Multi-channel user fusion and target population matching method and product

Abstract

The invention discloses a multi-channel user fusion and target population matching method and product, and relates to the technical field of the Internet, and the multi-channel user fusion method comprises the steps: searching an information node matched with the identity recognition information of a target user in a graph database; if all identity identification information of the target user does not have matched information nodes, creating an ONEID node of the target user and an information node connected with the ONEID node in a graph database; if the at least one kind of identity recognition information of the target user has the matched information node, determining an ONEID node of the target user according to the information node matched with the identity recognition information of the target user; based on an ONEID node of a target user, according to the first information, executing fusion operation to fuse the first information into the graph database; wherein the first information is user information of the target user without the matched information node. And the multi-channel user fusion speed is improved.

Description

Technical Field

[0001] The present application relates to the field of Internet technology, and in particular to a method and product for multi-channel user integration and target population matching. Background Art

[0002] User channel integration refers to the integration of different sources and different types of user contact points (such as online websites, mobile applications, social media, physical stores, etc.) to provide a seamless and consistent user experience. This strategy aims to improve user satisfaction, increase user stickiness and ultimately promote sales conversion by optimizing cross-channel information flow and interaction processes.

[0003] In traditional user channel fusion, when matching users, the same fields (such as mobile phone numbers, email addresses, etc.) in multiple user tables corresponding to multiple channels in the database are used to perform association calculations to determine the user's identity. Therefore, it is necessary to use the acquired user information to traverse each field in each user table in the database one by one to determine whether the user table of the user exists in the database, resulting in large calculations, high resource usage, and slow fusion speed. Summary of the invention

[0004] The purpose of this application is to provide a multi-channel user fusion and target population matching method and product to solve the problems of large computational complexity, high resource usage and slow fusion speed in existing user channel fusion technology.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a multi-channel user integration method, comprising:

[0007] Search the graph database for an information node that matches the identity information of the target user; wherein the graph database includes multiple ONEID nodes and information nodes, different ONEID nodes correspond to different users, and the information nodes of the same user are connected to the same ONEID node; any information node is used to store a type of user information; the name of the ONEID node is ONEID, and the ID value is the ONEID of the corresponding user; the name of any information node is the type name of the user information, and the ID value is the corresponding user information value; the various types of user information stored in each information node under the same ONEID node are obtained from different channels; the various types of user information include identity information and other information, and the other information includes at least one of data information and attribute information;

[0008] If there is no matching information node for all the identity identification information of the target user, create the ONEID node of the target user and the information node connected to its ONEID node in the graph database;

[0009] If at least one identity identification information of the target user has a matching information node, an information node matching the first information is created in the graph database to merge the first information into the graph database; wherein the first information is user information of the target user without a matching information node.

[0010] Optionally, creating the ONEID node of the target user and an information node connected to the ONEID node of the target user in the graph database includes:

[0011] Generate the ONEID of the target user according to the identity identification information of the target user;

[0012] According to the ONEID of the target user, create a ONEID node of the target user in the graph database;

[0013] According to the user information of the target user, an information node of the target user is created in the graph database, and each information node of the target user is connected to its ONEID node through an edge.

[0014] Optionally, the multi-channel user integration method further includes:

[0015] Before creating an information node matching the first information in the graph database to merge the first information into the graph database, determining the ONEID node of the target user according to the information node matching the identity identification information thereof:

[0016] If each identity identification information of the target user has a matching information node, the first ONEID node is used as the ONEID node of the target user; wherein the first ONEID node is a ONEID node connected to the information node matching each identity identification information of the target user; otherwise:

[0017] If the number of first ONEID nodes is 1, and the first ONEID node meets the first fusion condition, the first ONEID node is used as the ONEID node of the target user; wherein the first fusion condition includes that n is not greater than m, n is the minimum confidence level value of all first target information nodes, m is the number of first target information nodes with a confidence level value of n, and the first target information node refers to an information node on the first ONEID node that matches the identity identification information of the target user; the confidence level value of each information node is the confidence level value of the identity identification information stored thereon;

[0018] If the number of the first ONEID node is 1, and n is greater than m, create the ONEID node of the target user and an information node connected to the ONEID node in the graph database;

[0019] If the number of first ONEID nodes is greater than 1, retain the earliest created first ONEID node as the ONEID node of the target user; if there is a first ONEID node that meets the second fusion condition, delete the first ONEID node that meets the second fusion condition; if there is a third target information node, connect the third target node to the earliest created first ONEID node; wherein the second fusion condition includes N not greater than M, N is the minimum confidence level value of all second target information nodes, M is the number of second target information nodes with a confidence level value of N, the second target information node refers to the first target information node that is connected to the earliest created first ONEID node and any other first ONEID node, and the third target node refers to an information node that is connected to the first ONEID node that meets the second fusion condition but is not connected to the earliest created first ONEID node.

[0020] Optionally, creating an information node matching the first information in the graph database to merge the first information into the graph database includes:

[0021] According to the second information, the fusion operation is performed to fuse the second information into the graph database; wherein the second information is the identity identification information in the first information;

[0022] If each identity identification information of the target user has a matching information node, or after performing the fusion operation according to the second information, searching the graph database for an information node matching other information of the target user;

[0023] If at least one other information of the target user has no matching information node, performing the fusion operation according to the third information to fuse the third information into the graph database; wherein the third information is other information in the first information;

[0024] The fusion operation includes:

[0025] Creating a fourth target information node in the graph database according to the second information or the third information; wherein the fourth information includes the second information or the third information; and the fourth target information node refers to an information node matching the second information or the third information;

[0026] Each of the fourth target information nodes is connected to the corresponding ONEID node via an edge.

[0027] Optionally, the creating an information node matching the first information in the graph database to merge the first information into the graph database further includes:

[0028] If at least one other information of the target user has a matching information node, the other information includes attribute information, and the attribute information includes attribute tags, for each attribute tag with a matching information node:

[0029] If the attribute tag value of the attribute tag does not exist in the ID value of the information node matching the attribute tag, the attribute tag value is added to the ID value of the information node matching the attribute tag.

[0030] Optionally, the other information includes attribute information, and the attribute information includes an attribute tag;

[0031] Each information node in the graph database whose name is the name of the attribute label has its ID storing the corresponding attribute label value and its word vector value;

[0032] The multi-channel user integration method further includes:

[0033] After performing the fusion operation according to the third information to fuse the third information into the graph database, selecting a fifth target information node in the graph database; wherein the fifth target information node refers to an information node that matches the attribute label of the target user;

[0034] The ID value of the fifth target information node is converted into a word vector value, and the word vector value is added to the ID value of the fifth target information node.

[0035] In a second aspect, the present application provides a target population matching method, comprising:

[0036] Search the graph database for an information node that meets the matching requirements of any user; wherein the graph database includes multiple ONEID nodes and information nodes, different ONEID nodes correspond to different users, and the information nodes of the same user are connected to the same ONEID node; any information node is used to store a type of user information; the name of the ONEID node is ONEID, and the ID value is the ONEID of the corresponding user; the name of any information node is the type name of the user information, and the ID value is the corresponding user information value; the various types of user information stored in each information node under the same ONEID node are obtained from different channels; the various types of user information include identity identification information and other information, and the other information includes at least one of data information and attribute information;

[0037] The users corresponding to the target ONEID node are selected as the target population; wherein the target ONEID node refers to the ONEID node connected to the information node that meets the user matching requirements.

