User identity recognition method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a user identity recognition method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining at least one piece of identification data reported by at least one piece of terminal equipment, and carrying out the recognition of a user identity according to an association graph formed by an account identifier, an equipment identifier, a global wide area network identifier, a mobile phone identifier and a joint identifier; and determining the identification values with the association relationship in the at least one piece of identification data, and calculating the confidence degree between the identification values with the association relationship. Further, according to the identification values with the incidence relation and the confidence coefficient, at least one connected graph is generated, and for each connected graph, if the number of nodes in the connected graph is smaller than or equal to a preset number, a natural person identification value corresponding to the connected graph is generated; namely, unique identification is carried out on the identification values belonging to the same user and the confidence among the identification values through the natural person identification values, so that different identification values belonging to the same user are aggregated, and the same user is prevented from being mistaken as a plurality of independent individuals.

Description

Technical Field

[0001] The present disclosure relates to the field of information technology, and in particular, to a user identity recognition method, apparatus, electronic device, and storage medium. Background Art

[0002] With the development of information technology, users can install various application programs (APPs) with different functions on terminal devices, such as APPs providing automotive information, APPs providing instant messaging services, APPs providing short video services, etc.

[0003] However, the identities (IDs) corresponding to the same user on different APPs are different, resulting in the same user often being misjudged as multiple independent individuals, and thus accurate services cannot be provided for the user. Summary of the Invention

[0004] In order to solve the above technical problems or at least partially solve the above technical problems, embodiments of the present disclosure provide a user identity recognition method, apparatus, electronic device, and storage medium to provide accurate services for users.

[0005] Embodiments of the present disclosure provide a user identity recognition method, which includes:

[0006] Obtaining at least one piece of identification data respectively reported by at least one terminal device, each piece of identification data including at least one identification value;

[0007] According to the association graph composed of an account identifier, a device identifier, a global wide area network identifier, a mobile phone identifier, and a combined identifier, determining the identification values with an association relationship in the at least one piece of identification data respectively reported by the at least one terminal device, and calculating the confidence level between the identification values with the association relationship;

[0008] Generating at least one connected graph according to the identification values with the association relationship and the confidence level between the identification values with the association relationship, where the nodes in the connected graph represent the identification values, and the edges in the connected graph represent the confidence level;

[0009] For each connected graph in the at least one connected graph, if the number of nodes in the connected graph is less than or equal to a preset number, generating a natural person identification value corresponding to the connected graph.

[0010] Embodiments of the present disclosure further provide a user identity recognition apparatus, which includes:

[0011] An obtaining module, configured to obtain at least one piece of identification data respectively reported by at least one terminal device, each piece of identification data including at least one identification value;

[0012] A determination module, configured to determine, according to an association graph formed by an account identifier, a device identifier, a WAN identifier, a mobile phone identifier, and a combined identifier, identifier values having an association relationship in at least one identifier data respectively reported by the at least one terminal device, and calculate a confidence level between the identifier values having the association relationship;

[0013] A first generation module, configured to generate at least one connected graph according to the identifier values having the association relationship and the confidence level between the identifier values having the association relationship, where nodes in the connected graph represent the identifier values, and edges in the connected graph represent the confidence level;

[0014] A second generation module, configured to, for each of the at least one connected graph, if the number of nodes in the connected graph is less than or equal to a preset number, generate a natural person identifier value corresponding to the connected graph.

[0015] An embodiment of the present disclosure further provides an electronic device, where the electronic device includes:

[0016] One or more processors;

[0017] A storage device, configured to store one or more programs;

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement the user identity recognition method as described above.

[0019] An embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the user identity recognition method as described above is implemented.

