Method and related equipment for discovering express delivery and receiving organizations based on relationship mining

Abstract

The present disclosure provides a method, device, electronic device and storage medium for discovering express delivery and receiving organizations based on relationship mining. The method includes: obtaining customs declaration data including recipient addresses and recipients; forming address clusters according to the similarity between the recipient addresses; using the recipients and recipient addresses in the address clusters as nodes, and using the relationship between recipients, the relationship between recipient addresses, and the relationship between recipients and recipient addresses as edges to construct a graph; processing the nodes and edges in the graph according to the type of recipient addresses in the graph to obtain a connected subgraph; and determining the express delivery and receiving organization relationship network according to the connected subgraph. The method can quickly and accurately discover the express delivery and receiving organization relationship network.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a method, device, electronic device and storage medium for discovering express delivery organizations based on relationship mining. Background Art

[0002] With the development of international logistics, the number of inbound express shipments has increased rapidly. How to efficiently and accurately supervise and inspect the massive amount of express shipments is a major challenge facing the customs. In the process of customs supervision, how to identify the organized behavior of "breaking the whole into pieces" and "ant moving" and how to uncover the purchasing gangs is an important task.

[0003] In the business scenario of customs express inspection, the definition and construction of relationships involve complex business scenarios, and it is necessary to mine the contents of the customs declaration form to build relationships. The business scenario construction methods in related technologies are not applicable; moreover, in the process of discovering communities or organizations in the network, the clustering algorithms in related technologies are highly complex and cannot be effectively applied in actual business.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The purpose of the present disclosure is to provide a method, device, electronic device and storage medium for discovering express delivery receiving and sending organizations based on relationship mining, which can quickly and accurately discover the relationship network of express delivery receiving and sending organizations.

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

[0007] The disclosed embodiment provides a method for discovering express delivery receiving and sending organizations based on relationship mining, comprising: obtaining customs declaration data including recipient addresses and recipients; forming address clusters according to the similarities between the recipient addresses; taking the recipients and recipient addresses in the address clusters as nodes, and taking the relationships between recipients, the relationships between recipient addresses, and the relationships between recipients and recipient addresses as edges to construct a graph; processing the nodes and edges in the graph according to the types of recipient addresses in the graph to obtain a connected subgraph; and determining a relationship network of express delivery receiving and sending organizations according to the connected subgraph.

[0008] In some exemplary embodiments of the present disclosure, the above method also includes: cleaning the recipient address to obtain the recipient address information of the recipient address; segmenting the recipient address information to obtain the segmentation result of the recipient address; and determining the similarity between the recipient addresses based on the recipient address information and the segmentation result of the recipient address.

[0009] In some exemplary embodiments of the present disclosure, a recipient address is cleaned to obtain recipient address information, including: deleting preset characters in the recipient address, and converting the character format in the recipient address so that the character format in the recipient address is the same; obtaining the province information in the recipient address information according to the maximum forward matching method; if the recipient address information does not contain the province information, obtaining the city information in the recipient address information according to the maximum forward matching method, and determining the province information according to the city information; if the recipient address information does not contain the city information, obtaining the district information in the recipient address information according to the maximum forward matching method, and determining the city information according to the province information and the district information; if the recipient address information does not contain the district information, obtaining the town information in the recipient address information according to the maximum forward matching method, and determining the district information according to the province information, the city information and the town information.

[0010] In some exemplary embodiments of the present disclosure, the similarity between recipient addresses is determined based on recipient address information and word segmentation results of the recipient addresses, including: determining the similarity between the recipient address information of the recipient addresses; determining the similarity between each part in the word segmentation results of the recipient addresses; determining the similarity between the word segmentation results of the recipient addresses based on the similarity between each part in the word segmentation results; taking a weighted sum of the similarity between the recipient address information of the recipient addresses and the similarity between the word segmentation results of the recipient addresses to obtain the similarity between the recipient addresses.

[0011] In some exemplary embodiments of the present disclosure, the nodes and edges in the graph are processed according to the type of the recipient address in the graph to obtain a connected subgraph, including: if the recipient address in the connected subgraph is a public area, the target recipient connected to the recipient address is determined, and the nodes and the edges connected to the target recipients whose degrees among the target recipients are a preset constant are deleted to obtain a connected subgraph; if the recipient address in the connected subgraph is a non-public area, the target recipient connected to the recipient address is determined, and if there is no directly connected edge between the target recipients, and the target recipients are connected through the nodes corresponding to the recipient addresses, the edges between the recipient addresses and the target recipients are deleted to obtain a connected subgraph.

[0012] In some exemplary embodiments of the present disclosure, it is characterized in that determining the express delivery sending and receiving organization relationship network based on the connected subgraph includes: determining the connected subgraph with a total number of nodes greater than 2 in the connected subgraph as the express delivery sending and receiving organization relationship network.

[0013] In some exemplary embodiments of the present disclosure, the above method also includes: using the number of occurrences of customs declarations with the same consignee and recipient address as the edge weight of the corresponding edge between the consignee and the recipient address; using the number of occurrences of customs declarations with the same preset fields as the edge weight of the corresponding edge between the corresponding consignees; using the number of occurrences of customs declarations with the same preset fields as the edge weight of the corresponding edge between the corresponding recipients; wherein the preset fields include at least one of the consignee's phone number, sender, sender's city, waybill number, item list, and item value.

[0014] In some exemplary embodiments of the present disclosure, the above method also includes: determining the clustering coefficient and average clustering coefficient of each node in the sending and receiving organization relationship network according to the edges and edge weights corresponding to each node in the sending and receiving organization relationship network; if the average clustering coefficient of the sending and receiving organization relationship network is greater than a first average clustering coefficient threshold, the number of recipients in the sending and receiving organization relationship network is greater than a first number, the proportion of nodes in the sending and receiving organization relationship network whose clustering coefficient is the clustering coefficient threshold is less than a preset proportion, the types of items in the sending and receiving organization relationship network are greater than a second number, the frequency of item declarations in the sending and receiving organization relationship network is greater than a third number, and the average number of purchases in the sending and receiving organization relationship network is greater than a fourth number, then the risk level of the sending and receiving organization relationship network is determined to be the first risk level.

[0015] In some exemplary embodiments of the present disclosure, the above method also includes: normalizing the declaration frequency of the sending and receiving organization relationship network, the average declared value of the sending and receiving organization relationship network, the average purchase frequency of each item in the sending and receiving organization relationship network, and the item types of each recipient in the sending and receiving organization relationship network to obtain the normalized results of the declaration frequency, the normalized results of the average declared value, the normalized results of the average purchase frequency, and the normalized results of the item types; weighting and adding the normalized results of the declaration frequency, the normalized results of the average declared value, the normalized results of the average purchase frequency, the normalized results of the item types, and the average clustering coefficient to obtain the risk value of the sending and receiving organization relationship network; and sorting the sending and receiving organization relationship network according to the risk value of the sending and receiving organization relationship network.

[0016] In some exemplary embodiments of the present disclosure, the above method also includes: if the average clustering coefficient of the sending and receiving organization relationship network is greater than the second clustering coefficient threshold, the number of edges of the sending and receiving organization relationship network is greater than the fifth number, and the item declaration frequency of the sending and receiving organization relationship network is greater than the sixth number, then the risk level of the sending and receiving organization relationship network is determined to be the second risk level.

[0017] In some exemplary embodiments of the present disclosure, the above method also includes: if the number of edges of the sending and receiving organization relationship network is greater than the seventh number, and the number of nodes corresponding to the recipients of the sending and receiving organization relationship network is greater than the eighth number, then the risk level of the sending and receiving organization relationship network is determined to be the third risk level.

