details data obtained by parsing in step 101, the transaction behavior feature data of the tag address is obtained: specifically, the features are obtained and integrated from the virtual currency transaction details data from three aspects: the address transaction behavior, the cluster transaction behavior to which the address belongs, and the cluster network structure to which the address belongs;

[0077] The integrated feature data is input into the preset supervised multi-classification algorithm to train the label category recognition model, obtain the supervised multi-classification label recognition model, and set corresponding probability thresholds for different categories.

[0078] In this embodiment, virtual currency transaction addresses are classified and category labels are added, thereby used for category identification (major category) of virtual currency transaction addresses with unknown labels.

[0079] Step 105, based on the supervised multi-classification label recognition model, the category of the address clusters with unknown labels and the number of transactions exceeding the threshold is predicted, that is, more address clusters suspected of being exchanges are obtained.

[0080] In this embodiment, by calculating the similarity or distance between the address cluster of the unknown tag and the address cluster of the known tag, the tag of the address cluster of the known tag with the highest similarity or the shortest distance is used as the tag of the address cluster of the unknown tag.

[0081] This embodiment uses a supervised multi-classification label recognition model to locate the address clusters of unknown labels in large categories (coarse positioning), thereby finding all exchange address clusters of the same category (i.e., suspected exchange address clusters), and further performing exchange name recognition (fine positioning).

[0082] Step 106, extracting the set of incoming counterparty addresses of the suspected exchange address cluster.

[0083] Step 107: Identify the label name of the suspected exchange address cluster according to the extracted inbound counterparty address set of the suspected exchange address cluster.

[0084] Step 107 of this embodiment intersects the incoming counterparty address set of each suspected exchange address cluster with each known exchange platform address cluster to obtain the address intersection length, and calculates the corresponding intersection percentage score. According to the intersection percentage score threshold, the known exchange name corresponding to the suspected exchange is returned. Specifically, it includes the following sub-steps:

[0085] Step 401, intersecting the incoming counterparty address set of each suspected exchange address cluster with each known exchange platform address set;

[0086] Step 402, the intersection proportion score of each known exchange platform is calculated by the following formula:

[0087] Score=Length(C) / Min(Length(A),Length(B))

[0088] Among them, Length(C) represents the intersection length, Length(A) represents the length of the incoming counterparty address set of the suspected exchange address cluster, Length(B) represents the length of the known exchange platform address cluster, and Min() represents the minimum value.

[0089] Step 403, compare whether the calculated intersection proportion score is greater than the proportion score threshold. If so, return the small label of the currently known exchange platform, i.e., Huobi Exchange, Oken Exchange, etc.; if not, identify the next suspected exchange address cluster.

[0090] The percentage score threshold in this embodiment can be set to 0.2, 0.5, 0.7, etc.

[0091] This embodiment also provides a computer device for executing the above method of this embodiment.

[0092] Specific as Figure 3 As shown, the computer device includes a processor, an internal memory and a system bus; various device components including the internal memory and the processor are connected to the system bus. The processor is a piece of hardware used to execute computer program instructions through basic arithmetic and logical operations in the computer system. The internal memory is a physical device used to temporarily or permanently store computer programs or data (for example, program status information). The system bus can be any of the following types of bus structures, including a memory bus or a storage controller, a peripheral bus, and a local bus. The processor and the internal memory can communicate data through the system bus. The internal memory includes a read-only memory (ROM) or a flash memory (not shown in the figure), and a random access memory (RAM), and RAM usually refers to the main memory loaded with the operating system and computer programs.

[0093] Computer devices generally include an external storage device. The external storage device can be selected from a variety of computer-readable media, which refers to any available media that can be accessed by the computer device, including both removable and fixed media. For example, computer-readable media include, but are not limited to, flash memory (micro SD card), CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic tape cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by the computer device.

[0094] Computer devices can be logically connected to one or more network terminals in a network environment. Network terminals can be personal computers, servers, routers, smart phones, tablets, or other public network nodes. Computer devices are connected to network terminals through network interfaces (local area network LAN interfaces). Local area network (LAN) refers to a network of computers interconnected within a limited area, such as a home, school, computer laboratory, or office building using network media. WiFi and twisted pair cabling Ethernet are the two most commonly used technologies for building LANs.

[0095] It should be noted that other computer systems including more or fewer subsystems than a computer device may also be suitable for the invention.

[0096] As described in detail above, the computer device applicable to the present embodiment can perform the specified operations of the multidimensional space horizon search method. The computer device performs these operations in the form of software instructions running in a computer-readable medium through a processor. These software instructions can be read into the memory from a storage device or from another device through a local area network interface. The software instructions stored in the memory enable the processor to execute the above-mentioned group member information processing method. In addition, the present invention can also be implemented by hardware circuits or hardware circuits combined with software instructions. Therefore, the implementation of this embodiment is not limited to any specific combination of hardware circuits and software.

[0097] Example 2

[0098] This embodiment proposes a virtual currency exchange name recognition device based on counterparty matching, specifically, Figure 4 As shown, the recognition device of this embodiment includes: a parsing processing unit, a labeling unit, an address expansion unit, a model training unit, a category prediction unit, an extraction unit and a recognition unit.

[0099] Among them, the parsing processing unit constructs a virtual currency address cluster data set based on the virtual currency historical transaction details data.

[0100] The labeling unit is used to obtain all labels in the existing virtual currency transaction-related network, perform label processing on them, and construct a virtual currency address label data set.

[0101] The address expansion unit uses the virtual currency address label data set to perform label address expansion on the virtual currency address cluster data set.

