User information fusion method, terminal, storage medium and system under multi-layer association

Abstract

The present invention proposes a user information fusion method, terminal, storage medium and system under multi-layer association, which solves the problem that the current user information fusion method cannot realize user information fusion under multi-layer association. First, the data source of the user information to be integrated is selected, and the data source corresponds to the material table. Further, channel type information is obtained according to the association relationship in the material table. According to the channel type information, a user information graph is constructed, and the user information graph is split into independent user connection graphs, which are expanded from linear material table relationships to two-dimensional plane graph relationships, and conveniently support multi-layer complex associations. In the complex associations of a large amount of data, records belonging to the same natural user are found; on the basis of the user connection graph, combined with the historical material table, an entity unique ID connection graph is further constructed, and with the help of a new association provided by the newly added user information data, two independent user entities in the historical material table data are fused together.

Description

Technical Field

[0001] The present invention relates to the technical field of user information fusion, and more specifically, to a user information fusion method, terminal, storage medium and system under multi-layer association. Background Art

[0002] The Customer Data Platform (CDP) contains the relevant data of the audience, users or customers generated and accumulated by the enterprise during the operation process. In CDP, member (user entity) data is spread across multiple information systems / data sources of the enterprise, such as Customer Relationship Management (CRM), POS transactions, behavior log collection data, etc. The same natural user member will have different unique identifiers in different information systems or channels. For example, there is a JD user ID and mobile phone in the purchase record of JD.com, and there is a corresponding user ID in the enterprise's own CRM. The user ID of the same natural user member in JD.com and CRM is different, which brings difficulties to subsequent data processing, labeling, marketing campaign push, etc. Moreover, a natural user may have multiple accounts, which are marked as different users in the system. Therefore, a unique identifier is needed to identify a natural user, connect user data from different systems and channels, identify the user data that is actually the same natural user behind them, and assign it an entity unique ID (the entity unique identifier after fusion) for subsequent data processing, user label system construction, user operation statistical analysis, marketing campaign push, etc.

[0003] On March 14, 2015, a Chinese invention patent (publication number: CN104394118A) disclosed a user identity recognition method and system. The basic information formed by user registration, including user ID, user name, Email, telephone number, computer IP, etc., is used to extract website user behavior data. The user ID, user name, Email, telephone number, Cookie, computer IP, etc. in the comprehensive behavior data are used to establish a user information association relationship between the two and give a unique identification identity. It can perform unified identity recognition on the current B2B website users, establish identity feature relationships, and distinguish new and old users. It can effectively track user behavior, so as to establish a series of applications for users and improve user experience. However, this patent cannot handle multi-layer associated user information fusion scenarios. For example, a user's mobile phone number A (corresponding to user ID1) purchases a coupon, and then uses the coupon to make a purchase through a backup mobile phone number B (corresponding to user ID2). Mobile phone numbers A and B were originally recorded as two different user IDs, but through this purchase behavior, it can be considered that the two user IDs belong to the same natural user; however, this method uses the user ID as the user's unique identifier, and cannot generate an association for this scenario, and cannot fuse the two user information. Summary of the invention

[0004] In order to solve the problem that the current user information fusion method cannot realize user information fusion under multi-layer association, the present invention proposes a user information fusion method, terminal, storage medium and system under multi-layer association, which associates and identifies different accounts of natural users, and then integrates and determines a unique identifier to facilitate the push processing of subsequent related marketing activities.

[0005] In order to achieve the above technical effects, the technical solution of the present invention is as follows:

[0006] A user information fusion method under multi-layer association, the method at least comprising:

[0007] S1. Select the data source of user information to be integrated. Each data source corresponds to a material table. Determine the field in the material table that can identify the user and its corresponding channel type to obtain the channel type information;

[0008] S2. Determine vertices and edges according to the channel type information, construct a user information graph based on the vertices and edges, and then split the user information graph into independent user connectivity graphs;

[0009] S3. Using the channel type association table, query the historical data of the entity unique ID corresponding to each vertex of the user connectivity graph, obtain all the entity unique IDs corresponding to the user connectivity graph, further use the entity unique ID as the vertex and the relationship between multiple entity unique IDs in the user connectivity graph as the edge to construct the entity unique ID connectivity graph, thereby associating the user connectivity graph with the historical data;