[0018] In some exemplary embodiments of the present disclosure, the above method also includes: if the risk level of the sending and receiving organization relationship network is determined to be the first risk level, and the number of items in the sending and receiving organization relationship network is greater than the item number threshold or the number of item types in the sending and receiving organization relationship network is greater than the item type number threshold, then the use of items corresponding to the receiving address in the sending and receiving organization relationship network is determined to be commercial use; if the risk level of the sending and receiving organization relationship network is determined to be the second risk level, and the number of items in the sending and receiving organization relationship network is greater than the item number threshold or the number of item types in the sending and receiving organization relationship network is greater than the item type number threshold and the total value of the sending and receiving organization relationship network is greater than or equal to the preset value, then the use of items corresponding to the receiving address in the sending and receiving organization relationship network is determined to be commercial use; if the risk level of the sending and receiving organization relationship network is determined to be the third risk level, and the number of items in the sending and receiving organization relationship network is greater than the item number threshold or the number of item types in the sending and receiving organization relationship network is greater than the item type number threshold, the total value of the sending and receiving organization relationship network is greater than or equal to the preset value and the receiving frequency of the sending and receiving organization relationship network is greater than the frequency threshold, then the use of items corresponding to the receiving address in the sending and receiving organization relationship network is determined to be commercial use.

[0019] The disclosed embodiment provides a device for discovering express delivery receiving and sending organizations based on relationship mining, comprising: a data acquisition module for acquiring customs declaration data including a recipient address and a recipient; an address cluster formation module for forming an address cluster according to the similarity between the recipient addresses; a graph construction module for constructing a graph using the recipients and recipient addresses in the address cluster as nodes, and using the relationship between the recipients, the relationship between the recipient addresses, and the relationship between the recipients and the recipient addresses as edges; a connected subgraph acquisition module for processing the nodes and edges in the graph according to the type of the recipient addresses in the graph to obtain a connected subgraph; and a relationship network determination module for determining the relationship network of the express delivery receiving and sending organizations according to the connected subgraph.

[0020] An embodiment of the present disclosure provides an electronic device, comprising: at least one processor; a storage device for storing at least one program, and when the at least one program is executed by the at least one processor, the at least one processor implements any of the above-mentioned methods for discovering express delivery sending and receiving organizations based on relationship mining.

[0021] An embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any of the above-mentioned methods for discovering express delivery sending and receiving organizations based on relationship mining.

[0022] Some embodiments of the present disclosure provide a method for discovering express delivery receiving and sending organizations based on relationship mining. Address clusters are formed according to the similarities between receiving addresses in customs declaration data, so that closely connected address clusters can be obtained. Recipients and receiving addresses in the address clusters are taken as nodes, and the relationships between nodes are taken as edges. A graph based on recipients and receiving addresses in customs declaration data can be constructed. The nodes and edges in the graph are processed to obtain a connected subgraph. The relationship network of express delivery receiving and sending organizations is determined based on the connected subgraph. The relationship network of express delivery receiving and sending organizations can be quickly and accurately discovered, which facilitates efficient and accurate supervision and inspection of express deliveries.

[0023] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0025] Figure 1 A schematic diagram showing an exemplary system architecture of the express delivery organization discovery method based on relationship mining to which the embodiments of the present disclosure can be applied.

[0026] Figure 2 The present invention is a flowchart showing a method for discovering express delivery and receiving organizations based on relationship mining according to an exemplary embodiment.

[0027] Figure 3 is a schematic diagram showing a connected subgraph according to an exemplary embodiment.

[0028] Figure 4 The present invention is a flowchart showing another method for discovering express delivery organization based on relationship mining according to an exemplary embodiment.

[0029] Figure 5 The present invention is a flowchart showing another method for discovering express delivery organization based on relationship mining according to an exemplary embodiment.

[0030] Figure 6 The present invention is a flow chart showing a method of cleaning a delivery address according to an exemplary embodiment.

[0031] Figure 7 The present invention is a flowchart showing another method for discovering express delivery organization based on relationship mining according to an exemplary embodiment.

[0032] Figure 8The present invention is a flowchart showing a method for calculating the similarity of a receiving address according to an exemplary embodiment.

[0033] Fig. 9 The present invention is a flowchart showing another method for discovering express delivery organization based on relationship mining according to an exemplary embodiment.

[0034] Fig.10 It is a block diagram of a device for discovering express delivery and receiving organizations based on relationship mining according to an exemplary embodiment.

[0035] Fig.11 A schematic structural diagram of an electronic device is shown according to an exemplary embodiment. DETAILED DESCRIPTION

[0036] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0037] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0038] Figure 1 A schematic diagram showing an exemplary system architecture of the express delivery organization discovery method based on relationship mining to which the embodiments of the present disclosure can be applied.

[0039] like Figure 1 As shown, the system architecture may include a server 101, a network 102 and a terminal device 103. The network 102 is used to provide a medium for a communication link between the terminal device 103 and the server 101. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables.

[0040] The server 101 may be a server that provides various services, such as a background management server that provides support for devices operated by users using the terminal device 103. The background management server may analyze and process the received request data, and feed back the processing results to the terminal device.

[0041] Server 101 can, for example, obtain customs declaration data including recipient addresses and recipients; form address clusters based on the similarities between recipient addresses; server 101 can, for example, use recipients and recipient addresses in the address cluster as nodes, and use the relationships between recipients, the relationships between recipient addresses, and the relationships between recipients and recipient addresses as edges to construct a graph; server 101 can, for example, process the nodes and edges in the graph based on the types of recipient addresses in the graph to obtain a connected subgraph; server 101 can, for example, determine the relationship network of express delivery sending and receiving organizations based on the connected subgraph.

[0042] The server 101 can, for example, display the express delivery sending and receiving organization relationship network, and can also send the express delivery sending and receiving organization relationship network to the terminal device 103, and the terminal device 103 can be used to display the express delivery sending and receiving organization relationship network.

[0043] It should be understood that Figure 1 The number of terminal devices, networks and servers is merely illustrative. Server 101 may be a physical server, a server cluster consisting of multiple servers, or a cloud server. It may have any number of terminal devices, networks and servers according to actual needs.

[0044] Below, each step of the method for discovering express delivery and receiving organizations based on relationship mining in the exemplary embodiment of the present disclosure will be described in more detail with reference to the accompanying drawings and embodiments.

[0045] Figure 2 is a flowchart of a method for discovering a delivery organization based on relationship mining according to an exemplary embodiment. The method provided by the embodiment of the present disclosure can be performed as follows: Figure 1 The process is executed in the server shown in the figure, but the present disclosure is not limited to this.

[0046] like Figure 2 As shown, the method for discovering express delivery and receiving organizations based on relationship mining provided by the embodiment of the present disclosure may include the following steps.

[0047] In step S201, customs declaration data including the consignee address and consignee is obtained.

[0048] In the embodiment of the present disclosure, the customs declaration form data may be imported customs declaration form data of Class B, and Class B customs declaration forms are items that are exempted from duty within the limits specified by current customs regulations. The customs declaration form data may include the customs declaration form number, bill of lading number, sender, sender address, consignee, consignee address, consignee phone number, declared items list, declared items total value, declared items unit price, declared items number, declared items type, consignee ID number, etc., but the present disclosure is not limited thereto.

[0049] In step S202, address clusters are formed according to the similarities between the recipient addresses.

[0050] In the disclosed embodiment, a similarity threshold can be preset as needed, and the customs declaration data corresponding to the recipient addresses whose similarity between the recipient addresses is greater than the similarity threshold can be regarded as an address cluster. The customs declaration data corresponding to the recipient addresses with close recipient addresses can be regarded as an address cluster, for example, the customs declaration data of the recipient addresses in the same community can be regarded as an address cluster.

[0051] In step S203, the recipients and recipient addresses in the address cluster are taken as nodes, and the relationship between the recipients, the relationship between the recipient addresses, and the relationship between the recipients and the recipient addresses are taken as edges to construct a graph.