[0038] Optionally, the user matching requirement includes at least one of a target data requirement and a target attribute requirement, wherein the target data requirement refers to a condition that the data information of the target population must satisfy, and the target attribute requirement refers to a condition that the attribute information of the target population must satisfy;

[0039] The searching for information nodes that meet the user's matching requirements in the graph database includes:

[0040] Searching the graph database for information nodes that meet target data requirements and information nodes that meet target attribute requirements;

[0041] The information nodes that meet the user's matching requirements include: at least one of the information nodes that meet the target data requirements and the information nodes that meet the target attribute requirements;

[0042] Among them, the information nodes that meet the target attribute requirements are found in the following way:

[0043] Convert the attribute information value of the target attribute requirement into a word vector value to obtain the requirement word vector value;

[0044] Calculate the text similarity between the demand word vector value and the word vector value of the eighth target information node; wherein the eighth target information node refers to the information node whose name is the attribute information name of the target attribute demand;

[0045] The eighth target information node used for calculating the target text similarity is used as the information node that meets the target attribute requirements; wherein the target text similarity is a text similarity greater than a preset threshold or the largest M text similarities among all calculated text similarities.

[0046] Optionally, the target data requirement includes a derived indicator requirement;

[0047] The information nodes that meet the target data requirements include: information nodes that meet the derived indicator requirements;

[0048] The information nodes that meet the requirements of the derived indicators are found in the following way:

[0049] Determine a seventh target information node according to the data information required for calculating each derived indicator; wherein the seventh target information node refers to an information node that matches the data information required for calculating each derived indicator;

[0050] Calculate the corresponding derived index value according to the ID value of the seventh target information node connected to the same ONEID node according to the set calculation rule;

[0051] A seventh target information node whose derived indicator value meets the derived indicator requirement is selected as the information node that meets the derived indicator requirement.

[0052] In a third aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of any of the multi-channel user integration methods described above and / or any of the target population matching methods described above.

[0053] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0054] The present application provides a multi-channel user integration and target population matching method and product, which searches for information nodes matching the identity identification information of the target user in a graph database; wherein the graph database includes multiple ONEID nodes and information nodes, different ONEID nodes correspond to different users, and the information nodes of the same user are connected to the same ONEID node; any information node is used to store a type of user information; the name of the ONEID node is ONEID, and the ID value is the ONEID of the corresponding user; the name of any information node is the type name of the user information, and the ID value is the corresponding user information value; the various types of user information stored in each information node under the same ONEID node are obtained from different channels; the various types of user information include identity identification information and other information , the other information includes at least one of data information and attribute information; multiple channels share the same graph database, and the information of the same user obtained from multiple channels is associated with the same ONEID node in the form of information nodes, so that when matching, it is not necessary to match each field in N user tables, reducing the amount of calculation and improving the fusion speed; and if the target user has multiple identity identification information and other information, when fusion is performed, it is only necessary to merge the identity identification information that does not match the graph database and other information to the corresponding ONEID node. Compared with the prior art, the same ONEID node will not be associated with multiple information nodes that record the same content, so that it can occupy fewer resources, solving the problems of large amount of calculation, high resource occupation, and slow fusion speed in the existing user channel fusion technology. In addition, if all the identity identification information of the target user does not have a matching information node, the ONEID node of the target user and the information node connected to its ONEID node are created in the graph database, realizing the rapid identification and fusion of new users (target users whose all identity identification information does not have a matching information node). If at least one identity identification information of the target user has a matching information node, an information node matching the first information is created in the graph database to merge the first information into the graph database; wherein the first information is the user information of the target user without a matching information node, thus realizing the rapid identification and fusion of old users (target users with at least one identity identification information matching information node). At the same time, the graph database constructed by the present application supports more flexible data expansion, has strong multi-channel scalability, and is easy to maintain and manage. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0056] Figure 1 This is an application environment diagram of a multi-channel user integration method in an embodiment of the present application;

[0057] Figure 2 This is a schematic diagram of a derived indicator node in an embodiment of the present application;

[0058] Figure 3 A flowchart of a target population matching method provided in one embodiment of the present application;

[0059] Figure 4 A schematic diagram of functional modules of a multi-channel user integration device provided in one embodiment of the present application;

[0060] Figure 5 A schematic diagram of functional modules of a target population matching device provided in one embodiment of the present application;

[0061] Figure 6 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0063] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0064] In an exemplary embodiment, Figure 1 As shown, a multi-channel user integration method is provided, including the following steps 101 to 103. Among them:

[0065] Step 101, search for an information node matching the identity information of the target user in the graph database; wherein the graph database includes multiple ONEID nodes and information nodes, different ONEID nodes correspond to different users, and the information nodes of the same user are connected to the same ONEID node; any information node is used to store one type of user information; the name of the ONEID node is ONEID, and the ID value is the ONEID of the corresponding user; the name of any information node is the type name of the user information, and the ID value is the corresponding user information value; the various types of user information stored in each information node under the same ONEID node are obtained from different channels; the various types of user information include identity information and other information, and the other information includes at least one of data information and attribute information.

[0066] The target user refers to the user corresponding to the user information obtained from a certain channel or certain channels. The user information of the target user can be obtained through interfaces (APIs (application programming interfaces) with specific systems or platforms), data tables, data documents, registration information, crawlers, etc. When obtaining customer information of a certain channel, multiple corresponding channel ID values ​​and user information may be obtained. For example, the order information of a shopping platform is retrieved from the interface, which includes the user information of the order, and the customer information of a certain enterprise service is obtained from the data table. Search for an information node that matches the identity identification information of the target user in the graph database, that is, search for an information node in the graph database whose name is consistent with the name of a certain identity identification information of the target user and whose ID value is consistent with the value of the identity identification information. For example, if the identity identification information of the target user includes a mobile phone number, search for an information node whose name is the mobile phone number and whose ID value is the 11-digit mobile phone number of the target user in the graph database. If the information node is found, it means that the mobile phone number of the target user has a matching information node. Otherwise, it means that the mobile phone number of the target user does not have a matching information node. The user's unique identification number is an identifier obtained by mapping the user's identification information to a unified ID, that is, no matter what ID the user uses to log in, no matter whether the user logs in on the PC or the mobile APP, the user's unique identity can be identified through the unique identification number. Other information refers to the user's information other than the identification information, which may include at least one of data information and attribute information. Among them, the identification information is information that can identify the user, and may include at least one of the following information: mobile phone number, ID number, email address, shopping platform ID, device ID (such as mobile phone MAC, IMEI, IMSI, etc.), mobile phone device number, address, computer (desktop computer, laptop and tablet) device number, the last four digits of the mobile phone number, browsing history, activity history, login history, platform nickname, PC-side cookies, and user name. For example, for the user's mobile phone number, the name of the information node that matches it in the graph database is the mobile phone number, and the ID value is an 11-digit mobile phone number (i.e., mobile phone number value). Data information includes but is not limited to purchase information, and purchase information may include at least one of the total amount of consumption, the total number of purchases, and the price of a single item. For example, for the total amount of consumption of a user, the name of the information node matching it in the graph database is the total amount of consumption, and the ID value is the total amount of consumption K (total amount of consumption). Purchase information can be for a specific time period (such as a week, a month, a quarter, half a year, a year, etc.) or long-term. Other information except identity identification information and data information can belong to the user's attribute information. Attribute information can include at least one of purchase categories (such as maternal and child care, beauty products, etc.) and attribute tags.Attribute tags may include at least one of demographic tags (such as age, gender, occupation (such as student), income level, etc.), hobby tags (such as basketball), social network tags (such as number of friends, number of followers, activity, etc.), and risk tags (such as fake order users, verbal attacks, etc.). For example, for the user's purchase category, the name of the information node matching it in the graph database is "purchase category", and the ID value is the purchase category information (such as maternal and child products, beauty products, etc.).