[0010] S4. Determine whether each user entity unique ID appears only in one user connectivity graph. If so, fuse user information on demand for each user connectivity graph; otherwise, construct an entity unique ID connectivity graph and execute step S5;

[0011] S5. Determine the entity unique ID connected together in the entity unique ID connectivity graph, and associate the connected entity unique ID corresponding to the user connectivity graph;

[0012] S6. Read the entity unique IDs corresponding to all vertices of the user connected graph, and perform deduplication processing on the entity unique IDs;

[0013] S7. Update the channel type association table using the entity unique ID after deduplication processing.

[0014] In this technical solution, by constructing a user connectivity graph and splitting the user connectivity graph, multi-layer complex association relationships are supported rather than fixed simple association rules. Users only need to pay attention to the association between the material tables, without having to pay attention to how all the material tables are associated with each other. It also supports the operation of newly added user information. On the basis of the user connectivity graph, combined with the channel type association table (a table that can reflect the mapping relationship between the channel type value and the entity unique ID under historical data), an entity unique ID connectivity graph is further constructed. As a result, with the help of new associations provided by the incremental data, two independent user entities considered in the historical data can be merged together. The scope of application is not limited to very specific scenarios and business logic, but can organize channel type relationships according to business reality, adapt to different scenarios, and expand from user information fusion to information fusion of goods and places.

[0015] Preferably, in the enterprise's customer data platform CDP, user information is spread across multiple data sources of the enterprise, and each data source corresponds to a material table, which includes: enterprise member information table, mall user information table, coupon usage record table; channel types in the material table include: user mobile phone number, member ID, Email, WeChat unionID and WeChat openID, and records with the same channel type value in different material tables correspond to the same natural user; the channel type field is a combination of each channel type and the material table corresponding to each channel type.

[0016] Preferably, "channel type + channel type field value" is used as the vertex attribute, and the vertex attribute is hashed to obtain the vertex ID, which records the attributes of a natural user on a certain channel type;

[0017] Multiple "channel type field values" appearing in a record in the material table are used as edges. The edges are the connections between vertices, recording the association of a natural user on different channel types.

[0018] Connect vertices and edges to build a user information graph; then use the connected component algorithm to split the user information graph, obtain the minimum vertex ID of the user information graph where each vertex is located, group and aggregate the minimum vertex IDs, and obtain all the vertices of each independent user connected graph, that is, obtain each independent user connected graph.

[0019] Here, "channel type + channel type field value" is used as the vertex attribute. There is no need to record the information of the material table in which it is located. It is simple and easy to operate. It expands from the linear material table relationship to the two-dimensional plane graph relationship, and conveniently supports multi-layer complex associations.

[0020] Preferably, if each user entity unique ID only appears in one user connectivity graph, then multiple natural users generated by the newly added user information are also independent natural users in the historical material table, and user information is integrated as needed for each user connectivity graph.

[0021] Preferably, the entity unique ID connectivity graph is constructed with each entity unique ID as a vertex and the relationship between multiple entity unique IDs appearing in the same user connectivity graph as an edge, which greatly reduces the amount of data required for calculation, thereby effectively improving calculation efficiency.

[0022] Preferably, the method for determining the entity unique IDs connected together in the entity unique ID connected graph in step S5 is a connected component algorithm; after associating the user connected graphs corresponding to the connected entity unique IDs, each user connected graph corresponds to a natural user.

[0023] Preferably, the deduplication process of the entity unique ID in step S6 includes:

[0024] a. The user connectivity graph does not have an entity unique ID: The current natural user is a new user. An entity unique ID is generated based on the user connectivity graph, and UUID is used to ensure its uniqueness;

[0025] b. The user connectivity graph has only one entity unique ID: if the old and new user information data of the current natural user belong to the same natural user, the old entity unique ID will be used;

[0026] c. There are multiple entity unique IDs in the user connectivity graph: the newly added user information generates user information fusion, and only one entity unique ID is retained.