[0052] In the embodiment of the present disclosure, for each address cluster, the nodes constituting the graph in each address cluster may be extracted, and relationship edges may be generated according to rules to convert each address cluster into a graph.

[0053] In the disclosed embodiment, there may be one or more recipients in an address cluster, and there may also be one or more recipient addresses. The recipients and recipient addresses in the address cluster may be used as nodes of a graph, the relationship between two recipients may be used as an edge between the nodes corresponding to the two recipients, the relationship between two recipient addresses may be used as an edge between the nodes corresponding to the two recipient addresses, and the relationship between a recipient and a recipient address may be used as an edge between the node corresponding to the recipient and the node corresponding to the recipient address.

[0054] In step S204, the nodes and edges in the graph are processed according to the types of the recipient addresses in the graph to obtain a connected subgraph.

[0055] In the embodiment of the present disclosure, the influence of public areas can be considered, and the nodes and edges in the graph are processed differently according to whether the receiving address in the graph belongs to the public area to obtain a connected subgraph. The public area can be a hospital, a school, etc., but the present disclosure is not limited thereto.

[0056] The nodes and edges in the graph are processed. For example, the nodes and edges with weaker relationships in the graph can be deleted to obtain a connected subgraph.

[0057] Figure 3 is a schematic diagram showing a connected subgraph according to an exemplary embodiment.

[0058] like Figure 3 As shown, the nodes of the connected subgraph include recipient A, recipient B, recipient C, recipient D, recipient E, recipient F, recipient G, recipient H, recipient address 1 and recipient address 2.

[0059] In an exemplary embodiment, if the recipient address in the connected subgraph is a public area, the target recipient connected to the recipient address is determined, and the nodes corresponding to the target recipients whose degrees among the target recipients are preset constants and the edges connected thereto are deleted to obtain a connected subgraph.

[0060] In the embodiment of the present disclosure, the preset constant may be set to 1, and the degree may represent the number of edges of the node.

[0061] by Figure 3 Take the recipient address 1 in the example as an example. If the recipient address 1 is a public area, the target recipient connected to the recipient address 1 is the recipient D. The recipient D can be analyzed. Figure 3 It can be seen that the degree of the node corresponding to recipient D is 6, that is, there are 6 edges connected to the node corresponding to recipient D, so the node corresponding to recipient D and its connected edges are retained; if the degree of the node corresponding to recipient D is 1, that is, there is only one edge connected to the node corresponding to recipient D, then the node corresponding to recipient D and its connected edges are deleted.

[0062] In an exemplary embodiment, if the recipient address in the connected subgraph is a non-public area, the target recipient connected to the recipient address is determined. If there is no directly connected edge between the target recipients, and the target recipients are connected through the nodes corresponding to the recipient address, the edge between the recipient address and the target recipient is deleted to obtain a connected subgraph.

[0063] by Figure 3 Take the recipient address 2 in as an example to illustrate. If the recipient address 2 is not a public area, the target recipients connected to the recipient address 2 are recipient E, recipient F, recipient G, recipient H, recipient A, recipient B, and recipient C. Take the relationship between recipient E and recipient F as an example to illustrate. If there is no direct edge between recipient E and recipient F, and recipient E and recipient F are only connected through the node corresponding to recipient address 2, then the edge between recipient E and recipient address 2 can be deleted, and the edge between recipient F and recipient address 2 can be deleted.

[0064] In step S205, the express delivery sending and receiving organization relationship network is determined based on the connected subgraph.

[0065] In an exemplary embodiment, a connected subgraph in which the total number of nodes is greater than 2 may be determined as a relationship network of express delivery sending and receiving organizations.

[0066] In the embodiment of the present disclosure, a connected subgraph with a total number of nodes less than or equal to 2 can be deleted, and a connected subgraph with a total number of nodes greater than 2 can be determined as a sender-receiver organization relationship network, which can be regarded as a potential purchasing agent gang.

[0067] The method for discovering express delivery receiving and sending organizations based on relationship mining provided by the embodiments of the present disclosure forms address clusters according to the similarities between the receiving addresses in the customs declaration data, and can obtain closely connected address clusters; the recipients and receiving addresses in the address clusters are taken as nodes, and the relationships between the nodes are taken as edges, so as to construct a graph based on the recipients and receiving addresses in the customs declaration data, and the nodes and edges in the graph are processed to obtain a connected subgraph, and the relationship network of express delivery receiving and sending organizations is determined according to the connected subgraph, so that the relationship network of express delivery receiving and sending organizations can be quickly and accurately discovered, which is convenient for efficient and accurate supervision and inspection of express deliveries.

[0068] Figure 4 The present invention is a flowchart showing another method for discovering express delivery organization based on relationship mining according to an exemplary embodiment.

[0069] In the embodiment of the present disclosure, Figure 2 Based on the relationship mining-based express delivery organization discovery method shown in the figure, Figure 4 The method for discovering express delivery organizations based on relationship mining shown may also include the following steps.

[0070] In step S401, the recipient address is cleaned to obtain the recipient address information of the recipient address.

[0071] In the disclosed embodiment, the recipient address can be cleaned, non-standard characters in the recipient address can be deleted, characters with non-uniform formats in the recipient address can be converted, the postal code, mobile phone number, etc. in the recipient address can be deleted, and the province, city, district, and town information in the recipient address can be matched in sequence using the maximum forward matching method.

[0072] In step S402, the recipient address information is segmented to obtain a segmentation result of the recipient address.

[0073] In the disclosed embodiment, the cleaned recipient address information can be segmented, and multiple segmentation results can be obtained, such as province, city, district, town, region, road, community, factory (building, industrial park, etc.), school, other units, building number and house number, and other unknown information.

[0074] In step S403, the similarity between the recipient addresses is determined based on the recipient address information and the word segmentation result of the recipient addresses.

[0075] Taking recipient address A and recipient address B as an example, recipient address A is cleaned to obtain recipient address information of recipient address A, recipient address information of recipient address A is segmented to obtain segmentation results of recipient address A, recipient address B is cleaned to obtain recipient address information of recipient address B, recipient address information of recipient address B is segmented to obtain segmentation results of recipient address B. The similarity between recipient address information of recipient address A and recipient address information of recipient address B can be calculated, the similarity between segmentation results of recipient address A and recipient address B can be calculated, and the similarity between recipient addresses and segmentation results can be weighted and summed to obtain the similarity between recipient address A and recipient address B.

[0076] In the disclosed embodiments, regular expressions can be used to extract the quantity of items from the item names in the customs declaration data, and the quantity of items measured in sheets (such as sheet-packed facial masks) and the quantity of items with a unit price lower than 10 can be corrected. For example, the original quantity can be divided by 10 to obtain the corrected quantity of items, and the corrected quantity of items, total price of items, and unit price of items can be merged into the corresponding customs declaration data forming the address cluster through the customs declaration number.

[0077] Figure 5 The present invention is a flowchart showing another method for discovering express delivery organization based on relationship mining according to an exemplary embodiment.

[0078] In the disclosed embodiment, Figure 4 The difference between the express delivery organization discovery method based on relationship mining shown in the figure is that Figure 5 The method for discovering express delivery and receiving organizations based on relationship mining shown further provides how to clean the receiving address and obtain the receiving address information, that is, provides an embodiment of the above step S401.

[0079] like Figure 5 As shown, step S401 may include the following steps.

[0080] In step S4011, the preset characters in the recipient address are deleted, and the character formats in the recipient address are converted so that the character formats in the recipient address are the same.

[0081] In the disclosed embodiments, the receiving address can be cleaned to remove non-standard characters in the receiving address, such as " / ", space, etc. The characters in Chinese and English can be unified, and the Chinese characters can be converted into English characters, such as converting the Chinese '()' into the English '()'. The uppercase and lowercase letters can be unified, converting uppercase letters into lowercase letters, or converting lowercase letters into uppercase letters. The postal code, mobile phone number, etc. in the receiving address can be deleted.