[0067] Step 102: If there is no matching information node for all the identity identification information of the target user, create the ONEID node of the target user and the information node connected to the ONEID node in the graph database.

[0068] In an embodiment of the present application, if all identity identification information of the target user does not have a matching information node, the target user is considered to be a new user, and user information fusion of the new user is achieved by creating a ONEID node of the target user and an information node connected to its ONEID node in the graph database.

[0069] Step 103, if at least one identity identification information of the target user has a matching information node, create an information node matching the first information in the graph database to merge the first information into the graph database; wherein the first information is the user information of the target user without a matching information node.

[0070] Implement the above steps 101 to 103 by searching the graph database for an information node that matches the identity information of the target user; wherein the graph database includes the ONEID node and information node of each first user, and each information node of each first user is connected to its ONEID node via an edge, the first user is a user who has completed the fusion or storage of user information, the ONEID node is a node with the name ONEID and the ID value being the ONEID of the first user; the information node is a node with the name of the corresponding user information and the ID value being the corresponding user information value; the user information includes identity information and other information obtained from different channels, and the other information includes data information and attribute information At least one; multiple channels share the same graph database, and the information of the same user obtained from multiple channels is associated with the same ONEID node in the form of an information node. In this way, when matching, it is not necessary to match each field in N user tables, reducing the amount of calculation and increasing the fusion speed; and if the target user has multiple identity identification information and other information, when fusion is performed, it is only necessary to merge the identity identification information that does not match the graph database and other information to the corresponding ONEID node. Compared with the prior art, the same ONEID node will not be associated with multiple information nodes that record the same content, so that it can occupy fewer resources, solving the problems of large amount of calculation, high resource occupation, and slow fusion speed in the existing user channel fusion technology. In addition, if all the identity identification information of the target user does not have a matching information node, the ONEID node of the target user and the information node connected to its ONEID node are created in the graph database, realizing the rapid identification and fusion of new users (target users whose all identity identification information does not have a matching information node). If at least one identity identification information of the target user has a matching information node, an information node matching the first information is created in the graph database to merge the first information into the graph database; wherein the first information is the user information of the target user without a matching information node, thus realizing the rapid identification and fusion of old users (target users with at least one identity identification information matching information node). At the same time, the graph database constructed by the present application supports more flexible data expansion, has strong multi-channel scalability, and is easy to maintain and manage.

[0071] In another exemplary embodiment of the present application, the above step 102 includes the following steps 201 to 203. Among them:

[0072] Step 201, generating the ONEID of the target user according to the identity identification information of the target user.

[0073] Step 202: Create a ONEID node of the target user in the graph database according to the ONEID of the target user.

[0074] In the embodiment of the present application, when creating the ONEID node of the target user in the graph database, ONEID is used as the name of the ONEID node, and the ONEID of the target user is used as the ID value of the ONEID node. The timestamp of creating the ONEID node can also be used as the Timestamp of the ONEID node.

[0075] For example, the target user's identity identification information obtained includes a mobile phone number and a shopping platform ID. When creating the target user's ONEID node in the graph database, the mobile phone number and the shopping platform ID are first used to generate the target user's ONEID, and then the target user's ONEID is used as the ID value of its ONEID node.

[0076] Step 203: Create an information node of the target user in the graph database according to the user information of the target user, and connect each information node of the target user with its ONEID node through an edge.

[0077] In the embodiment of the present application, when creating the information node of the target user in the graph database, the name of each user information is used as the name of the information node matching it, and the value of each user information is used as the ID value of the information node matching it. The timestamp of creating the corresponding information node can also be used as the TimeStamp of the information node.

[0078] For example, if the user information of the target user obtained includes identity information, data information and attribute information, the identity information includes the mobile phone number, the data information includes the total consumption amount and the total number of purchases, and the attribute information includes the purchase category (such as maternal and child products), then it is necessary to create 1 identification information node, 2 data information nodes and 1 attribute information node in the graph database. Among them, the name of the identification information node is "mobile phone number" and the ID value is an 11-digit mobile phone number. The names of the two data information nodes are "total consumption amount" and "total number of purchases", and the ID values ​​are the total consumption amount and the total number of purchases, respectively. The name of the attribute information node is "purchase category" and the ID value is maternal and child products.

[0079] In another exemplary embodiment of the present application, in the graph database of the above step 101, the time attribute value TimeStamp of each first node is the timestamp of its creation, and the first node includes at least one of a ONEID node and an information node.

[0080] The time attribute value TimeStamp of each second node created in step 102 is the timestamp of its creation, and the second node includes at least one of the ONEID node of the target user and the information node connected to its ONEID node.

[0081] In the embodiment of the present application, the timestamp is the total number of seconds or milliseconds that have passed since a fixed starting point (such as midnight on January 1, 1970, i.e., the Unix epoch).

[0082] In another exemplary embodiment of the present application, the above step 103 further includes:

[0083] An information node matching the first information is created in the graph database, so as to determine the ONEID node of the target user according to the information node matching the identity identification information thereof before integrating the first information into the graph database.

[0084] In another exemplary embodiment of the present application, the above-mentioned determination of the ONEID node of the target user according to the information node matching its identity identification information includes the following steps 301 to 304. Among them:

[0085] Step 301, if each identity identification information of the target user has a matching information node, the first ONEID node is used as the ONEID node of the target user; wherein the first ONEID node is the ONEID node connected to the information node matching the identity identification information of the target user.

[0086] Step 302, otherwise (i.e., if part of the identity identification information of the target user has a matching information node), if the number of the first ONEID node is 1, and the first ONEID node satisfies the first fusion condition, the first ONEID node is used as the ONEID node of the target user; wherein the first fusion condition includes that n is not greater than m, n is the minimum confidence level value of all first target information nodes, m is the number of first target information nodes with a confidence level value of n, and the first target information node refers to the information node on the first ONEID node that matches the identity identification information of the target user; the confidence level value of each information node is the confidence level value of the identity identification information stored thereon.

[0087] In the embodiment of the present application, different confidence levels are set for various identity identification information according to the contribution (strength) of each identity identification information to identifying two sets of user information as the same user. The specific division of confidence levels and the setting of the confidence level of each identity identification information can be set according to the needs, and are not specifically limited here. For example, the identity identification information is set to include mobile phone number, mobile phone device number, shopping platform ID, address, computer device number, the last four digits of the mobile phone number, browsing history, activity history, login history, platform nickname and other identity identification information. The confidence level of various identity identification information can be set according to Table 1. In Table 1, 1 is the highest confidence level and 4 is the lowest confidence level.

[0088] Table 1 Example of confidence level classification

[0089]

[0090]

[0091] For example, the information nodes connected to the first ONEID node that match the identity identification information of the target user include a mobile phone number node and an address node. If the confidence level of the mobile phone number is 1 and the confidence level of the address is 2, the confidence level value of the mobile phone number node that matches the identity identification information of the target user is 1, and the confidence level value of the address node is 2, then n=1, m=1, satisfying that n is not greater than m, and the first ONEID node is used as the ONEID node of the target user.