[0027] The present invention proposes a terminal, comprising a processor, a memory and a computer program stored in the memory, wherein the processor executes the computer program stored in the memory to implement the steps of the user information fusion method under multi-layer association.

[0028] The present invention proposes a computer storage medium, wherein the computer readable storage medium stores computer program instructions, and when the instructions are executed by a processor, the steps of the user information fusion method under multi-layer association are implemented.

[0029] The present invention also proposes a user information fusion system under multi-layer association, and the system is used to implement the user information fusion method under multi-layer association, comprising:

[0030] The channel type information acquisition module selects the data source of the user information to be integrated. Each data source corresponds to a material table. The field in the material table that can identify the user and its corresponding channel type are determined to obtain the channel type information.

[0031] The user connectivity graph construction module determines vertices and edges according to the channel type information, constructs a user information graph based on the vertices and edges, and then splits the user information graph into independent user connectivity graphs;

[0032] The historical data association module uses the channel type association table to query the historical data of the entity unique ID corresponding to each vertex of the user connectivity graph, obtain all the entity unique IDs corresponding to the user connectivity graph, and further use the entity unique ID as the vertex and the relationship between multiple entity unique IDs in the user connectivity graph as the edge to construct the entity unique ID connectivity graph, thereby using the historical data to associate the user connectivity graph;

[0033] A judgment module is used to judge whether each user entity unique ID appears in only one user connectivity graph. If so, user information is integrated into each user connectivity graph as needed; otherwise, a unique entity ID connectivity graph is constructed;

[0034] A user connectivity graph association module determines the entity unique IDs connected together in the entity unique ID connectivity graph, and associates the user connectivity graphs corresponding to the connected entity unique IDs;

[0035] The deduplication processing module reads the entity unique IDs corresponding to all vertices of the user connected graph and performs deduplication processing on the entity unique IDs;

[0036] The update module updates the channel type association table using the entity unique ID after deduplication processing.

[0037] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0038] The present invention proposes a user information fusion method, terminal, storage medium and system under multi-layer association. First, a data source of user information to be integrated is selected, and the data source corresponds to a material table. Channel type information is further obtained according to the association relationship in the material table. According to the channel type information, a user information graph is constructed, and then the user information graph is split into independent user connectivity graphs, which are expanded from linear material table relationships to two-dimensional plane graph relationships, and conveniently support multi-layer complex associations. In the complex associations of a large amount of data, records belonging to the same natural user are found, which can be used for subsequent data processing, user tag system construction, user operation statistical analysis, marketing activity push and other operations; on the basis of the user connectivity graph, combined with the historical material table, an entity unique ID connectivity graph is further constructed, and with the help of a new association provided by the newly added user information data, two independent user entities in the historical material table data are fused together, which greatly reduces the amount of data required to be calculated, thereby effectively improving the calculation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A schematic diagram showing a flow chart of a method for fusing user information under multi-layer association proposed in Embodiment 1 of the present invention;

[0040] Figure 2 It represents the independent user connectivity graphs into which the user information graph is split in Embodiment 1 of the present invention;

[0041] Figure 3 It represents a user connectivity graph corresponding to the historical data in the historical material table proposed in Embodiment 1 of the present invention;

[0042] Figure 4 A schematic diagram showing the method of searching for a user entity unique ID in a channel type association table corresponding to each vertex in a current user connectivity graph proposed in Embodiment 1 of the present invention;

[0043] Figure 5 A schematic diagram showing associating user connectivity graphs corresponding to connected entity unique IDs proposed in Embodiment 1 of the present invention;

[0044] Figure 6 It represents the final user connectivity graph proposed in Embodiment 1 of the present invention;

[0045] Figure 7 A schematic diagram showing the structure of a user information fusion system under multi-layer association proposed in Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0046] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0047] It is understandable to those skilled in the art that descriptions of certain well-known contents in the drawings may be omitted.