[0082] In step S4012, the province information in the recipient address information is obtained according to the maximum forward matching method.

[0083] Figure 6 The present invention is a flow chart showing a method of cleaning a delivery address according to an exemplary embodiment.

[0084] In the embodiments of the present disclosure, reference Figure 6 , the province information in the recipient address information can be obtained by matching according to the maximum forward matching method, if the recipient address contains the province information, or, if the recipient address does not contain the province information, the city information in the recipient address information can be obtained by matching according to the maximum forward matching method.

[0085] In step S4013, if the recipient address information does not contain the province information, the city information in the recipient address information is obtained according to the maximum forward matching method, and the province information is determined according to the city information.

[0086] In the embodiments of the present disclosure, reference Figure 6 If the recipient address does not contain the province information, the city information in the recipient address information is matched according to the maximum forward matching method, and the province information is determined and supplemented according to the obtained city information. For example, if the city information in the recipient address information is Shijiazhuang, it can be determined that the province information should be Hebei Province, and the province information is supplemented into the recipient address information.

[0087] In the embodiments of the present disclosure, reference Figure 6 If the recipient address contains both province information and city information, continue to match the city information in the recipient address using the maximum forward matching method.

[0088] In step S4014, if the recipient address information does not contain city information, the district information in the recipient address information is obtained according to the maximum forward matching method, and the city information is determined according to the province information and the district information.

[0089] In the embodiments of the present disclosure, reference Figure 6 If the receiving address does not contain the city information, the district information in the receiving address information is matched according to the maximum forward matching method, and the city information is determined and supplemented according to the obtained district information. For example, if the district information in the received address information is Luquan District, it can be determined that the city information should be Shijiazhuang City, and the city information is supplemented into the receiving address information.

[0090] In the embodiments of the present disclosure, reference Figure 6 If the recipient address contains province information but not city information, the district information in the recipient address information can be obtained according to the maximum forward matching method, and the city information can be determined according to the province information and district information.

[0091] In the embodiments of the present disclosure, reference Figure 6 If the recipient address contains province information, city information and district information, continue to match the town information in the recipient address according to the maximum forward matching method.

[0092] In step S4015, if the receiving address information does not contain district information, the town information in the receiving address information is obtained according to the maximum forward matching method, and the district information is determined according to the province information, city information and town information.

[0093] In the embodiments of the present disclosure, reference Figure 6 If the receiving address does not contain district information, the town information in the receiving address information is matched according to the maximum forward matching method, and the district information is determined and supplemented according to the obtained town information. For example, if the town information in the receiving address information is Sijiazhuang Town, it can be determined that the district information should be Luquan District, and the district information is supplemented into the receiving address information.

[0094] In the embodiments of the present disclosure, reference Figure 6 If the recipient address contains province information and city information but does not contain district information, the town information in the recipient address information can be obtained according to the maximum forward matching method, and the city information can be determined according to the province information, city information and town information. In the embodiment of the present disclosure, after the recipient address is cleaned, the recipient address information can be obtained, and the recipient address information can include province information, city information, district information, town information and remaining address information.

[0095] In the disclosed embodiment, by matching the next level of recipient address information to determine and supplement the previous level of recipient address information, complete and accurate recipient address information can be obtained.

[0096] Figure 7 The present invention is a flowchart showing another method for discovering express delivery organization based on relationship mining according to an exemplary embodiment.

[0097] In the present disclosure, Figure 4 The difference between the express delivery organization discovery method based on relationship mining shown in the figure is that Figure 7 The method for discovering express delivery organizations based on relationship mining shown further provides how to determine the similarity between recipient addresses based on the recipient address information and word segmentation results of the recipient addresses, that is, provides an embodiment of the above step S403.

[0098] like Figure 7 As shown, step S403 may include the following steps.

[0099] In step S4031, the similarity between the recipient address information of the recipient addresses is determined.

[0100] Figure 8The present invention is a flowchart showing a method for calculating the similarity of a receiving address according to an exemplary embodiment.

[0101] In the embodiments of the present disclosure, reference Figure 8 , taking the calculation of the similarity between the recipient address A and the recipient address B as an example, after obtaining the recipient address information of the recipient address A and the recipient address information of the recipient address B, the similarity between the recipient address information of the recipient address A and the recipient address information of the recipient address B can be calculated according to the Jaro-Winkler distance algorithm.

[0102] It should be noted that in the embodiments of the present disclosure, the similarity of the recipient addresses may also be calculated according to other similarity calculation methods, and the present disclosure does not limit this.

[0103] In step S4032, the similarity between the parts in the word segmentation result of the recipient address is determined.

[0104] In the embodiments of the present disclosure, reference Figure 8 , still taking the calculation of the similarity between the recipient address A and the recipient address B as an example, after obtaining the word segmentation results of the recipient address A and the word segmentation results of the recipient address B, the similarities between each part of the word segmentation result of the recipient address A and each part of the word segmentation result of the recipient address B can be calculated respectively according to the Jaro-Winkler distance algorithm.

[0105] In step S4033, the similarity between the word segmentation results of the recipient addresses is determined based on the similarity between the parts in the word segmentation results.

[0106] In the embodiments of the present disclosure, reference Figure 8 , the similarities between the parts of the word segmentation results can be weighted and summed to obtain the similarities between the word segmentation results of the recipient address.

[0107] In step S4034, the similarity between the recipient address information and the similarity between the word segmentation results of the recipient addresses are weighted and summed to obtain the similarity between the recipient addresses.

[0108] In the embodiment of the present disclosure, the similarity between the recipient addresses can be determined according to the following formula:

[0109]

[0110] Where similarity indicates the similarity between the recipient addresses. all Represents the similarity between the recipient address information, α represents the weighted harmonic coefficient less than 1, similarityp art_i Indicates the similarity of the string in the i-th part of the word segmentation result, where i is an integer greater than or equal to 1. Indicates the weight coefficient of each part in the word segmentation result.

[0111] In the disclosed embodiment, after determining the similarity between the receiving addresses, similar receiving addresses can be replaced. For example, "Room 210, 3rd floor, 4th floor" and "Room 4210, 3rd floor" are actually one receiving address. Based on the similarity between the receiving addresses, similar receiving addresses can be replaced. For example, "Room 210, 3rd floor, 4th floor" can be replaced with "Room 210, 3rd floor, 4th floor" to unify similar receiving addresses into the same receiving address.

[0112] In the embodiment of the present disclosure, the recipient address may be processed according to the following situations:

[0113] (1) If all characters in the two recipient addresses are the same, the similarity result returned is 1.

[0114] (2) If the similarity between the two recipient addresses is greater than 0.9, and the last three characters of the character strings of the two recipient addresses are the same, the last three characters of the two recipient addresses are judged. If the last three characters of the two recipient addresses do not contain letters or numbers, the similarity result is returned as 1; if the last three characters of the two recipient addresses contain letters or numbers, the building numbers and house numbers of the two recipient addresses are digitally extracted, such as Room 210 on the 4th floor of the 3rd floor. The extracted result is 34210. If the extracted results of the two recipient addresses are the same, the similarity result is returned as 1.

[0115] (3) If the two delivery addresses end with the word "school" or "factory" in the word segmentation results, and the word segmentation results of the school or "factory" in the two delivery addresses are the same and not empty, and the similarity of the two delivery addresses is greater than 0.8, then the similarity result returned is 1.

[0116] (4) If the similarity result is greater than or equal to 0.9, it means that the two addresses are very close. Determine whether the building number and house number in the two addresses are the same. When using regular matching to extract the building number and house number in the address, firstly, the road and road number must be matched regularly, and then the road and road number must be deleted to eliminate the influence of the road number in the address. If the extracted results are not null and are the same, the returned similarity is 1; if the results are not null and are not the same, the returned similarity is 0; otherwise, the similarity value calculated by the algorithm is returned.