[0092] Step 303, otherwise (i.e., if part of the target user's identity identification information has a matching information node), if the number of the first ONEID node is 1, and n is greater than m, create the target user's ONEID node and the information node connected to its ONEID node in the graph database.

[0093] For example, the information nodes connected to the first ONEID node that match the identity identification information of the target user include address nodes and platform nickname nodes. If the confidence level of the address is 2 and the confidence level of the platform nickname is 3, the confidence level value of the address node that matches the identity identification information of the target user is 2, and the confidence level value of the platform nickname node is 3. Then n=2, m=1, and n is greater than m. Then, the ONEID node of the target user and the information node connected to its ONEID node are created in the graph database.

[0094] Step 304, otherwise (i.e., if part of the identity identification information of the target user has a matching information node), if the number of first ONEID nodes is greater than 1, retain the first ONEID node created earliest as the ONEID node of the target user, if there is a first ONEID node that meets the second fusion condition, delete the first ONEID node that meets the second fusion condition, if there is a third target information node, connect the third target node to the first ONEID node created earliest; wherein the second fusion condition includes that N is not greater than M, N is the minimum confidence level value of all second target information nodes, M is the number of second target information nodes with a confidence level value of N, the second target information node refers to the first target information node that is connected to the first ONEID node created earliest and any other first ONEID node, and the third target node refers to an information node that is connected to the first ONEID node that meets the second fusion condition but is not connected to the first ONEID node created earliest.

[0095] In an embodiment of the present application, if for the ONEID nodes in the graph database, the timestamp of creating each ONEID node is used as its TimeStamp, then the earliest first ONEID node is created, that is, the first ONEID node with the smallest TimeStamp value.

[0096] For example, the target user's identity information includes the mobile phone number, the last four digits of the mobile phone number and the platform nickname. ONEID node A connects the last four digits of the mobile phone number node and the platform nickname node. ONEID node B connects the address node, the mobile phone number node, the last four digits of the mobile phone number node and the platform nickname node. ONEID node C connects the address node, the mobile phone number node and the last four digits of the mobile phone number node. The last four digits of the mobile phone number node and the platform nickname node connected to ONEID node A match the target user's identity information. The mobile phone number node, the last four digits of the mobile phone number node and the platform nickname node connected to ONEID node B match the target user's identity information. The mobile phone number node and the last four digits of the mobile phone number node connected to ONEID node C match the target user's identity information. Then the number of the first ONEID node is 3, which satisfies the requirement that the number of the first ONEID node is greater than 1. At this time, the earliest established ONEID node A is used as the target user's ONEID node, and it is determined that the ONEID node B and ONEID node C need to be deleted. Specifically, for ONEID node A and ONEID node C, they are connected to the node with the last four digits of the mobile phone number and the mobile phone number node at the same time. If the confidence level value of the node with the last four digits of the mobile phone number is 2 and the confidence level value of the mobile phone number node is 1, then N=1, M=1, and N is not greater than M, that is, ONEID node C meets the second fusion condition. It can be considered that ONEID node A and ONEID node C correspond to the same person, then ONEID node C is deleted, and the address node connected to ONEID node C but not connected to ONEID node A is connected to ONEID node A as the third target information node. Then, for ONEID node A and ONEID node B, they are connected to the last four digits of the mobile phone number node, the platform nickname node and the address node at the same time. If the confidence level values ​​of the last four digits of the mobile phone number node and the address node are both 2, and the confidence level value of the platform nickname node is 3, then N=2, M=2, satisfying N is not greater than M, that is, ONEID node B meets the second fusion condition, and it can be considered that ONEID node A and ONEID node B correspond to the same person, because there is no information node connected to ONEID node B but not connected to ONEID node A, that is, there is no third target node, then ONEID node B can be deleted.

[0097] In another exemplary embodiment of the present application, in the above step 103, creating an information node matching the first information in the graph database to integrate the first information into the graph database includes the following steps 401 to 403. Among them:

[0098] Step 401, performing a fusion operation based on the second information to fuse the second information into the graph database; wherein the second information is the identity identification information in the first information.

[0099] Step 402: If each identity identification information of the target user has a matching information node, or after performing a fusion operation based on the second information, search the graph database for an information node that matches other information of the target user.

[0100] Step 403: If at least one other information of the target user has no matching information node, perform a fusion operation based on the third information to fuse the third information into the graph database; wherein the third information is other information in the first information.

[0101] In an embodiment of the present application, by performing a fusion operation, for target users whose partial identity identification information has matching information nodes, identity identification information and other information that do not have matching information nodes can be fused into the graph database; for target users whose all identity identification information has matching information nodes, other information that does not have matching information nodes can be fused into the graph database.

[0102] In another exemplary embodiment of the present application, the fusion operation includes the following steps 501 to 502.

[0103] in:

[0104] Step 501, creating a fourth target information node in the graph database according to the second information or the third information; wherein the fourth target information node refers to an information node that matches the second information or the third information.

[0105] In an embodiment of the present application, a fourth target information node is created in the graph database according to the second information or the third information, that is, an information node is created in the graph database with the name of the second information or the name of the third information and the ID value is the information value of the second information or the information value of the third information. For example, for example, the user information of the target user obtained includes identity information and data information, the identity information includes the mobile phone number and the shopping platform ID, the data information includes the total amount of consumption and the total number of purchases, the shopping platform ID and the total number of purchases do not have a matching information node in the graph database, that is, the first information includes the shopping platform ID and the total number of purchases, the second information includes the shopping platform ID, and the third information includes the total number of purchases, then a fourth target information node is created in the graph database, that is, an information node matching the shopping platform ID and an information node matching the total number of purchases are created in the graph database. The timestamp of creating the fourth target information node can also be used as the TimeStamp value of the fourth target information node.

[0106] Step 502: Connect each fourth target information node to the corresponding ONEID node via an edge.

[0107] In another exemplary embodiment of the present application, the above step 103 further includes:

[0108] If at least one other information of the target user has a matching information node, where the other information includes attribute information, and the attribute information includes attribute tags, for each attribute tag with a matching information node:

[0109] If the attribute tag value of the attribute tag does not exist in the ID value of the information node matching the attribute tag, the attribute tag value is added to the ID value of the information node matching the attribute tag.

[0110] For example, in the graph database, for an information node named Purchase Category, if its ID value is maternal and child products, but the target user's other information includes attribute information, and the attribute information includes Purchase Category, and the value of the Purchase Category is beauty and cosmetics, then it is necessary to add beauty and cosmetics to the ID value of the information node named Purchase Category.

[0111] In another exemplary embodiment of the present application, the above step 103 further includes:

[0112] If at least one other information of the target user has a matching information node, where the other information includes data information, and the data information includes purchase information, for each purchase information with a matching information node:

[0113] The purchase information value of the target user and the ID value of the sixth target information node are calculated according to the set calculation rule, and the ID value of the sixth target information node is updated as the calculation result.

[0114] In the embodiment of the present application, the sixth target information node refers to the information node that matches the purchase information of the target user. The purchase information and the set calculation rules are not specifically limited here, and can be set according to actual needs. For example, the purchase information of the target user includes the total consumption amount and the total number of purchases. There are information nodes that match the total consumption amount and the total number of purchases of the target user in the graph database. The set calculation rule is to add the purchase information of the target user to the ID value of the sixth target information node, then the ID value of the information node that matches the total consumption amount of the target user is added to the total consumption amount of the target user, the ID value of the information node that matches the total number of purchases of the target user is added to the total number of purchases of the target user, and the ID value of the information node that matches the total consumption amount of the target user is updated to the total consumption amount obtained by the addition, and the ID value of the information node that matches the total number of purchases of the target user is updated to the total number of purchases obtained by the addition.