[0048] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0049] The positional relationships described in the drawings are only for illustrative purposes and should not be construed as limiting the present patent;

[0050] Example 1

[0051] like Figure 1 As shown, in this embodiment, a user information fusion method under multi-layer association is first proposed, and the method includes:

[0052] S1. Select the data source of user information to be integrated. Each data source corresponds to a material table. Determine the field in the material table that can identify the user and its corresponding channel type to obtain the channel type information;

[0053] Taking the enterprise's customer data platform CDP as the background, user information is spread across multiple data sources of the enterprise, and each data source corresponds to a material table. The material table includes: enterprise member information table, mall user information table, and coupon usage record table; the channel types in the material table include: user mobile phone number, member ID, Email, WeChat unionID, and WeChat openID. Records with the same channel type value in different material tables correspond to the same natural user; the channel type field is a combination of each channel type and the material table corresponding to each channel type. That is, first find out which data sources in the system need to integrate information, analyze the material tables corresponding to these data sources, and find the fields that can identify users; a certain value of this field corresponds to a user, but a user can have multiple values ​​in this field in the material table, such as the ID of a coupon. A user can purchase and use multiple coupons, but a coupon will only be purchased and used by one user; such a field is a channel type field, such as coupon ID, Email, and mobile phone number. The channel type represents the scene in which the user entity has a relationship with the outside world, and determines the channel type fields that can be associated between material tables. Each such field set is recorded as a "channel type". For example, the "mobile phone number" field in the coupon usage record table, the "mobile phone number" in the user information table, and the fields representing the consumer user's mobile phone number in other material tables constitute a channel type. Similarly, there may be channel types such as ID card and Email.

[0054] S2. Determine vertices and edges according to the channel type information, construct a user information graph based on the vertices and edges, and then split the user information graph into independent user connectivity graphs;

[0055] In the same sense as data structure Figure 1Similarly, the user information graph is also composed of vertices and lines between vertices. According to the definition of channel type in step S1, records with the same channel type value in different material tables should correspond to the same natural user. For example, the record with the mobile phone number 13344445555 in the coupon usage record table and the record with the mobile phone number 13344445555 in the mall user information table belong to the same natural user. Therefore, "channel type + channel type field value" is used as the vertex attribute, and the vertex attribute is hashed to obtain the vertex ID. The vertex ID records the attributes of a natural user on a certain channel type. The hash calculation in this process is implemented using the currently disclosed MurMusHash algorithm, which is a conventional technology in this field and will not be repeated here.

[0056] Multiple "channel type field values" appearing in a record in the material table are taken as edges. The edges are the connections between vertices, which record the association of a natural user in different channel types. In addition, the edited weighted value can be the update time of this record, which can be used for subsequent user connectivity graphs to resolve conflicts.

[0057] Connect vertices and edges to build a user information graph. Ideally, an independent user connectivity graph corresponds to a unique natural user, and there is no edge connecting it to other user connectivity graphs. Use the connected component algorithm to split the user information graph. The connected component algorithm can be directly applied in relatively mature programming languages, such as SparkGraphX. The direct result of the calculation is the minimum vertex ID of the user information graph where each vertex is located. The minimum vertex ID is grouped and aggregated, which can be achieved using the GROUP function to obtain all vertices of each independent user connectivity graph, that is, the information of each independent user connectivity graph (vertices and corresponding edges) is obtained.

[0058] S3. Using the channel type association table, query the historical data of the entity unique ID corresponding to each vertex of the user connectivity graph, obtain all the entity unique IDs corresponding to the user connectivity graph, further use the entity unique ID as the vertex and the relationship between multiple entity unique IDs in the user connectivity graph as the edge to construct the entity unique ID connectivity graph, thereby associating the user connectivity graph with the historical data;

[0059] This step is the process of associating historical data. If the method proposed by the present invention is executed for the first time, there is no need to associate history and this step can be skipped. For the data source of the user information to be integrated, it is a new computing task and it can be assumed that the mapping already exists. The user connectivity graph formed by it does not represent comprehensive user channel type information. It is necessary to associate historical data to perform accurate user information fusion calculation;

[0060] S4. Determine whether each user entity unique ID appears in only one user connectivity graph. If so, fuse user information for each user connectivity graph as needed; otherwise, construct an entity unique ID connectivity graph and execute step S5; if each user entity unique ID appears in only one user connectivity graph, then the multiple natural users generated by the newly added user information are also independent natural users in the historical material table. Therefore, it is only necessary to fuse user information for each user connectivity graph as needed in the future; if multiple user connectivity graphs have the same user entity unique ID, then the multiple natural users generated by the data of the newly added user information actually belong to the same natural user in the old data and can be fused. It is necessary to connect multiple connectivity graphs with the same entity unique ID together, which can be achieved by constructing a connectivity graph of the entity unique ID.