[0117] (5) If the recipient address cannot be matched in the previous four cases, the similarity calculated according to the above method is returned.

[0118] After processing the recipient addresses by the above method, the similarity threshold can be set to 0.97. If the similarity is greater than or equal to 0.97, the two recipient addresses are unified as the same recipient address.

[0119] Fig. 9 The present invention is a flowchart showing another method for discovering express delivery organization based on relationship mining according to an exemplary embodiment.

[0120] In the embodiment of the present disclosure, Figure 2 Based on the relationship mining-based express delivery organization discovery method shown in the figure, Fig. 9 The method for discovering express delivery organizations based on relationship mining shown may also include the following steps.

[0121] In step S901, the number of occurrences of the customs declaration forms with the same consignee and consignment address is used as the edge weight of the edge corresponding to the consignee and the consignment address.

[0122] In the embodiment of the present disclosure, there may be two types of nodes in the graph: recipient and recipient address. The graph may include three types of relationships: recipient and recipient address, recipient and recipient, and recipient address and recipient address.

[0123] In the disclosed embodiment, the edge weight of the edge corresponding to the consignee and the consignee address can be determined according to the number of customs declarations in which the consignee and the consignee address appear at the same time.

[0124] In step S902, the number of occurrences of the customs declaration forms with the same preset fields is used as the edge weight of the corresponding edge between the corresponding recipients.

[0125] The preset fields include at least one of the recipient's phone number, the sender, the sender's city, the waybill number, the item list, and the item value.

[0126] In the disclosed embodiment, the number of customs declarations with the same fields in the customs declaration data can be used as the edge weight of the corresponding edge between the recipients:

[0127] (1)[Recipient’s phone number]

[0128] (2) [Shipper, Shipper City, Waybill Number]

[0129] (3) [Shipper, Shipper City, Waybill Number, Value]

[0130] (4) [Shipper, Shipper City, List of Goods]

[0131] (5) [Shipper, Shipper City, Goods List, Value]

[0132] In step S903, the number of occurrences of the customs declaration forms with the same preset fields is used as the edge weight of the corresponding edge between the corresponding delivery addresses.

[0133] In the disclosed embodiment, the method for determining the edge weights of the edges corresponding to the recipient addresses is similar to the method for determining the edge weights of the edges corresponding to the recipients, and will not be repeated here.

[0134] In an exemplary embodiment, the express delivery sending and receiving organization discovery method based on relationship mining may also include: determining the clustering coefficient and average clustering coefficient of each node in the sending and receiving organization relationship network according to the edges and edge weights corresponding to each node in the sending and receiving organization relationship network.

[0135] In the disclosed embodiments, the clustering coefficient can be used to describe the coefficient of the degree of clustering between nodes in a graph, specifically the degree of interconnection between adjacent nodes of a node. A clustering coefficient of 0 can indicate that the degree of connection between adjacent nodes of the node is low, and the connection between nodes is sparse. The average clustering coefficient can be the mean of the clustering coefficients of all nodes in the graph.

[0136] In an exemplary embodiment, the method for discovering express delivery sending and receiving organizations based on relationship mining may also include: if the average clustering coefficient of the sending and receiving organization relationship network is greater than a first average clustering coefficient threshold, the number of recipients in the sending and receiving organization relationship network is greater than a first number, the proportion of nodes in the sending and receiving organization relationship network whose clustering coefficient is the clustering coefficient threshold is less than a preset proportion, the types of items in the sending and receiving organization relationship network are greater than a second number, the frequency of item declarations in the sending and receiving organization relationship network is greater than a third number, and the average number of purchases in the sending and receiving organization relationship network is greater than a fourth number, then the risk level of the sending and receiving organization relationship network is determined to be the first risk level.

[0137] The first risk level may be a high risk level. The first average clustering coefficient threshold may be set as needed, for example, it may be set to 0.05; the first number may be set as needed, for example, it may be set to 4; the preset ratio may be set as needed, for example, it may be set to 1 / 3; the second number may be set as needed, for example, it may be set to 3; the third number may be set as needed, for example, it may be set to 1; and the fourth number may be set as needed, for example, it may be set to 1.

[0138] In the disclosed embodiment, at a high risk level, the density between recipients is high, and the overall declaration frequency and average declaration value of the sender-receiver relationship network are high. At a high risk level, there are at least 4 different recipient nodes in the sender-receiver relationship network, the average clustering coefficient of the network is greater than 0.05, and the number of nodes in the network with a clustering coefficient of 0 is less than 1 / 3 of the number of all nodes in the network (making the network more dense), the number of types of goods in the network is at least 3, the frequency of goods declaration is greater than or equal to 1, and the average number of purchases of different goods in the network is greater than 1.

[0139] In an exemplary embodiment, the method for discovering express delivery organizations based on relationship mining may also include: normalizing the declaration frequency of the delivery organization relationship network, the average declaration value of the delivery organization relationship network, the average purchase frequency of each item in the delivery organization relationship network, and the item types of each recipient in the delivery organization relationship network to obtain the normalized results of the declaration frequency, the normalized results of the average declaration value, the normalized results of the average purchase frequency, and the normalized results of the item types; weighting and adding the normalized results of the declaration frequency, the normalized results of the average declaration value, the normalized results of the average purchase frequency, the normalized results of the item types, and the average clustering coefficient to obtain the risk value of the delivery organization relationship network; and sorting the delivery organization relationship network according to the risk value of the delivery organization relationship network.

[0140] In the disclosed embodiment, after the risk level of the express delivery receiving and sending organization relationship network is determined, the express delivery receiving and sending organization relationship network under the risk level may be internally sorted.

[0141] Taking the express delivery and receiving organization relationship network of high risk level as an example, the risk value can be determined according to the average clustering coefficient of the network, the network's declaration frequency, the network's average declaration value, the average number of purchases of each commodity in the network, and the average type of commodities purchased by each recipient. The weight of the average clustering coefficient can be set to 0.05, the weight of the declaration frequency can be set to 0.3, the weight of the average declaration value can be set to 0.25, the weight of the average number of purchases of each commodity can be set to 0.15, and the average type of commodities purchased by each recipient can be set to 0.25. The weight of the average clustering coefficient, the declaration frequency, the average declaration value, the average number of purchases, and the types of commodities purchased can be normalized, and these values ​​can be weighted and added according to the above weights to obtain the risk value of the express delivery and receiving organization relationship network. The express delivery and receiving organization relationship network can be sorted according to its risk value.

[0142] In an exemplary embodiment, the express delivery sending and receiving organization discovery method based on relationship mining may also include: if the average clustering coefficient of the sending and receiving organization relationship network is greater than the second clustering coefficient threshold, the number of edges of the sending and receiving organization relationship network is greater than the fifth number, and the item declaration frequency of the sending and receiving organization relationship network is greater than the sixth number, then the risk level of the sending and receiving organization relationship network is determined to be the second risk level.

[0143] The second risk level may be a medium risk level. The second average clustering coefficient threshold may be set as needed, for example, it may be set to 0; the fifth number may be set as needed, for example, it may be set to 4; and the sixth number may be set as needed, for example, it may be set to 0.51.

[0144] In the disclosed embodiment, at a medium risk level, the average clustering coefficient of the network is greater than 0, which may indicate that the connection density between some nodes is high; the number of recipient nodes in the network is at least 2, the number of connected edges in the network is at least 4, and the overall commodity declaration frequency of the network is greater than or equal to 0.5.

[0145] In an exemplary embodiment, the express delivery sending and receiving organization discovery method based on relationship mining may also include: if the number of edges in the sending and receiving organization relationship network is greater than the seventh number, and the number of nodes corresponding to the recipients of the sending and receiving organization relationship network is greater than the eighth number, then the risk level of the sending and receiving organization relationship network is determined to be the third risk level.