[0020] The technical solution provided by the embodiment of the present disclosure has at least the following advantages compared with the prior art:

[0021] The user identity recognition method provided by the embodiments of the present disclosure obtains at least one piece of identification data reported by at least one terminal device respectively, determines the identification values with an association relationship in the at least one piece of identification data according to an association map composed of an account identifier, a device identifier, a global wide area network identifier, a mobile phone identifier, and a combined identifier, and calculates the confidence level between the identification values with the association relationship. Further, according to the identification values with the association relationship and the confidence level between the identification values with the association relationship, at least one connected graph is generated. For each connected graph, if the number of nodes in the connected graph is less than or equal to a preset number, a natural person identification value corresponding to the connected graph is generated, that is, the identification values belonging to the same user and the confidence level between the identification values are uniquely identified by the natural person identification value, so as to aggregate different identification values belonging to the same user and avoid the same user being misidentified as multiple independent individuals. In addition, after aggregating different identification values belonging to the same user, more information related to the user can be obtained according to the multiple identification values of the user, so as to provide accurate services for the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In combination with the accompanying drawings and with reference to the following specific embodiments, the above and other features, advantages and aspects of the embodiments of the present disclosure will become more obvious. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the original and elements are not necessarily drawn to scale.

[0023] Figure 1 is a flowchart of a user identity recognition method in an embodiment of the present disclosure;

[0024] Figure 2 is a schematic diagram of an application scenario in an embodiment of the present disclosure;

[0025] Figure 3 is a schematic diagram of an association map in an embodiment of the present disclosure;

[0026] Figure 4 is a flowchart of another user identity recognition method in an embodiment of the present disclosure;

[0027] Figure 5 is a flowchart of another user identity recognition method in an embodiment of the present disclosure;

[0028] Figure 6 is a flowchart of another user identity recognition method in an embodiment of the present disclosure;

[0029] Figure 7 is a flowchart of another user identity recognition method in an embodiment of the present disclosure;

[0030] Figure 8Flowchart of another user identity recognition method in the embodiments of the present disclosure;

[0031] Figure 9 Flowchart of another user identity recognition method in the embodiments of the present disclosure;

[0032] Figure 10 Flowchart of another user identity recognition method in the embodiments of the present disclosure;

[0033] Figure 11 Overall architecture diagram in the embodiments of the present disclosure;

[0034] Figure 12 Structural schematic diagram of a user identity recognition device in the embodiments of the present disclosure;

[0035] Figure 13 Structural schematic diagram of an electronic device in the embodiments of the present disclosure. Detailed implementation manners

[0036] The embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0037] It should be understood that the various steps recorded in the method embodiments of the present disclosure can be executed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this regard.

[0038] The term "including" and its variants used herein are open-ended, that is, "including but not limited to". The term "based on" is "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0039] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order of functions executed by these devices, modules or units or their interdependent relationships.

[0040] It should be noted that the modifiers "one" and "multiple" mentioned in this disclosure are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise clearly specified in the context, it should be understood as "one or more".

[0041] The names of the messages or information exchanged between multiple devices in the embodiments of this disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0042] Figure 1 FIG. is a flowchart of a user identity recognition method in an embodiment of this disclosure. This method can be executed by a user identity recognition device, which can be implemented in software and / or hardware and can be configured in a server. As Figure 2 shown, in this application scenario, the server 21 can communicate with multiple terminal devices. For example, the multiple terminal devices include, as Figure 2 shown, the terminal device 22, the terminal device 23, and the terminal device 24. Among them, the terminal device 22, the terminal device 23, and the terminal device 24 are not limited to smart phones, personal digital assistants, tablet computers, wearable devices with a display screen, desktop computers, laptop computers, all-in-one computers, smart home devices, etc. In addition, the multiple terminal devices are not limited to Figure 2 the terminal device 22, the terminal device 23, and the terminal device 24 shown, and may also include more terminal devices.

[0043] As Figure 1 shown, this method can specifically include the following steps:

[0044] S101. Obtain at least one piece of identification data reported by each of at least one terminal device, where each piece of identification data includes at least one identification value.

[0045] The embodiments of this disclosure involve multiple different identifiers and the corresponding identification values for each identifier. Among them, the multiple different identifiers are abstract concepts, and the corresponding identification values for each identifier are specific numerical values. For example, identifiers such as account identifier, device identity (DID), web identity (web-ID), mobile identity (MID), union identity (Union-ID), and session identity (SID) are abstract concepts. Identification values such as account identifier value, device identity value, web identity value, mobile identity value, union identity value, and session identity value are specific numerical values, also called actual numerical values. For example, the account identifier value is the actual numerical value or specific numerical value corresponding to the abstract concept of the account identifier.