[0061] Among them, the entity unique ID connectivity graph is constructed with each entity unique ID as a vertex and the relationship between multiple entity unique IDs in the same user connectivity graph as an edge, which greatly reduces the amount of data required for calculation, thereby effectively improving computing efficiency.

[0062] S5. Determine the entity unique ID connected together in the entity unique ID connectivity graph, and associate the connected entity unique ID corresponding to the user connectivity graph;

[0063] S6. Read the entity unique IDs corresponding to all vertices of the user connectivity graph, and deduplicate the entity unique IDs; the method for determining the entity unique IDs connected together in the entity unique ID connectivity graph described in step S5 is also a connected component algorithm, which is a type of graph algorithm. Connectivity means that in an undirected graph, if there is a path from vertex v to vertex w, then v and w are called connected. If any two vertices in graph G are connected, graph G is called a connected graph, otherwise it becomes a non-connected graph. After associating the user connectivity graphs corresponding to the connected entity unique IDs, each user connectivity graph corresponds to a natural user;

[0064] When deduplicating the entity unique ID, it includes:

[0065] a. The user connectivity graph does not have an entity unique ID: The current natural user is a new user. An entity unique ID is generated based on the user connectivity graph, and UUID is used to ensure its uniqueness. UUID is the abbreviation of Universally Unique Identifier, which is a relatively common technology available.

[0066] b. The user connectivity graph has only one entity unique ID: if the old and new user information data of the current natural user belong to the same natural user, the old entity unique ID will be used;

[0067] c. There are multiple entity unique IDs in the user connectivity graph: The newly added user information generates user information fusion, and only one entity unique ID is retained. Because the specific one saved has little impact on the calculation result, the first one is generally saved.

[0068] S7. Update the channel type association table using the entity unique ID after deduplication processing;

[0069] Specifically, traverse all vertices. The attributes of the vertex contain information about the channel type and its value. The relationship between the channel type and the entity unique ID is stored in the channel type association table. The name of the association table is channel_entity_relation_<channel type>, such as channel_entity_relation_mobile number, and the structure is:

[0070] Channel Type Field Value Entity unique ID 13344445555 33f5ec7daf694f9ca09c66767272d415 …… ……

[0071] In summary, by constructing a user connectivity graph and splitting the user connectivity graph, multi-layer complex association relationships are supported rather than fixed simple association rules. Users only need to pay attention to the associations between material tables, without having to pay attention to how all material tables are associated with each other. It also supports the operation of newly added user information. On the basis of the user connectivity graph, combined with the historical material table (historical data), an entity unique ID connectivity graph is further constructed. As a result, with the help of new associations provided by the incremental data, two independent user entities considered in the historical data can be merged together. The scope of application is not limited to very specific scenarios and business logic, but can be based on business reality. Channel type relationships can be sorted out to adapt to different scenarios, expanding from user information fusion to information fusion of goods and places.

[0072] The method proposed in this embodiment is further explained below in conjunction with a specific implementation scenario. Taking the CDP system as an example, the scenario is simplified as follows: a user receives a coupon in the membership system and then uses the coupon to shop on JD.com.

[0073] Assume that the data source contains only three material tables: coupon usage record table, JD user information table, and internal member information table.

[0074] Analyze the association relationship of the above material table and express it in the form of channel type and channel type field table, as shown in Table 1.

[0075] Table 1

[0076]

[0077] Assume that the coupon usage record table has the records shown in Table 2.

[0078] Table 2

[0079] Phone number Coupon Number Member ID …… 13344445555 T00003 C001 13455556666 T00004 C001 13344445555 T00005 C001 13566667777 T00006 C002

[0080] The JD user information table has the records shown in Table 3.