[0102] The model training unit extracts features based on the historical transaction details of virtual currency, and trains a label category recognition model based on the extracted features to obtain a supervised multi-classification label recognition model.

[0103] The category prediction unit uses the trained supervised multi-classification label recognition model to predict the category of address clusters with unknown labels and transaction times exceeding the threshold, that is, to obtain suspected exchange address clusters.

[0104] The extraction unit is used to extract the set of incoming counterparty addresses of the suspected exchange address cluster.

[0105] The identification unit identifies the label name of the suspected exchange address cluster according to the extracted inbound counterparty address set of the suspected exchange address cluster.

[0106] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying the name of a virtual currency exchange based on counterparty matching, characterized in that: include: Construct a virtual currency address cluster dataset based on the historical transaction details of virtual currency; Obtain all labels in the existing virtual currency transaction-related network, perform label processing on them, and construct a virtual currency address label dataset; Using the virtual currency address label data set to expand the label address of the virtual currency address cluster data set; Extracting transaction behavior feature data based on the historical transaction details of virtual currency, and training a label category recognition model based on the extracted feature data to obtain a supervised multi-classification label recognition model; The supervised multi-classification label recognition model obtained through training is used to predict the category of address clusters with unknown labels and transaction times exceeding a threshold, that is, to obtain a suspected exchange address cluster; Extracting the incoming counterparty address set of the suspected exchange address cluster; According to the extracted inbound counterparty address set of the suspected exchange address cluster, the label name of the suspected exchange address cluster is identified, specifically including: Intersect the incoming counterparty address set of the suspected exchange address cluster with each known exchange platform address cluster to obtain the length of the address intersection; Based on the length of the address intersection, the intersection proportion score is calculated; Based on the intersection ratio score and the ratio score threshold, the label name of the suspected exchange address cluster can be identified.

2. The method for identifying the name of a virtual currency exchange based on counterparty matching according to claim 1, characterized in that: The steps for constructing a virtual currency address cluster dataset based on the historical transaction details of virtual currency include: Obtaining virtual currency historical transaction data and parsing and processing it to obtain virtual currency historical transaction details; Based on the virtual currency transaction details data obtained by parsing, the "multi-input" address clustering method is used to cluster virtual currency addresses to form a virtual currency address cluster data set; The "multi-input" address clustering method is to cluster all input addresses in a multi-input transaction into an address cluster data set.

3. The method for identifying the name of a virtual currency exchange based on counterparty matching according to claim 1, characterized in that: The steps of obtaining all labels in the existing virtual currency transaction-related network and labeling them to construct a virtual currency address label data set include: By crawling all the tags in the WalletExplorer website, get the small tags of each tag address; The obtained small labels are classified and labeled to obtain a virtual currency address label dataset.

4. The method for identifying the name of a virtual currency exchange based on counterparty matching according to claim 1, characterized in that: The step of using the virtual currency address label data set to expand the label address of the virtual currency address cluster data set specifically includes: The "multi-input" address clustering method is used to expand the virtual currency address cluster dataset with virtual currency label addresses to obtain the exchange address cluster dataset; The exchange address cluster dataset is aggregated according to small tags to obtain the exchange platform address cluster dataset.

5. The method for identifying the name of a virtual currency exchange based on counterparty matching according to claim 1, characterized in that: The transaction behavior feature data is extracted based on the historical transaction details of virtual currency, and the label category recognition model is trained based on the extracted feature data to obtain the supervised multi-classification label recognition model. The specific steps include: Obtain and integrate features from the virtual currency transaction details data from three aspects: address transaction behavior, address cluster transaction behavior, and address cluster network structure; The integrated feature data is input into the preset supervised multi-classification algorithm to train the label category recognition model, obtain the supervised multi-classification label recognition model, and set corresponding probability thresholds for different categories.

6. The method for identifying the name of a virtual currency exchange based on counterparty matching according to claim 1, characterized in that: The intersection proportion score calculation formula is as follows: Score = Length(C) / Min(Length(A), Length(B)) Among them, Length(C) represents the intersection length, Length(A) represents the length of the incoming counterparty address set of the suspected exchange address cluster, Length(B) represents the length of the known exchange platform address cluster, and Min() represents taking the minimum value.

7. A virtual currency exchange name identification device based on counterparty matching, characterized in that: include: Parsing unit, labeling unit, address expansion unit, model training unit, category prediction unit, extraction unit and recognition unit; The analysis processing unit constructs a virtual currency address cluster data set based on the virtual currency historical transaction details data; The labeling unit is used to obtain all labels in the existing virtual currency transaction related network, and perform labeling processing on them to construct a virtual currency address label data set; The address expansion unit uses the virtual currency address label data set to perform label address expansion on the virtual currency address cluster data set; The model training unit extracts features based on the historical transaction details of virtual currency, and trains a label category recognition model based on the extracted features to obtain a supervised multi-classification label recognition model; The category prediction unit uses the trained supervised multi-classification label recognition model to predict the category of the address cluster with unknown labels and the number of transactions exceeding the threshold, that is, to obtain the suspected exchange address cluster; The extraction unit is used to extract the set of incoming counterparty addresses of the suspected exchange address cluster; The identification unit identifies the label name of the suspected exchange address cluster according to the extracted inbound counterparty address set of the suspected exchange address cluster, specifically including: Intersect the incoming counterparty address set of the suspected exchange address cluster with each known exchange platform address cluster to obtain the length of the address intersection; Based on the length of the address intersection, the intersection proportion score is calculated; Based on the intersection ratio score and the ratio score threshold, the label name of the suspected exchange address cluster can be identified.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images