[0046] As Figure 2As shown, the terminal devices 22, 23, and 24 can be terminal devices belonging to the same user or terminal devices belonging to different users. For example, the terminal devices 22, 23, and 24 belong to different users respectively. Specifically, the terminal device 22 belongs to user A, the terminal device 23 belongs to user B, and the terminal device 24 belongs to user C.

[0047] Taking the terminal device 22 as an example, various application programs (Apps) with different functions can be installed on the terminal device 22, such as an App providing automotive information, an App providing instant messaging services, an App providing short video services, etc. Suppose the App providing automotive information is denoted as App1, the App providing instant messaging services is denoted as App2, and the App providing short video services is denoted as App3. User A can operate on various Apps on the terminal device 22. For example, user A can perform user behaviors such as user registration, login, browsing, coupon collection, leaving contact information (abbreviated as leaving data), test drive reservation, purchasing goods (such as purchasing a car), vehicle owner authentication, car wash and maintenance reservation on various Apps. The terminal device 22 can record user behaviors and report user behavior information to the server 21. The user behavior information includes the type of user behavior and the identification data corresponding to the user behavior. Specifically, the code corresponding to the App in the terminal device 22 is the front-end code, and the front-end code can be preset with buried point codes, or the service platform corresponding to the App can embed buried point codes in the front-end code. The buried point codes can be used to collect user behavior information and report user behavior information to the server 21. Specifically, in the buried point codes, each user behavior can correspond to a buried point field, a reporting trigger condition, and the collected identification data. Among them, the reporting trigger condition can be the trigger condition for reporting user behavior information. For example, for any user behavior, the corresponding reporting trigger condition can be reporting when the user opens the relevant page, or reporting after the user finishes the relevant operation, or reporting after the user closes the relevant page. In addition, the identification data is also called ID data, and the identification data includes at least one of an account identification value, a device identification value, a global wide area network identification value, a mobile phone identification value, a combined identification value, and a session identification value.

[0048] In a feasible implementation, the identification data may fixedly include a device identification value. If the user registers on the APP, the identification data may further include an account identification value and a mobile phone identification value. Additionally, if the user has a session with the APP, for example, the user and the APP have an intelligent Q&A session, then the identification data may further include a session identification value. Moreover, the APP providing instant messaging services contains a variety of different mini-programs. For example, the APP providing instant messaging services contains a mini-program providing automotive information. That is to say, users can not only learn about automotive information through the APP providing automotive information, but also learn about automotive information through the mini-program providing automotive information contained in the APP providing instant messaging services. The identification values assigned to the same user by the APP and the mini-program are different. For example, the identification value assigned to the user by the APP providing automotive information is the account identification value, and the identification value assigned to the user by the mini-program providing automotive information is the global wide area network identification value. Specifically, if the user browses automotive information through the mini-program providing automotive information, the identification data includes the global wide area network identification value and the combined identification value. If the user browses automotive information through the APP providing automotive information, the identification data includes the account identification value and the device identification value.

[0049] It can be understood that since the user behavior of the same user on the same terminal device changes in real time, therefore, the same terminal device can collect and report user behavior information in real time. Since the user behavior information includes identification data corresponding to the user behavior, therefore, it is equivalent to that the same terminal device can collect and report identification data corresponding to the user behavior in real time. That is to say, the same terminal device can report at least one piece of identification data to the server 21. Or, in some embodiments, the same terminal device can collect identification data corresponding to the user behavior in real time and directly report the identification data to the server 21.

[0050] As Figure 2 shown, the server 21 can communicate with multiple terminal devices. Therefore, the terminal device 22, the terminal device 23, and the terminal device 24 can respectively report the identification data corresponding to the user behavior to the server 21 in real time. Since the user behaviors collected by different terminal devices are different, the identification data reported by different terminal devices to the server 21 in real time are also different. That is to say, the server 21 can receive at least one piece of identification data reported by at least one terminal device respectively.

[0051] In addition, to ensure the security of the identification data, each terminal device can use the Hypertext Transfer Protocol Secure (HTTPS) to encrypt and transmit the collected user behavior information to the server 21, which can be a data middle platform. In addition, after the server 21 receives the user behavior information reported by each terminal device, it can further perform format verification on the user behavior information to query whether the user behavior information conforms to the predefined format specification, and store the verified user behavior information. The storage format includes key information such as identification data, collection time, and collection platform.