[0081] Table 3

[0082] JD User ID Phone number …… JD001 13344445555 JD002 13455556666 JD003 13566667777

[0083] The internal member information table has the records shown in Table 4.

[0084] Table 4

[0085]

[0086]

[0087] All channel types are: "Mobile phone number", "Member ID" and "JD user ID".

[0088] According to the sorted channel type information, the user information graph is constructed and split into independent user connectivity graphs. According to the material table information, the vertices of the user information graph are calculated, as shown in Table 5. Here, for the convenience of example description and to improve the recognition of the example, relatively simple numerical values ​​are used to mention the actual vertex attribute hash values.

[0089] Table 5

[0090] Vertex ID Vertex attributes 1111 [Mobile, 13344445555] 2222 [Mobile, 13455556666] 3333 [Mobile_13566667777] 4444 [Member ID_, C001] 5555 [Member ID_, C002] 6666 [JD user ID, JD001] 7777 [JD user ID, JD002] 8888 [JD user ID, JD003]

[0091] The minimum vertex ID of the connected graph where each vertex is located is calculated using the connected component algorithm. The calculation results are shown in Table 6:

[0092] Table 6

[0093] Vertex ID The minimum vertex ID of the connected graph Vertex attributes 1111 1111 [Mobile, 13344445555] 2222 1111 [Mobile, 13455556666] 3333 3333 [Mobile, 13566667777] 4444 1111 [Member ID, C001] 5555 3333 [Member ID, C002] 6666 1111 [JD user ID, JD001] 7777 1111 [JD user ID, JD002] 8888 3333 [JD user ID, JD003]

[0094] According to the above results, the user connectivity graph is obtained as follows: Figure 2 As shown by Figure 2 It can be seen that according to the "Minimum Vertex ID of the Connected Graph" column, the entire user information graph is split into two independent user connected graphs, corresponding to two natural users, including the user connected Figure 1 Connect with users Figure 2 .

[0095] Assume that this calculation is not the first one, the historical data in the historical material table has been calculated, and a channel type association table has been generated. The historical data is the coupon usage record table with the records shown in Table 7:

[0096] Table 7

[0097] Phone number Coupon Number Member ID …… 13566667777 T00001 C001 13344445555 T00002 C002

[0098] The result of the last calculation is Figure 3 As shown, the two natural unique IDs corresponding to the records shown in Table 7 are obtained, so the content of the "mobile phone number" channel type association table is shown in Table 8.

[0099] Table 8

[0100] Phone number Entity unique ID 13566667777 5109b4d4e0e7411c992da7d5ea112538 13344445555 73c823904daa4c20a4c2035132d0e7c2

[0101] The contents of the “Member ID” channel type association table are shown in Table 9.

[0102] Table 9

[0103] Member ID Entity unique ID C001 5109b4d4e0e7411c992da7d5ea112538 C002 73c823904daa4c20a4c2035132d0e7c2

[0104] At this time, all vertex attributes of each user connectivity graph are queried in the corresponding channel type association table to obtain the entity unique ID and recorded as the attributes of the user connectivity graph, such as Figure 4 As shown in the figure, the entity unique ID connectivity graph is constructed based on the entity unique ID, and the corresponding user connectivity graph is connected to obtain a new user connectivity graph as shown in the figure. Figure 5 shown.

[0105] After associating the historical data in the historical material table, since some vertices of the two new user connection graphs belong to the same user connection graph in the historical data, the two can be associated together, such as Figure 5 As shown in the figure, all the data currently constitutes the same natural user, whose entity unique ID, after deduplication, contains two:

[0106] 5109b4d4e0e7411c992da7d5ea112538

[0107] 73c823904daa4c20a4c2035132d0e7c2

[0108] It is believed that in this calculation, due to the new user coupon usage record, the two independent users in the old user information are associated and are the same user, and user information fusion has occurred.

[0109] At this time, only one unique user ID is reserved, such as: 5109b4d4e0e7411c992da7d5ea112538. Figure 6 shown.