[0115] In another exemplary embodiment of the present application, the other information includes attribute information, and the attribute information includes an attribute tag;

[0116] In the above step 101, each information node in the graph database whose name is the name of the attribute label has its ID storing the corresponding attribute label value and its word vector value.

[0117] In the embodiment of the present application, a word vector conversion method including but not limited to Word2Vec, GloVe, etc. can be used to convert the attribute label value (text) stored in the ID of the information node whose name is the name of the attribute label into a word vector value to obtain the word vector value of the attribute label value. For example, in the graph database, the ID value of the information node named "Purchase Category" of a first user is "Mother and Baby", and "Mother and Baby" is converted into a word vector value to obtain [0.34834, 0.67673, 0.14313, 0.565634, 0.56778, 0.11122], and "Mother and Baby" and its word vector value [0.34834, 0.67673, 0.14313, 0.565634, 0.56778, 0.11122] are stored simultaneously in the ID value of the information node named "Purchase Category" of the first user. Saving the converted word vector values ​​facilitates fast and accurate matching of similar attribute labels when calculating text similarity when matching the target population.

[0118] Accordingly, the above multi-channel user integration method further includes the following steps 601 to 602. Among them:

[0119] Step 601, after performing a fusion operation based on the third information to fuse the third information into the graph database, a fifth target information node is selected in the graph database.

[0120] In the embodiment of the present application, the seventh target information node refers to an information node that matches the attribute tag of the target user, that is, an information node whose name is the name of the attribute tag of the target user and whose ID value is the attribute tag value of the attribute tag of the target user.

[0121] Step 602: Convert the ID value of the fifth target information node into a word vector value, and add the word vector value to the ID value of the fifth target information node.

[0122] In another exemplary embodiment of the present application, the multi-channel user integration method further includes:

[0123] After the above steps 102, 303, 304, 401, 402 and 403, the logs are updated and saved as log files, entity tables or external zipper tables for future query or provision to other business parties.

[0124] In another exemplary embodiment of the present application, the other information includes data information, and the data information includes a derivative indicator;

[0125] The above step 403 further includes:

[0126] Step 701: If at least one derived indicator of the target user has no matching information node, determine the seventh target information node based on the data information required to calculate each first target derived indicator. The first target derived indicator refers to each derived indicator of the target user that has no matching information node; the seventh target information node refers to an information node that matches the data information required to calculate each derived indicator.

[0127] In the embodiment of the present application, a derived indicator refers to an indicator that does not have an information node with the name of the derived indicator in the graph database, and needs to be calculated according to the ID value of the information node in the graph database according to the set calculation rules.

[0128] For example, the average unit price of the target user does not have a matching information node, and the data information required to calculate the average unit price is the total consumption amount and the total number of purchases. In this case, in the graph database, select the information nodes named total consumption amount and total number of purchases that are connected to the ONEID node of the target user.

[0129] Step 702: Calculate the data information required for calculating each first target derived indicator of the target user and the ID value of the seventh target information node according to the set calculation rule, and update the ID value of the seventh target information node as the calculation result.

[0130] Step 703, calculating the first target derived indicator according to the calculation result of step 702 and the set calculation rules.

[0131] In the embodiment of the present application, there is no specific limitation on the types of derived indicators and the calculation rules of each derived indicator, and they are set according to actual needs. For example, if the average unit price of a target user does not have a matching information node, and the calculation rule is set to the total amount of consumption divided by the total number of purchases, the total amount of consumption calculated in step 702 is divided by the total number of purchases calculated in step 702 to obtain the average unit price of the target user.

[0132] Step 704: Create an information node matching the calculated first target derived indicator in the graph database, and connect the created information node matching the calculated first target derived indicator to the associated seventh target information node through an edge.

[0133] In the embodiment of the present application, an information node matching the calculated first target derived indicator is created in the graph database, that is, an information node with the name of each derived indicator of the target user that has no matching information node and an ID value of the first target derived indicator value is created. Figure 2 As shown, the created information node named unit price (referring to the average unit price) is connected to the information node named total consumption amount and the information node named total number of purchased items of the target user associated with it.

[0134] In another exemplary embodiment of the present application, Figure 3 As shown, a target population matching method is provided, including the following steps 801 to 802. Among them:

[0135] Step 801, search for information nodes that meet the user matching requirements in the graph database; wherein the fusion operation graph database includes multiple ONEID nodes and information nodes, different ONEID nodes correspond to different users, and the information nodes of the same user are connected to the same ONEID node; any information node is used to store a type of user information; the name of the fusion operation ONEID node is ONEID, and the ID value is the ONEID of the corresponding user; the name of any information node is the type name of the user information, and the ID value is the corresponding user information value; the various user information stored in each information node under the same ONEID node is obtained from different channels; the fusion operation of various user information includes identity identification information and other information, and the fusion operation of other information includes at least one of data information and attribute information.

[0136] Step 802, select users corresponding to the target ONEID node as the target population; wherein the target ONEID node refers to the ONEID node connected to the information node that meets the user matching requirements.

[0137] In the embodiment of the present application, the user matching requirement refers to the conditions that the user information of the target population must meet. If the user matching requirement includes multiple conditions, satisfying the user matching requirement may be satisfying at least one condition of the user matching requirement, and the target ONEID node may be a ONEID node connected to at least one information node that satisfies the user matching requirement. For example, the acquired user matching requirement includes two conditions: 1. The total purchase amount is greater than X, 2. The total number of purchases is greater than Y. Satisfying at least one of condition 1 and condition 2 satisfies the user matching requirement. The target ONEID node may be a ONEID node connected to an information node that satisfies condition 1, may be a ONEID node connected to an information node that satisfies condition 2, or may be a ONEID node connected to both an information node that satisfies condition 1 and an information node that satisfies condition 2.

[0138] In another exemplary embodiment of the application, the user matching requirement includes at least one of a target data requirement and a target attribute requirement, wherein the target data requirement refers to a condition that the data information of the target population must satisfy, and the target attribute requirement refers to a condition that the attribute information of the target population must satisfy;

[0139] Accordingly, the above step 802 includes:

[0140] Search the graph database for information nodes that meet the target data requirements and information nodes that meet the target attribute requirements.

[0141] The information nodes that meet the user's matching requirements include: at least one of an information node that meets the target data requirements and an information node that meets the target attribute requirements.

[0142] In an embodiment of the present application, if the target data requirement includes multiple conditions, satisfying the target data requirement may be at least one condition that satisfies the target data requirement. If the target attribute requirement includes multiple conditions, satisfying the target attribute requirement may be at least one condition that satisfies the target attribute requirement. The target ONEID node may be a ONEID node connected to at least one of the information nodes that meet the target data requirement and the information nodes that meet the target attribute requirement. For example, a user matching requirement is obtained, wherein the target data requirement includes conditions 1 and 2, condition 1 is that the total purchase amount is greater than X, condition 2 is that the total number of purchases is greater than Y (purchase information requirement), and the target attribute requirement includes conditions 3 and 4, condition 3 is that the purchase category preference is "baby products", and condition 4 is that the purchase category preference is "beauty". Information node a in which the total purchase amount count is greater than X in the information node named total purchase amount, or information node b in which the total number of purchases is greater than Y in the information node named total number of purchases can be selected as the information node that meets the target data requirement. Information node c named as purchase category and with ID value including "baby products" or information node d named as purchase category and with ID value including "beauty" can be selected as the information node that meets the target attribute requirements. Selecting at least one of information nodes a, b, c and d is the information node that meets the user's matching requirements.