[0052] Since the server 21 can receive a large amount of identification data, it is necessary to exclude outliers and duplicate data from the large amount of identification data to ensure the consistency and accuracy of the identification data. Specifically, the server 21 can clean the large amount of identification data according to the preset cleaning rules.

[0053] In one possible implementation, the server 21 analyzes the large amount of identification data received within a preset duration. If there are multiple account identification values corresponding to the same device identification value, and the number of the multiple account identification values is greater than or equal to the first preset number, the corresponding identification data is deleted. For example, the server 21 receives 100 identification data within a preset duration, and each identification data includes a device identification value and an account identification value. Among them, more than 50 identification data include the same device identification value, and the account identification values included in the more than 50 identification data are all different, that is, the same device has started more than 50 account identification values. Therefore, the more than 50 identification data are deleted.

[0054] In another possible implementation, the server 21 analyzes the large amount of identification data received within a preset duration. If there are multiple mobile phone identification values corresponding to the same account identification value, and the number of the multiple mobile phone identification values is greater than or equal to the second preset number, the corresponding identification data is deleted. For example, the server 21 receives 100 identification data within a preset duration. Among them, more than 3 identification data respectively include an account identification value and a mobile phone identification value, and the account identification values included in the more than 3 identification data are the same, but the mobile phone identification values are different, that is, the same account identification value is bound to more than 3 mobile phone identification values, then the more than 3 identification data are deleted.

[0055] In yet another possible implementation, the server 21 analyzes a large amount of identification data received within a preset duration. If multiple mobile identification values correspond to the same device identification value and the number of these multiple mobile identification values is greater than or equal to a third preset number, for example, if the same device is bound to more than 4 mobile identification values, the corresponding identification data is deleted. Additionally, if multiple device identification values correspond to the same mobile identification value and the number of these multiple device identification values is greater than or equal to a fourth preset number, for example, if the same mobile identification value is used by more than 4 devices, the corresponding identification data is deleted. The device in the embodiments of the present disclosure may specifically be a terminal device, such as a mobile phone.

[0056] S102. Determine the identification values with an association relationship in the at least one piece of identification data respectively reported by the at least one terminal device according to the association graph composed of the account identification, device identification, global wide area network identification, mobile identification, and combined identification, and calculate the confidence level between the identification values with the association relationship.

[0057] In the embodiments of the present disclosure, there is an association relationship between different identifications in the account identification, device identification (Device Identity, DID), global wide area network identification (web Identity, web-ID), mobile identification (Mobile Identity, MID), and combined identification (UnionIdentity, Union-ID), and the association relationships between different identifications can form an association graph as shown in Figure 3 shown, where the account identification may also be denoted as the user identification (User Identity, UID). The device identification may be the identification of a terminal device, and the device identification is generated according to information such as the factory information of the terminal device and the media access control address (MediaAccess Control Address, MAC). The mobile identification is a mapping of the mobile phone number to keep the user's mobile phone number confidential. The combined identification may be the same identification corresponding to different applications under the instant messaging open platform. For example, the mobile application, website application, official account, mini-program, etc. under the instant messaging open platform correspond to the same combined identification.