[0146] The third risk level may be a low risk level. The seventh number may be set as needed, for example, it may be set to 3; the eighth number may be set as needed, for example, it may be set to 2.

[0147] In the disclosed embodiment, under low risk, the density of the network is low and it is not easy to form a small network. The number of recipient nodes in the network is at least 2, and the network has at least 3 connected edges.

[0148] The above three risk levels, their corresponding level descriptions, risk treatment and risk judgment can be shown in Table 1.

[0149] Table 1

[0150]

[0151]

[0152] In the disclosed embodiment, each sending and receiving organization relationship network can be analyzed to determine the purpose of the item corresponding to the receiving address in the initial organization relationship network. The purpose of the item can include commercial use and personal use.

[0153] In an exemplary embodiment, the express delivery sending and receiving organization discovery method based on relationship mining may also include: if the risk level of the sending and receiving organization relationship network is determined to be the first risk level, and the number of items in the sending and receiving organization relationship network is greater than the item number threshold or the number of item types in the sending and receiving organization relationship network is greater than the item type number threshold, then the purpose of the item corresponding to the receiving address in the sending and receiving organization relationship network is determined to be commercial purpose.

[0154] The item quantity threshold can be set as needed, for example, it can be set to 6; the item type quantity threshold can be set as needed, for example, it can be set to 10.

[0155] In the disclosed embodiment, if the level of the sending and receiving organization relationship network is classified as high risk, and the number of a certain commodity in the network is greater than 6 or the number of commodity types is greater than 10, it can be determined that the use of the item corresponding to the receiving address in the sending and receiving organization relationship is commercial use, otherwise it is for personal use.

[0156] In an exemplary embodiment, the method for discovering express delivery organizations based on relationship mining may also include: if the risk level of the delivery organization relationship network is determined to be the second risk level, and the number of items in the delivery organization relationship network is greater than the item number threshold or the number of item types in the delivery organization relationship network is greater than the item type number threshold and the total value of the delivery organization relationship network is greater than or equal to the preset value, then the purpose of the item corresponding to the receiving address in the delivery organization relationship network is determined to be commercial purpose.

[0157] The preset value can be set as needed, for example, it can be set to 1000.

[0158] In the disclosed embodiment, if the level of the sending and receiving organization relationship network is classified as medium risk, and the number of a certain commodity in the network is greater than 6 or the number of commodity types is greater than 10, and the total value of the commodities in the network is greater than or equal to 1000, then the commodity under the address is judged to be for commercial use, otherwise it is for personal use.

[0159] In an exemplary embodiment, the method for discovering express delivery organizations based on relationship mining may also include: if the risk level of the delivery organization relationship network is determined to be the third risk level, and the number of items in the delivery organization relationship network is greater than the item quantity threshold or the number of item types in the delivery organization relationship network is greater than the item type quantity threshold, the total value of the delivery organization relationship network is greater than or equal to the preset value and the delivery frequency of the delivery organization relationship network is greater than the frequency threshold, then the use of the item corresponding to the delivery address in the delivery organization relationship network is determined to be commercial use.

[0160] The frequency threshold can be set as needed, for example, it can be set to 1.

[0161] In the disclosed embodiment, if the level of the sending and receiving organization relationship network is classified as low risk, and the number of a certain commodity in the network is greater than 6 or the number of commodity types is greater than 10, and the total value of the commodities in the network is greater than or equal to 1000, and the receiving frequency of the sending and receiving organization relationship network is greater than 1, then the commodity under the address is judged to be for commercial use, otherwise it is for personal use.

[0162] The method for discovering express delivery receiving and sending organizations based on relationship mining provided by the embodiments of the present disclosure can determine the risk level of the delivery receiving and sending organization relationship network. When inspecting parcels to be cleared, the parcels declared by the recipients and receiving addresses corresponding to the delivery receiving and sending organization relationship network with higher risk levels can be inspected in particular, thereby improving inspection efficiency.

[0163] It should be noted that the above figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0164] The following are embodiments of the device disclosed herein, which can be used to execute the method embodiments disclosed herein. For details not disclosed in the device embodiments disclosed herein, please refer to the method embodiments disclosed herein.

[0165] Fig.10 It is a block diagram of a device for discovering express delivery and receiving organizations based on relationship mining according to an exemplary embodiment.

[0166] like Fig.10 As shown, the express delivery organization discovery device 1000 based on relationship mining may include: a data acquisition module 1001, an address cluster formation module 1002, a graph construction module 1003, a connected subgraph acquisition module 1004 and a relationship network determination module 1005.

[0167] Among them, the data acquisition module 1001 can be used to obtain customs declaration data including recipient addresses and recipients; the address cluster formation module 1002 can be used to form address clusters based on the similarity between recipient addresses; the graph construction module 1003 can be used to construct a graph using recipients and recipient addresses in the address cluster as nodes, and the relationship between recipients, the relationship between recipient addresses, and the relationship between recipients and recipient addresses as edges; the connected subgraph acquisition module 1004 can be used to process the nodes and edges in the graph according to the type of recipient addresses in the graph to obtain a connected subgraph; the relationship network determination module 1005 can be used to determine the relationship network of express delivery sending and receiving organizations based on the connected subgraph.

[0168] In an exemplary embodiment, the express delivery organization discovery device 1000 based on relationship mining may also include: an address cleaning module, which can be used to clean the receiving address and obtain the receiving address information of the receiving address; a word segmentation result acquisition module, which can be used to segment the receiving address information and obtain the word segmentation result of the receiving address; a similarity determination module, which can be used to determine the similarity between the receiving addresses based on the receiving address information and the word segmentation result of the receiving address.

[0169] In an exemplary embodiment, the address cleaning module may include: a character conversion unit, which can be used to delete preset characters in the recipient address and convert the character format in the recipient address so that the character format in the recipient address is the same; a province information acquisition module, which can be used to obtain the province information in the recipient address information according to the maximum forward matching method; a city information acquisition module, which can be used to obtain the city information in the recipient address information according to the maximum forward matching method if the recipient address information does not contain the province information, and determine the province information according to the city information; a district information acquisition module, which can be used to obtain the district information in the recipient address information according to the maximum forward matching method if the recipient address information does not contain the city information, and determine the city information according to the province information and the district information; a district information acquisition module, which can be used to obtain the town information in the recipient address information according to the maximum forward matching method if the recipient address information does not contain the district information, and determine the district information according to the province information, the city information and the town information.

[0170] In an exemplary embodiment, the similarity determination module may include: a first similarity determination unit, which can be used to determine the similarity between the recipient address information of the recipient address; a second similarity determination unit, which can be used to determine the similarity between the parts in the word segmentation results of the recipient address; a third similarity determination unit, which can be used to determine the similarity between the word segmentation results of the recipient address based on the similarity between the parts in the word segmentation results; a similarity acquisition unit, which can be used to weighted sum the similarity between the recipient address information of the recipient address and the similarity between the word segmentation results of the recipient address to obtain the similarity between the recipient addresses.

[0171] In an exemplary embodiment, the connected subgraph obtaining module 1004 may include: a first connected subgraph obtaining unit, which can be used to determine the target recipient connected to the recipient address if the recipient address in the connected subgraph is a public area, delete the nodes corresponding to the target recipients whose degrees in the target recipients are a preset constant and the edges connected to them, to obtain a connected subgraph; a second connected subgraph obtaining unit, which can be used to determine the target recipient connected to the recipient address if the recipient address in the connected subgraph is a non-public area, and if there is no directly connected edge between the target recipients, and the target recipients are connected through the nodes corresponding to the recipient addresses, delete the edge between the recipient address and the target recipient to obtain a connected subgraph.

[0172] In an exemplary embodiment, the relationship network determination module may include: a relationship network determination unit, which may be used to determine a connected subgraph with a total number of nodes greater than 2 as a relationship network of express delivery sending and receiving organizations.