[0110] Finally, the channel type association table is updated to facilitate subsequent calculations. The updated channel type association table includes the "mobile phone number" channel type association table shown in Table 10, the "member ID" channel type association table shown in Table 11, and the "JD user ID" channel type association table shown in Table 12.

[0111] The subsequent use of CDP data is based on the entity unique ID. For example, to calculate the user portrait, the entity unique ID is calculated to obtain the user (entity) portrait of each entity unique ID; if you want to do marketing activities in the membership system later, you can filter the population based on the user portrait, and get a series of entity unique IDs corresponding to the user portrait population. Then, you can find the corresponding JD user ID based on the association table of the JD user ID channel, so as to carry out marketing activities and other contact operations.

[0112] Example 2

[0113] The present invention proposes a terminal, comprising a processor, a memory and a computer program stored in the memory, wherein the processor executes the computer program stored in the memory to implement the steps of the user information fusion method under multi-layer association described in Example 1, wherein the memory can be a disk, a flash memory or any other non-volatile storage medium, and the processor is connected to the memory and can be implemented as one or more integrated circuits, specifically a microprocessor or a microcontroller, and when executing the computer program stored in the memory, the user information fusion under multi-layer association is implemented.

[0114] The present invention proposes a computer storage medium, wherein the computer-readable storage medium stores computer program instructions, and when the instructions are executed by a processor, the steps of the user information fusion method under multi-layer association described in Example 1 are implemented.

[0115] Example 2

[0116] See also Figure 7 The present invention further proposes a user information fusion system under multi-layer association, the system is used to implement the user information fusion method under multi-layer association described in Example 1, including:

[0117] The channel type information acquisition module selects the data source of the user information to be integrated, each data source corresponds to a material table, determines the channel type and channel type field that identify the association relationship in the material table, and obtains the channel type information;

[0118] The user connectivity graph construction module determines vertices and edges according to the channel type information, constructs a user information graph based on the vertices and edges, and then splits the user information graph into independent user connectivity graphs;

[0119] The historical data association module is used to collect the historical material table, determine the mapping between each channel type in the historical material table and the unique ID of the user entity, form an initial channel type association table, and further find the user entity unique ID in the channel type association table corresponding to each vertex of the user connectivity graph;

[0120] A judgment module is used to judge whether each user entity unique ID appears in only one user connectivity graph. If so, user information is integrated into each user connectivity graph as needed; otherwise, a unique entity ID connectivity graph is constructed;

[0121] A user connectivity graph association module determines the entity unique IDs connected together in the entity unique ID connectivity graph, and associates the user connectivity graphs corresponding to the connected entity unique IDs;

[0122] The deduplication processing module reads the entity unique IDs corresponding to all vertices of the user connected graph and performs deduplication processing on the entity unique IDs;

[0123] The update module updates the channel type association table using the entity unique ID after deduplication processing.

[0124] Obviously, the above embodiments of the present invention are only examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A user information fusion method under multi-layer association, characterized in that: The method at least comprises: S1. Select the data source of user information to be integrated. Each data source corresponds to a material table. Determine the field in the material table that can identify the user and its corresponding channel type to obtain the channel type information; In the enterprise's customer data platform CDP, user information is spread across multiple data sources of the enterprise. Each data source corresponds to a material table, which includes: enterprise member information table, mall user information table, and coupon usage record table; channel types in the material table include: user mobile phone number, member ID, Email, WeChat unionID, and WeChat openID. Records with the same channel type value in different material tables correspond to the same natural user; the channel type field is a combination of each channel type and the material table corresponding to each channel type; S2. Determine vertices and edges according to the channel type information, construct a user information graph based on the vertices and edges, and then split the user information graph into independent user connectivity graphs; In step S2, "channel type + channel type field value" is used as the vertex attribute, and the vertex attribute is hashed to obtain the vertex ID. The vertex ID records the attributes of a natural user on a certain channel type; Multiple "channel type field values" appearing in a record in the material table are used as edges. The edges are the connections between vertices, recording the association of a natural user on different channel types. Connect vertices and edges to build a user information graph; then use the connected component algorithm to split the user information graph, obtain the minimum vertex ID of the user information graph where each vertex is located, group and aggregate the minimum vertex IDs, and obtain all the vertices of each independent user connected graph, that is, obtain each independent user connected graph; S3. Using the channel type association table, query the historical data of the entity unique ID corresponding to each vertex of the user connectivity graph, obtain all the entity unique IDs corresponding to the user connectivity graph, further use the entity unique ID as the vertex and the relationship between multiple entity unique IDs in the user connectivity graph as the edge to construct the entity unique ID connectivity graph, so as to associate the user connectivity graph with the historical data; S4. Determine whether each user entity unique ID appears only in one user connectivity graph. If so, fuse user information on demand for each user connectivity graph; otherwise, construct an entity unique ID connectivity graph and execute step S5; S5. Determine the entity unique ID connected together in the entity unique ID connectivity graph, and associate the connected entity unique ID corresponding to the user connectivity graph; S6. Read the entity unique IDs corresponding to all vertices of the user connected graph, and perform deduplication processing on the entity unique IDs; S7. Update the channel type association table using the entity unique ID after deduplication processing.