[0058] As Figure 3As shown, there is a binding relationship between the account identifier (UID) and the mobile phone identifier (MID). There is an APP startup relationship between the account identifier (UID) and the device identifier (DID), that is, when the APP is started on the device, the identifier data reported by the device includes the account identifier value and the device identifier value. The association relationship between the device identifier (DID) and the mobile phone identifier (MID) is obtained indirectly through the association relationship between UID and MID, and the association relationship between UID and DID. The relationship between the account identifier (UID) and the third-party account identifier (UID) is the relationship of third-party account binding. For example, the account identifier (UID) is the account identifier of the user in the APP that provides automotive information, and the third-party account identifier (UID) is the account identifier of the user in the APP that provides instant messaging services. The APP that provides automotive information can authorize the user to log in through the third-party account identifier. Therefore, when the user logs in to the APP that provides automotive information through the third-party account identifier (UID), there is a binding relationship between the account identifier (UID) and the third-party account identifier (UID). The association relationship between the account identifier (UID) and the account identifier (UID) refers to the association relationship between the account identifiers of the user on different APPs within the same system. The same system can refer to the same company. For example, the same company has developed multiple APPs, and the same user has registered accounts on these multiple APPs. Then, the association relationship between the account identifiers registered by the same user on any two different APPs among these multiple APPs is as Figure 3 shown in the association relationship between the account identifier (UID) and the account identifier (UID). Specifically, the account identifiers of the user on different APPs within the same system can be associated through the same DID. In addition, when the identifier data reported by the device includes the global wide area network identifier value, there is also an association relationship between the web-ID and the DID. In addition, since the union identifier value is carried simultaneously when the device reports the global wide area network identifier value, there is a binding relationship between the web-ID and the Union-ID. It can be understood that the association graph as Figure 3 shown is an abstract concept, that is, the various different identifiers as Figure 3 shown are abstract concepts, while the different identifier values included in the identifier data reported by the terminal device are specific numerical values. That is to say, even if the server 21 does not receive the identifier data reported by each terminal device, this association graph still exists. In addition, in the association graph as Figure 3 shown, the association relationship between different identifiers can also be marked. For example, the association relationship between the account identifier and the mobile phone identifier can be marked as a binding relationship. Therefore, when the server 21 receives the identifier data reported by each terminal device, according to as Figure 3For the associated graph shown, determine which identification values in the identification data reported by each terminal device are associated, and calculate the confidence level between the associated identification values.

[0059] In the embodiments of the present disclosure, if there is a binding relationship, a fixed collocation relationship, or a relationship of co-occurrence between two identification values, the confidence level between the two identification values is 1; otherwise, the confidence level between the two identification values needs to be calculated separately. For example, the identification data reported by terminal device 22 includes an account identification value and a mobile phone identification value. The server 21 queries according to the Figure 3 associated graph shown and determines that the account identification and the mobile phone identification are in a binding relationship, then determines that the confidence level between the account identification value and the mobile phone identification value in the identification data reported by terminal device 22 is 1.

[0060] It can be understood that the associated identification values are not limited to appearing in one piece of identification data. For example, there may also be associated identification values in different pieces of identification data reported by the same terminal device. There may also be associated identification values in different pieces of identification data reported by different terminal devices.

[0061] For example, the identification data reported by terminal device 22 at the first time includes an account identification value (e.g., x) and a mobile phone identification value (e.g., y). At this time, the server 21 can determine that the account identification value (e.g., x) and the mobile phone identification value (e.g., y) are associated identification values, and calculate the confidence level between the account identification value (e.g., x) and the mobile phone identification value (e.g., y). The identification data reported by terminal device 22 at the second time includes an account identification value (e.g., x) and a device identification value (e.g., z). At this time, the server 21 can not only determine that the account identification value (e.g., x) and the device identification value (e.g., z) are associated identification values, and calculate the confidence level between the account identification value (e.g., x) and the device identification value (e.g., z), but also according to the Figure 3 associated graph shown, determine that the mobile phone identification value (e.g., y) and the device identification value (e.g., z) are also associated identification values, and calculate the confidence level between the mobile phone identification value (e.g., y) and the device identification value (e.g., z).

[0062] S103. Generate at least one connected graph according to the associated identification values and the confidence level between the associated identification values, where the nodes in the connected graph represent the identification values, and the edges in the connected graph represent the confidence level.

[0063] Specifically, when the server 21 receives the identification data, it processes the identification data through an algorithm to obtain identification values with an association relationship. Further, a connectivity graph is constructed based on the identification values with an association relationship and the confidence levels between the identification values with an association relationship. In theory, the identification values belonging to the same user will be aggregated together to form a connectivity graph. Therefore, as the number of identification data from different users increases, the number of connectivity graphs will also increase. Specifically, the nodes of the connectivity graph represent the identification values with an association relationship, and the edges of the connectivity graph represent the confidence levels between the identification values with an association relationship.

[0064] S104. For each of the at least one connectivity graph, if the number of nodes in the connectivity graph is less than or equal to a preset number, a natural person identification value corresponding to the connectivity graph is generated.