[0143] In another exemplary embodiment of the present application, the information node that meets the target attribute requirements is searched in the following manner:

[0144] Step a: convert the attribute information value of the target attribute requirement into a word vector value to obtain the requirement word vector value.

[0145] Step b, calculating the text similarity between the demand word vector value and the word vector value of the eighth target information node; wherein the eighth target information node refers to the information node whose name is the attribute information name of the target attribute requirement.

[0146] In an embodiment of the present application, if multiple word vector values ​​are stored in the ID of the eighth target information node, step b calculates the text similarity between the demand word vector value and each word vector value stored in the ID value of the eighth target information node. For example, if a certain target attribute demand is to purchase a category preference for baby products, then "purchase category" is the attribute information name of the target attribute demand, and the information node named "purchase category" is used as the eighth target information node. Step a converts "baby products" into a word vector value to obtain a demand word vector value. If the ID value in the eighth target information node includes baby products, beauty makeup, and the word vector values ​​of both, step b calculates the text similarity between the demand word vector value and each word vector value stored in the ID value of the eighth target information node.

[0147] Step c, using the eighth target information node used to calculate the target text similarity as the information node that meets the target attribute requirements; the target text similarity is a text similarity greater than a preset threshold or the largest M text similarities among all calculated text similarities.

[0148] In the embodiment of the present application, the preset threshold and M are not specifically limited and can be set according to actual needs or set as default values. For example, the words with a text similarity greater than 0.7 matched by "baby products" in the user matching needs are "mother and baby products", "newborns", and "low months old". Users corresponding to the ONEID nodes connected to at least one of the information nodes where the three groups of attribute label values ​​are located are selected as the target population.

[0149] In another exemplary embodiment of the present application, the target data demand includes a purchase information demand, and the purchase information demand includes a derivative indicator demand.

[0150] The information nodes that meet the target data requirements include: information nodes that meet the derived indicator requirements. Accordingly, the information nodes that meet the derived indicator requirements are searched through the following steps (1) to (3).

[0151] in:

[0152] Step (1), determining the seventh target information node according to the data information required for calculating each derived indicator; wherein the seventh target information node refers to an information node that matches the data information required for calculating each derived indicator.

[0153] In the embodiment of the present application, a derived indicator refers to an indicator that does not have an information node named as the derived indicator in the graph database, and needs to be calculated according to the set calculation rules based on the ID value of the information node in the graph database. The derived indicator requirement refers to the conditions that the derived indicator of the target population must meet. The information node that matches the data information required to calculate each derived indicator refers to the information node whose name is the name of the data information required to calculate each derived indicator. For example, a certain target data requirement is that the average unit price is greater than z, and the calculation rule of the average unit price is the total purchase amount divided by the total number of purchases. There is no information node named "average unit price" in the graph database. The average unit price needs to be calculated based on the ID values ​​of the information nodes named "total purchase amount" and "total number of purchases" in the graph database. Then the average unit price is a derived indicator, and the average unit price is greater than z is a derived indicator requirement. The information nodes named "total purchase amount" and "total number of purchases" are the seventh target information nodes.

[0154] Step (2), according to the ID value of the seventh target information node connected to the same ONEID node, calculate the corresponding derived index value according to the set calculation rules.

[0155] For example, a ONEID node in the graph database is connected to an information node named "Total purchase amount" and an information node named "Total number of purchases" at the same time. According to the calculation rules of the average unit price, the ID value (total purchase amount) of the information node named "Total purchase amount" is divided by the ID value (total number of purchases) of the information node named "Total number of purchases" to obtain the corresponding average unit price.

[0156] Step (3), selecting the seventh target information node whose derived indicator value meets the derived indicator requirement as the information node that meets the derived indicator requirement.

[0157] In the embodiment of the present application, if the derived indicator requirement includes multiple conditions, satisfying the derived indicator requirement may be satisfying at least one condition of the derived indicator requirement. For example, a target data requirement includes two conditions: 1. the total purchase amount is greater than X, and 2. the average unit price is greater than z. Then, the information node named "total purchase amount" and the information node named "total number of purchases" connected to the same ONEID node and with an average unit value greater than z are selected as the information nodes that meet the derived indicator requirements; or the information node named "total purchase amount" whose total purchase amount is greater than X is selected as the information node that meets the derived indicator requirements; or the information node named "total purchase amount" and the information node named "total number of purchases" connected to the same ONEID node and with an average unit value greater than z, and the information node named "total purchase amount" whose total purchase amount is greater than X are selected as the information nodes that meet the derived indicator requirements; or, among the information node named "total purchase amount" and the information node named "total number of purchases" connected to the same ONEID node and with an average unit value greater than z, the information node whose total purchase amount is greater than X and the information node named "total purchase amount" connected to the same ONEID node are selected as the information nodes that meet the derived indicator requirements. There is no specific limitation here and you can choose according to your actual needs.

[0158] Based on the same inventive concept, the embodiment of the present application also provides a multi-channel user fusion device for implementing the multi-channel user fusion method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more multi-channel user fusion device embodiments provided below can refer to the limitations of the multi-channel user fusion method above, and will not be repeated here.

[0159] In an exemplary embodiment, Figure 4 As shown, a multi-channel user integration device 90 is provided, comprising:

[0160] The first identification and matching module 901 is used to search for an information node matching the identity identification information of the target user in the graph database; wherein the graph database includes multiple ONEID nodes and information nodes, different ONEID nodes correspond to different users, and the information nodes of the same user are connected to the same ONEID node; any information node is used to store a type of user information; the name of the ONEID node is ONEID, and the ID value is the ONEID of the corresponding user; the name of any information node is the type name of the user information, and the ID value is the corresponding user information value; the various types of user information stored in each information node under the same ONEID node are obtained from different channels; the various types of user information include identity identification information and other information, and the other information includes at least one of data information and attribute information;

[0161] The first fusion module 902 is used to create a ONEID node of the target user and an information node connected to the ONEID node of the target user in the graph database if all the identity identification information of the target user has no matching information node;

[0162] The second creation module 903 is used to create an information node matching the first information in the graph database if at least one identity identification information of the target user has a matching information node, so as to integrate the first information into the graph database; wherein the first information is the user information of the target user without a matching information node.

[0163] In another exemplary embodiment of the present application, the first fusion module 902 is further configured to:

[0164] Generate the target user's ONEID based on the target user's identity identification information;

[0165] According to the target user's ONEID, create the target user's ONEID node in the graph database;

[0166] According to the user information of the target user, an information node of the target user is created in the graph database, and each information node of the target user is connected to its ONEID node through an edge.