[0173] In an exemplary embodiment, the express delivery organization discovery device 1000 based on relationship mining may also include: a first edge weight determination module, which can be used to use the number of occurrences of customs declarations with the same recipient and recipient address as the edge weight of the edge corresponding to the recipient and the recipient address; a second edge weight determination module, which can be used to use the number of occurrences of customs declarations with the same preset fields as the edge weight of the edge corresponding to the corresponding recipients; a third edge weight determination module, which can be used to use the number of occurrences of customs declarations with the same preset fields as the edge weight of the edge corresponding to the corresponding recipients; wherein the preset fields include at least one of the recipient's phone number, sender, sender's city, waybill number, item list, and item value.

[0174] In an exemplary embodiment, the express delivery organization discovery device 1000 based on relationship mining may also include: a clustering coefficient determination module, which can be used to determine the clustering coefficient and average clustering coefficient of each node in the sending and receiving organization relationship network according to the edges and edge weights corresponding to each node in the sending and receiving organization relationship network; a first risk level determination module, which can be used to determine the risk level of the sending and receiving organization relationship network as the first risk level if the average clustering coefficient of the sending and receiving organization relationship network is greater than the first clustering coefficient threshold, the number of recipients in the sending and receiving organization relationship network is greater than the first number, the proportion of nodes in the sending and receiving organization relationship network whose clustering coefficient is the clustering coefficient threshold is less than a preset proportion, the types of items in the sending and receiving organization relationship network are greater than the second number, the frequency of item declarations in the sending and receiving organization relationship network is greater than the third number, and the average number of purchases in the sending and receiving organization relationship network is greater than the fourth number.

[0175] In an exemplary embodiment, the express delivery organization discovery device 1000 based on relationship mining may also include: a normalization module, which can be used to normalize the declaration frequency of the delivery organization relationship network, the average declaration value of the delivery organization relationship network, the average purchase frequency of each item in the delivery organization relationship network, and the item type of each recipient in the delivery organization relationship network, and obtain the normalized results of the declaration frequency, the normalized results of the average declaration value, the normalized results of the average purchase frequency, and the normalized results of the item type; a risk value acquisition module, which can be used to add the normalized results of the declaration frequency, the normalized results of the average declaration value, the normalized results of the average purchase frequency, the normalized results of the item type and the average clustering coefficient after weighting, to obtain the risk value of the delivery organization relationship network; a sorting module, which can be used to sort the delivery organization relationship network according to the risk value of the delivery organization relationship network.

[0176] In an exemplary embodiment, the express delivery organization discovery device 1000 based on relationship mining may also include: a second risk level determination module, which can be used to determine the risk level of the sending and receiving organization relationship network as the second risk level if the average clustering coefficient of the sending and receiving organization relationship network is less than or equal to the first clustering coefficient threshold and greater than the clustering coefficient threshold, the number of edges of the sending and receiving organization relationship network is greater than the fifth number, and the item declaration frequency of the sending and receiving organization relationship network is greater than the sixth number.

[0177] In an exemplary embodiment, the express delivery sending and receiving organization discovery device 1000 based on relationship mining may also include: a third risk level determination module, which can be used to determine the risk level of the sending and receiving organization relationship network as the third risk level if the number of edges in the sending and receiving organization relationship network is greater than the seventh number, and the number of nodes corresponding to the recipients of the sending and receiving organization relationship network is greater than the eighth number.

[0178] In an exemplary embodiment, the express delivery sending and receiving organization discovery device 1000 based on relationship mining may also include: a first purpose determination module, which can be used to determine the purpose of the item corresponding to the receiving address in the sending and receiving organization relationship network as commercial purpose if the risk level of the sending and receiving organization relationship network is determined to be high risk, and the number of items in the sending and receiving organization relationship network is greater than the item number threshold or the number of item types in the sending and receiving organization relationship network is greater than the item type threshold; a second purpose determination module, which can be used to determine the purpose of the item corresponding to the receiving address in the sending and receiving organization relationship network as commercial purpose if the risk level of the sending and receiving organization relationship network is determined to be medium risk, and the number of items in the sending and receiving organization relationship network is greater than the item number threshold or the number of item types in the sending and receiving organization relationship network is greater than the item number threshold; If the quantity is greater than the threshold value of the number of item types and the total value of the sending and receiving organization relationship network is greater than or equal to the preset value, the purpose of the item corresponding to the receiving address in the sending and receiving organization relationship network is determined to be commercial purpose; the third purpose determination module can be used to determine the purpose of the item corresponding to the receiving address in the sending and receiving organization relationship network as commercial purpose if the risk level of the sending and receiving organization relationship network is determined to be low risk, and the number of items in the sending and receiving organization relationship network is greater than the threshold value of the number of items or the number of item types in the sending and receiving organization relationship network is greater than the threshold value of the number of item types, the total value of the sending and receiving organization relationship network is greater than or equal to the preset value and the receiving frequency of the sending and receiving organization relationship network is greater than the frequency threshold.

[0179] It should be noted that the block diagrams shown in the above figures are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0180] Fig.11 is a schematic diagram of a structure of an electronic device according to an exemplary embodiment. It should be noted that: Fig.11 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0181] like Fig.11 As shown, the electronic device 1100 includes a central processing unit (CPU) 1101, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1102 or a program loaded from a storage part 1108 into a random access memory (RAM) 1103. In the RAM 1103, various programs and data required for the operation of the system 1100 are also stored. The CPU 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0182] The following components are connected to the I / O interface 1105: an input section 1106 including a keyboard, a mouse, etc.; an output section 1107 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1108 including a hard disk, etc.; and a communication section 1109 including a network interface card such as a LAN card, a modem, etc. The communication section 1109 performs communication processing via a network such as the Internet. A drive 1110 is also connected to the I / O interface 1105 as needed. A removable medium 1111, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1110 as needed, so that a computer program read therefrom is installed into the storage section 1108 as needed.

[0183] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 1109, and / or installed from a removable medium 1111. When the computer program is executed by a central processing unit (CPU) 1101, the above-mentioned functions defined in the system of the present invention are executed.

[0184] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0185] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0186] The units involved in the embodiments of the present invention may be implemented in software or hardware. The units described may also be provided in a processor. For example, they may be described as follows: a processor includes a sending unit, an acquiring unit, a determining unit, and a first processing unit. The names of these units do not, in some cases, limit the units themselves. For example, the sending unit may also be described as a "unit that sends a request for obtaining an image to a connected server."

[0187] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: obtaining customs declaration data including a recipient address and a recipient; forming an address cluster according to the similarity between the recipient addresses; taking the recipients and recipient addresses in the address cluster as nodes, and taking the relationship between the recipients, the relationship between the recipient addresses, and the relationship between the recipients and the recipient addresses as edges to construct a graph; processing the nodes and edges in the graph according to the type of the recipient address in the graph to obtain a connected subgraph; determining the express delivery organization relationship network according to the connected subgraph. The exemplary embodiments of the present invention are specifically shown and described above.