2. The user information fusion method under multi-layer association according to claim 1 is characterized in that: In step S4, if each user entity unique ID only appears in one user connectivity graph, then the multiple natural users generated by the newly added user information are also independent natural users in the historical material table, and user information is integrated as needed for each user connectivity graph.

3. The user information fusion method under multi-layer association according to claim 2 is characterized in that: The entity unique ID connectivity graph is constructed with each entity unique ID as a vertex and with the relationship between multiple entity unique IDs appearing in the same user connectivity graph as an edge.

4. The user information fusion method under multi-layer association according to claim 3 is characterized in that: The method for determining the entity unique IDs connected together in the entity unique ID connected graph in step S5 is a connected component algorithm; after associating the user connected graphs corresponding to the connected entity unique IDs, each user connected graph corresponds to a natural user.

5. The user information fusion method under multi-layer association according to claim 4 is characterized in that: The deduplication process of the entity unique ID described in step S6 includes: a. The user connectivity graph does not have an entity unique ID: The current natural user is a new user. An entity unique ID is generated based on the user connectivity graph, and UUID is used to ensure its uniqueness; b. The user connectivity graph has only one entity unique ID: if the old and new user information data of the current natural user belong to the same natural user, the old entity unique ID will be used; c. There are multiple entity unique IDs in the user connectivity graph: the newly added user information generates user information fusion, and only one entity unique ID is retained.

6. A terminal, characterized in that: The invention comprises a processor, a memory and a computer program stored in the memory, wherein the processor executes the computer program stored in the memory to implement the steps of the method for fusion of user information under multi-layer association as described in any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the steps of the method for fusion of user information under multi-layer association as described in any one of claims 1 to 5.

8. A user information fusion system under multi-layer association, the system is used to implement the user information fusion method under multi-layer association according to claim 1, characterized in that: include: The channel type information acquisition module selects the data source of the user information to be integrated. Each data source corresponds to a material table. The field in the material table that can identify the user and its corresponding channel type are determined to obtain the channel type information. The user connectivity graph construction module determines vertices and edges according to the channel type information, constructs a user information graph based on the vertices and edges, and then splits the user information graph into independent user connectivity graphs; The historical data association module uses the channel type association table to query the historical data of the entity unique ID corresponding to each vertex of the user connectivity graph, obtain all the entity unique IDs corresponding to the user connectivity graph, and further use the entity unique ID as the vertex and the relationship between multiple entity unique IDs in the user connectivity graph as the edge to construct the entity unique ID connectivity graph, thereby using the historical data to associate the user connectivity graph; A judgment module is used to judge whether each user entity unique ID appears in only one user connectivity graph. If so, user information is integrated into each user connectivity graph as needed; otherwise, a unique entity ID connectivity graph is constructed; A user connectivity graph association module determines the entity unique IDs connected together in the entity unique ID connectivity graph, and associates the user connectivity graphs corresponding to the connected entity unique IDs; The deduplication processing module reads the entity unique IDs corresponding to all vertices of the user connected graph and performs deduplication processing on the entity unique IDs; The update module updates the channel type association table using the entity unique ID after deduplication processing.

Images