[0167] In another exemplary embodiment of the present application, the second creation module 903 is further used to:

[0168] An information node matching the first information is created in the graph database, so as to determine the ONEID node of the target user according to the information node matching the identity identification information thereof before integrating the first information into the graph database:

[0169] If each identity identification information of the target user has a matching information node, the first ONEID node is used as the ONEID node of the target user;

[0170] otherwise:

[0171] If the number of first ONEID nodes is 1, and the first ONEID node meets the first fusion condition, the first ONEID node is used as the ONEID node of the target user; wherein the first fusion condition includes that n is not greater than m, n is the minimum confidence level value of all first target information nodes, m is the number of first target information nodes with a confidence level value of n, and the first target information node refers to an information node on the first ONEID node that matches the identity identification information of the target user; the confidence level value of each information node is the confidence level value of the identity identification information stored thereon;

[0172] If the number of the first ONEID node is 1, and n is greater than m, create the target user's ONEID node and the information node connected to its ONEID node in the graph database;

[0173] If the number of first ONEID nodes is greater than 1, retain the earliest created first ONEID node as the ONEID node of the target user; if there is a first ONEID node that meets the second fusion condition, delete the first ONEID node that meets the second fusion condition; if there is a third target information node, connect the third target node to the earliest created first ONEID node; wherein the second fusion condition includes that N is not greater than M, N is the minimum confidence level value of all second target information nodes, M is the number of second target information nodes with a confidence level value of N, the second target information node refers to the first target information node that is connected to the earliest created first ONEID node and any other first ONEID node, and the third target node refers to an information node that is connected to the first ONEID node that meets the second fusion condition but is not connected to the earliest created first ONEID node.

[0174] In another exemplary embodiment of the present application, the second creation module 904 is further used to:

[0175] According to the second information, a fusion operation is performed to fuse the second information into the graph database; wherein the second information is the identity identification information in the first information;

[0176] If each identity identification information of the target user has a matching information node, or after performing a fusion operation based on the second information, searching the graph database for information nodes that match other information of the target user;

[0177] If at least one other information of the target user has no matching information node, performing a fusion operation according to the third information to fuse the third information into the graph database; wherein the third information is other information in the first information;

[0178] The fusion operation includes:

[0179] Creating a fourth target information node in the graph database according to the second information or the third information;

[0180] Each of the fourth target information nodes is connected to the corresponding ONEID node via an edge.

[0181] In another exemplary embodiment of the present application, the second creation module 903 is further used to:

[0182] If at least one other information of the target user has a matching information node, where the other information includes attribute information, and the attribute information includes attribute tags, for each attribute tag with a matching information node:

[0183] If the attribute tag value of the attribute tag does not exist in the ID value of the information node matching the attribute tag, the attribute tag value is added to the ID value of the information node matching the attribute tag.

[0184] In another exemplary embodiment of the present application, the second creation module 903 is further used to:

[0185] If at least one other information of the target user has a matching information node, where the other information includes data information, and the data information includes purchase information, for each purchase information with a matching information node:

[0186] The purchase information value of the target user and the ID value of the sixth target information node are calculated according to the set calculation rule, and the ID value of the sixth target information node is updated as the calculation result.

[0187] In another exemplary embodiment of the present application, the other information includes data information, and the data information includes a derivative indicator;

[0188] The second creation module 903 is further used for:

[0189] If at least one derived indicator of the target user has no matching information node, determining a seventh target information node according to data information required for calculating each first target derived indicator;

[0190] Calculate the data information required for calculating each first target derived indicator of the target user and the ID value of the seventh target information node according to the set calculation rules, and update the ID value of the seventh target information node as the calculation result; wherein the first target derived indicator refers to each derived indicator of the target user that has no matching information node; the seventh target information node refers to the information node that matches the data information required for calculating each derived indicator;

[0191] Calculate the data information required for calculating each first target derived indicator of the target user and the ID value of the seventh target information node according to the set calculation rules, and update the ID value of the seventh target information node as the calculation result;

[0192] Calculate the first target derived indicator according to the calculation results and the set calculation rules;

[0193] An information node matching the calculated first target derived indicator is created in the graph database, and the created information node matching the calculated first target derived indicator is connected to the associated seventh target information node through an edge.

[0194] In another exemplary embodiment of the present application, each information node in the graph database whose name is the name of the attribute label has its ID storing the corresponding attribute label value and its word vector value;

[0195] The multi-channel user integration device 90 further includes:

[0196] The word vector conversion module 904 is used to perform a fusion operation according to the third information to fuse the third information into the graph database, and then select a fifth target information node in the graph database;

[0197] The ID value of the fifth target information node is converted into a word vector value, and the word vector value is added to the ID value of the fifth target information node.

[0198] Based on the same inventive concept, the embodiment of the present application also provides a target population matching device for implementing the target population matching method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more target population matching device embodiments provided below can refer to the limitations of the target population matching method above, and will not be repeated here.

[0199] In an exemplary embodiment, Figure 5 As shown, a target population matching device 100 is provided, comprising:

[0200] The second identification and matching module 1001 is used to search for information nodes that meet the user matching requirements in the graph database; wherein the graph database includes multiple ONEID nodes and information nodes, different ONEID nodes correspond to different users, and the information nodes of the same user are connected to the same ONEID node; any information node is used to store a type of user information; the name of the ONEID node is ONEID, and the ID value is the ONEID of the corresponding user; the name of any information node is the type name of the user information, and the ID value is the corresponding user information value; the various user information stored in each information node under the same ONEID node is obtained from different channels; the various user information includes identity identification information and other information, and the other information includes at least one of data information and attribute information;

[0201] The target population matching module 1002 is used to select users corresponding to the target ONEID node as the target population; wherein the target ONEID node refers to the ONEID node connected to the information node that meets the matching requirements.

[0202] In another exemplary embodiment of the present application, the user matching requirement includes at least one of a target data requirement and a target attribute requirement, wherein the target data requirement refers to a condition that the data information of the target population must satisfy, and the target attribute requirement refers to a condition that the attribute information of the target population must satisfy;

[0203] Accordingly, the second identification and matching module 1001 is further used for:

[0204] Search the graph database for information nodes that meet the target data requirements and information nodes that meet the target attribute requirements.

[0205] The information nodes that meet the user's matching requirements include: at least one of an information node that meets the target data requirements and an information node that meets the target attribute requirements.

[0206] In another exemplary embodiment of the present application, the second identification and matching module 1001 is further used to:

[0207] Find the information node that meets the target attribute requirements in the following ways:

[0208] Convert the attribute information value of the target attribute requirement into a word vector value to obtain the requirement word vector value;

[0209] Calculate the text similarity between the demand word vector value and the word vector value of the eighth target information node; wherein the eighth target information node refers to the information node whose name is the attribute information name of the target attribute demand;

[0210] The eighth target information node used for calculating the target text similarity is used as the information node that meets the target attribute requirements; the target text similarity is a text similarity greater than a preset threshold or the largest M text similarities among all calculated text similarities.

[0211] In another exemplary embodiment of the present application, the target data requirement includes a derived indicator requirement; the information node that meets the target data requirement includes: the information node that meets the derived indicator requirement, and accordingly, the second identification and matching module 1001 is further used to:

[0212] Find the information nodes that meet the requirements of the derived indicators in the following ways:

[0213] Determine the seventh target information node according to the data information required for calculating each derivative indicator; the seventh target information node refers to an information node that matches the data information required for calculating each derivative indicator;

[0214] According to the ID value of the seventh target information node connected to the same ONEID node, the corresponding derived index value is calculated according to the set calculation rules;

[0215] The seventh target information node whose derived indicator value meets the derived indicator requirement is selected as the information node that meets any target attribute requirement.

[0216] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store multi-channel user fusion and / or target population matching data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a multi-channel user fusion method and / or a target population matching method is implemented.