[0188] It should be understood that the present invention is not limited to the detailed structure, arrangement or implementation described herein; on the contrary, the present invention is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

1. A method for discovering express delivery and receiving organizations based on relationship mining, characterized in that: include: Obtain customs declaration data including delivery address and recipient; According to the similarity between the recipient addresses, an address cluster is formed; The recipients and recipient addresses in the address cluster are taken as nodes, and the relationship between the recipients, the relationship between the recipient addresses, and the relationship between the recipients and the recipient addresses are taken as edges to construct a graph; According to the type of the recipient address in the graph, the nodes and edges in the graph are processed to obtain a connected subgraph; Determine the express delivery sending and receiving organization relationship network according to the connected subgraph; Normalizing the declaration frequency of the sending and receiving organization relationship network, the average declared value of the sending and receiving organization relationship network, the average purchase frequency of each item in the sending and receiving organization relationship network, and the item types of each recipient in the sending and receiving organization relationship network, respectively, to obtain the normalized result of the declaration frequency, the normalized result of the average declared value, the normalized result of the average purchase frequency, and the normalized result of the item types; The normalized result of the declaration frequency, the normalized result of the average declared value, the normalized result of the average number of purchases, the normalized result of the item types and the average clustering coefficient are weighted and added together to obtain the risk value of the sending and receiving organization relationship network; sorting the sending and receiving organization relationship network according to the risk value of the sending and receiving organization relationship network; According to the type of the recipient address in the graph, the nodes and edges in the graph are processed to obtain a connected subgraph, including: If the recipient address in the connected subgraph is a public area, determine the target recipient connected to the recipient address, delete the node corresponding to the target recipient whose degree among the target recipients is a preset constant and the edges connected thereto, and obtain the connected subgraph; If the recipient address in the connected subgraph is a non-public area, determine the target recipient connected to the recipient address. If there is no directly connected edge between the target recipients, and the target recipients are connected through the nodes corresponding to the recipient address, delete the edge between the recipient address and the target recipient to obtain the connected subgraph.

2. The method according to claim 1, characterized in that Also includes: Cleaning the receiving address to obtain the receiving address information of the receiving address; Performing word segmentation on the receiving address information to obtain a word segmentation result of the receiving address; The similarity between the receiving addresses is determined according to the receiving address information and the word segmentation result of the receiving addresses.

3. The method according to claim 2, characterized in that The receiving address is cleaned to obtain the receiving address information, including: Deleting preset characters in the recipient address and converting the character formats in the recipient address so that the character formats in the recipient address are the same; Obtaining the province information in the receiving address information according to the maximum forward matching method; If the receiving address information does not contain the province information, then obtaining the city information in the receiving address information according to the maximum forward matching method, and determining the province information according to the city information; If the receiving address information does not contain city information, obtaining the district information in the receiving address information according to the maximum forward matching method, and determining the city information according to the province information and the district information; If the receiving address information does not contain district information, the town information in the receiving address information is obtained according to the maximum forward matching method, and the district information is determined according to the province information, the city information and the town information.

4. The method according to claim 2, characterized in that: Determining the similarity between the receiving addresses according to the receiving address information and the word segmentation result of the receiving addresses includes: Determining the similarity between the receiving address information of the receiving addresses; Determine the similarity between the parts in the word segmentation result of the recipient address; Determining the similarity between the word segmentation results of the recipient addresses according to the similarity between the parts in the word segmentation results; The similarity between the recipient address information of the recipient address and the similarity between the word segmentation results of the recipient address are weighted and summed to obtain the similarity between the recipient addresses.

5. The method according to claim 1, characterized in that Determining the express delivery sending and receiving organization relationship network according to the connected subgraph includes: The connected subgraph with a total number of nodes greater than 2 in the connected subgraph is determined as the express delivery sending and receiving organization relationship network.

6. The method according to claim 1, characterized in that Also includes: The number of occurrences of the customs declaration forms with the same recipient and the same receiving address is used as the edge weight of the edge corresponding to the recipient and the receiving address; The number of occurrences of customs declarations with the same preset fields is used as the edge weight of the corresponding edge between the corresponding recipients; The number of occurrences of customs declarations with the same preset fields is used as the edge weight of the corresponding edge between the corresponding delivery addresses; The preset fields include at least one of the recipient's phone number, the sender, the sender's city, the waybill number, the item list, and the item value.

7. The method according to claim 6, characterized in that Also includes: Determine the clustering coefficient and average clustering coefficient of each node in the sending and receiving organization relationship network according to the edges and edge weights corresponding to each node in the sending and receiving organization relationship network; If the average clustering coefficient of the sending and receiving organization relationship network is greater than the first average clustering coefficient threshold, the number of recipients in the sending and receiving organization relationship network is greater than the first number, the proportion of nodes in the sending and receiving organization relationship network whose clustering coefficient is the clustering coefficient threshold is less than a preset proportion, the types of items in the sending and receiving organization relationship network are greater than the second number, the frequency of item declarations in the sending and receiving organization relationship network is greater than the third number, and the average number of purchases in the sending and receiving organization relationship network is greater than the fourth number, then the risk level of the sending and receiving organization relationship network is determined to be the first risk level.

8. The method according to claim 7, characterized in that Also includes: If the average clustering coefficient of the sending and receiving organization relationship network is greater than the second clustering coefficient threshold, the number of edges of the sending and receiving organization relationship network is greater than the fifth number, and the item declaration frequency of the sending and receiving organization relationship network is greater than the sixth number, then the risk level of the sending and receiving organization relationship network is determined to be the second risk level.

9. The method according to claim 8, characterized in that Also includes: If the number of edges of the sending and receiving organization relationship network is greater than the seventh number, and the number of nodes corresponding to the recipients of the sending and receiving organization relationship network is greater than the eighth number, the risk level of the sending and receiving organization relationship network is determined to be the third risk level.

10. The method according to claim 1, characterized in that Also includes: If the risk level of the sending and receiving organization relationship network is determined to be the first risk level, and the number of items in the sending and receiving organization relationship network is greater than the item number threshold or the number of item types in the sending and receiving organization relationship network is greater than the item type number threshold, the purpose of the item corresponding to the receiving address in the sending and receiving organization relationship network is determined to be commercial purpose; If the risk level of the sending and receiving organization relationship network is determined to be the second risk level, and the number of items in the sending and receiving organization relationship network is greater than the item quantity threshold or the number of item types in the sending and receiving organization relationship network is greater than the item type quantity threshold and the total value of the sending and receiving organization relationship network is greater than or equal to the preset value, the purpose of the item corresponding to the receiving address in the sending and receiving organization relationship network is determined to be commercial purpose; If the risk level of the sending and receiving organization relationship network is determined to be the third risk level, and the number of items in the sending and receiving organization relationship network is greater than the item quantity threshold or the number of item types in the sending and receiving organization relationship network is greater than the item type quantity threshold, the total value of the sending and receiving organization relationship network is greater than or equal to the preset value and the receiving frequency of the sending and receiving organization relationship network is greater than the frequency threshold, then the use of the items corresponding to the receiving address in the sending and receiving organization relationship network is determined to be commercial use.

11. A device for discovering express delivery and receiving organizations based on relationship mining, characterized in that: Used to perform the method according to any one of claims 1 to 10, comprising: A data acquisition module, used to acquire customs declaration data including the consignee address and consignee; An address cluster forming module, used for forming an address cluster according to the similarity between the recipient addresses; A graph construction module, used to construct a graph by taking the recipients and recipient addresses in the address cluster as nodes, and taking the relationship between the recipients, the relationship between the recipient addresses, and the relationship between the recipients and the recipient addresses as edges; A connected subgraph obtaining module, used for processing the nodes and edges in the graph according to the type of the recipient address in the graph to obtain a connected subgraph; A relationship network determination module, used to determine the relationship network of express delivery sending and receiving organizations according to the connected subgraph; Wherein, the connected subgraph obtaining module is set to: If the recipient address in the connected subgraph is a public area, determine the target recipient connected to the recipient address, delete the node corresponding to the target recipient whose degree among the target recipients is a preset constant and the edges connected thereto, and obtain the connected subgraph; If the recipient address in the connected subgraph is a non-public area, determine the target recipient connected to the recipient address. If there is no directly connected edge between the target recipients, and the target recipients are connected through the nodes corresponding to the recipient address, delete the edge between the recipient address and the target recipient to obtain the connected subgraph.

12. An electronic device, characterized in that: include: at least one processor; A storage device for storing at least one program, which, when executed by the at least one processor, enables the at least one processor to implement the method according to any one of claims 1 to 10.

13. A computer-readable storage medium having computer-executable instructions stored thereon, characterized in that: When the executable instructions are executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images