[0217] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0218] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0219] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0220] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0221] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0222] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0223] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0224] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0225] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A multi-channel user integration method, characterized in that: include: Search the graph database for an information node that matches the identity information of the target user; wherein the graph database includes multiple ONEID nodes and information nodes, different ONEID nodes correspond to different users, and the information nodes of the same user are connected to the same ONEID node; any information node is used to store a type of user information; the name of the ONEID node is ONEID, and the ID value is the ONEID of the corresponding user; the name of any information node is the type name of the user information, and the ID value is the corresponding user information value; the various types of user information stored in each information node under the same ONEID node are obtained from different channels; the various types of user information include identity information and other information, and the other information includes at least one of data information and attribute information; If there is no matching information node for all the identity identification information of the target user, create the ONEID node of the target user and the information node connected to its ONEID node in the graph database; If at least one identity identification information of the target user has a matching information node, an information node matching the first information is created in the graph database to merge the first information into the graph database; wherein the first information is user information of the target user without a matching information node.

2. The multi-channel user integration method according to claim 1, characterized in that: The step of creating the target user's ONEID node and an information node connected to the target user's ONEID node in the graph database includes: Generate the ONEID of the target user according to the identity identification information of the target user; According to the ONEID of the target user, create a ONEID node of the target user in the graph database; According to the user information of the target user, an information node of the target user is created in the graph database, and each information node of the target user is connected to its ONEID node through an edge.

3. The multi-channel user integration method according to claim 1, characterized in that: Also includes: Before creating an information node matching the first information in the graph database to merge the first information into the graph database, determining the ONEID node of the target user according to the information node matching the identity identification information thereof: If each identity identification information of the target user has a matching information node, the first ONEID node is used as the ONEID node of the target user; wherein the first ONEID node is a ONEID node connected to the information node matching each identity identification information of the target user; otherwise: If the number of first ONEID nodes is 1, and the first ONEID node meets the first fusion condition, the first ONEID node is used as the ONEID node of the target user; wherein the first fusion condition includes that n is not greater than m, n is the minimum confidence level value of all first target information nodes, m is the number of first target information nodes with a confidence level value of n, and the first target information node refers to an information node on the first ONEID node that matches the identity identification information of the target user; the confidence level value of each information node is the confidence level value of the identity identification information stored thereon; If the number of the first ONEID node is 1, and n is greater than m, create the ONEID node of the target user and an information node connected to the ONEID node in the graph database; If the number of first ONEID nodes is greater than 1, retain the earliest created first ONEID node as the ONEID node of the target user; if there is a first ONEID node that meets the second fusion condition, delete the first ONEID node that meets the second fusion condition; if there is a third target information node, connect the third target node to the earliest created first ONEID node; wherein the second fusion condition includes N not greater than M, N is the minimum confidence level value of all second target information nodes, M is the number of second target information nodes with a confidence level value of N, the second target information node refers to the first target information node that is connected to the earliest created first ONEID node and any other first ONEID node, and the third target node refers to an information node that is connected to the first ONEID node that meets the second fusion condition but is not connected to the earliest created first ONEID node.

4. The multi-channel user integration method according to claim 1, characterized in that: The step of creating an information node matching the first information in the graph database to integrate the first information into the graph database includes: According to the second information, the fusion operation is performed to fuse the second information into the graph database; wherein the second information is the identity identification information in the first information; If each identity identification information of the target user has a matching information node, or after performing the fusion operation according to the second information, searching the graph database for an information node matching other information of the target user; If at least one other information of the target user has no matching information node, performing the fusion operation according to the third information to fuse the third information into the graph database; wherein the third information is other information in the first information; The fusion operation includes: Creating a fourth target information node in the graph database according to the second information or the third information; the fourth target information node refers to an information node matching the second information or the third information; Each of the fourth target information nodes is connected to the corresponding ONEID node via an edge.

5. The multi-channel user integration method according to claim 4, characterized in that: The step of creating an information node matching the first information in the graph database to integrate the first information into the graph database further includes: If at least one other information of the target user has a matching information node, the other information includes attribute information, and the attribute information includes attribute tags, for each attribute tag with a matching information node: If the attribute tag value of the attribute tag does not exist in the ID value of the information node matching the attribute tag, the attribute tag value is added to the ID value of the information node matching the attribute tag.

6. The multi-channel user integration method according to claim 4, characterized in that: The other information includes attribute information, and the attribute information includes an attribute tag; Each information node in the graph database whose name is the name of the attribute label has its ID storing the corresponding attribute label value and its word vector value; The multi-channel user integration method further includes: After performing the fusion operation according to the third information to fuse the third information into the graph database, selecting a fifth target information node in the graph database; wherein the fifth target information node refers to an information node that matches the attribute label of the target user; The ID value of the fifth target information node is converted into a word vector value, and the word vector value is added to the ID value of the fifth target information node.

7. A target population matching method, characterized in that: include: Search the graph database for information nodes that meet the user matching requirements; wherein the graph database includes multiple ONEID nodes and information nodes, different ONEID nodes correspond to different users, and the information nodes of the same user are connected to the same ONEID node; any information node is used to store a type of user information; the name of the ONEID node is ONEID, and the ID value is the ONEID of the corresponding user; the name of any information node is the type name of the user information, and the ID value is the corresponding user information value; the various user information stored in each information node under the same ONEID node is obtained from different channels; the various user information includes identity identification information and other information, and the other information includes at least one of data information and attribute information; The users corresponding to the target ONEID node are selected as the target population; wherein the target ONEID node refers to the ONEID node connected to the information node that meets the matching requirements of the users.

8. The target population matching method according to claim 7, characterized in that: The user matching requirement includes at least one of a target data requirement and a target attribute requirement, wherein the target data requirement refers to a condition that the data information of the target population must satisfy, and the target attribute requirement refers to a condition that the attribute information of the target population must satisfy; The searching for information nodes that meet the user's matching requirements in the graph database includes: Searching the graph database for information nodes that meet target data requirements and information nodes that meet target attribute requirements; The information nodes that meet the user's matching requirements include: at least one of the information nodes that meet the target data requirements and the information nodes that meet the target attribute requirements; Among them, the information nodes that meet the target attribute requirements are found in the following way: Convert the attribute information value of the target attribute requirement into a word vector value to obtain the requirement word vector value; Calculate the text similarity between the demand word vector value and the word vector value of the eighth target information node; wherein the eighth target information node refers to the information node whose name is the attribute information name of the target attribute demand; The eighth target information node used for calculating the target text similarity is used as the information node that meets the target attribute requirements; wherein the target text similarity is a text similarity greater than a preset threshold or the largest M text similarities among all calculated text similarities.

9. The target population matching method according to claim 8, characterized in that: The target data requirements include derived indicator requirements; The information nodes that meet the target data requirements include: information nodes that meet the derived indicator requirements; The information nodes that meet the requirements of the derived indicators are found in the following way: Determine a seventh target information node according to the data information required for calculating each derived indicator; wherein the seventh target information node refers to an information node that matches the data information required for calculating each derived indicator; Calculate the corresponding derived index value according to the ID value of the seventh target information node connected to the same ONEID node according to the set calculation rule; A seventh target information node whose derived indicator value meets the derived indicator requirement is selected as the information node that meets the derived indicator requirement.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the multi-channel user integration method described in any one of claims 1-6 or the target population matching method described in any one of claims 7-9 are implemented.

Images