[0065] For example, after generating multiple connectivity graphs according to the algorithm, the multiple connectivity graphs are further traversed. If the number of nodes in the currently traversed connectivity graph is less than or equal to the preset number, it is determined that the connectivity graph belongs to the data of the same user. Therefore, a natural person identification value is assigned to the connectivity graph, and the natural person identification values of each connectivity graph are unique. If the number of nodes in the currently traversed connectivity graph is greater than the preset number, the large connectivity graph needs to be split into multiple subgraphs, and then a natural person identification value is assigned to each subgraph. Similarly, the natural person identification values of each subgraph are also unique. That is to say, in this embodiment, the identification values belonging to the same user and the confidence levels between the identification values are uniquely identified by the natural person identification value.

[0066] The user identity recognition method provided by the embodiments of the present disclosure obtains at least one piece of identification data reported by at least one terminal device respectively, determines the identification values with an association relationship in the at least one piece of identification data according to the association graph composed of the account identification, device identification, global wide area network identification, mobile phone identification, and combined identification, and calculates the confidence levels between the identification values with an association relationship. Further, at least one connectivity graph is generated according to the identification values with an association relationship and the confidence levels between the identification values with an association relationship. For each connectivity graph, if the number of nodes in the connectivity graph is less than or equal to a preset number, a natural person identification value corresponding to the connectivity graph is generated, that is, the identification values belonging to the same user and the confidence levels between the identification values are uniquely identified by the natural person identification value, so as to aggregate different identification values belonging to the same user and avoid the same user being misidentified as multiple independent individuals. In addition, after aggregating different identification values belonging to the same user together, more information related to the user can be obtained according to the multiple identification values of the user, so as to provide accurate services for the user.

[0067] Optionally, according to the association graph composed of the account identifier, device identifier, global wide area network identifier, mobile phone identifier, and combined identifier, determine the identifier values with an association relationship among the at least one identifier data respectively reported by the at least one terminal device, and calculate the confidence level between the identifier values with an association relationship, including: according to the association relationship between the account identifier and the device identifier in the association graph, when it is determined that there are multiple device identifier values corresponding to the same account identifier value among the at least one identifier data respectively reported by the at least one terminal device, calculate the confidence level between the same account identifier value and each device identifier value among the multiple device identifier values.

[0068] Specifically, the same user may log in to the same account identifier value on different devices. For example, when the user logs in to the account identifier value on Device 1, the identifier data reported by Device 1 to Server 21 includes the account identifier value and the device identifier value of Device 1. When the user logs in to the account identifier value on Device 2, the identifier data reported by Device 2 to Server 21 includes the account identifier value and the device identifier value of Device 2. That is, the same account identifier value corresponds to multiple device identifier values. In this case, it is necessary to calculate the confidence level between the same account identifier value and each device identifier value. For example, taking days as the granularity, calculate the number of days the account identifier value is logged in on Device 1 and the number of days it is logged in on Device 2. According to the number of days the account identifier value is logged in on Device 1, calculate the confidence level between the account identifier value and the device identifier value of Device 1. According to the number of days the account identifier value is logged in on Device 2, calculate the confidence level between the account identifier value and the device identifier value of Device 2. Since the number of days the account identifier value is logged in on Device 1 or Device 2 changes, the confidence level also changes with the number of days. That is, in the embodiments of the present disclosure, the confidence level is calculated taking days as the granularity.

[0069] In addition, the higher the confidence level, the more stable the association relationship. Specifically, the confidence level is a value between 0 and 1. When the account identifier value is only logged in on one device and not on other devices, the confidence level is 1. In addition, as long as the account identifier value has been logged in on one device, the confidence level between the account identifier value and the device identifier value of that device is greater than 0.

[0070] Optionally, the multiple device identifier values include a first device identifier value and a second device identifier value; the confidence level between the same account identifier value and the first device identifier value increases as the number of days the same account identifier value is logged in on the first device increases; when the same account identifier value changes to be logged in on the second device, the confidence level between the same account identifier value and the first device identifier value decreases, and the confidence level between the same account identifier value and the second device identifier value increases as the number of days the same account identifier value is logged in on the second device increases.

[0071] For example, a positive event can trigger an increase in the confidence value, and a negative event can trigger a decrease in the confidence value. Among them, a positive event is an increase in the number of login days, and a negative event can be that the number of login days no longer increases. For example, when the number of login days of the account identification value increases on device 1, the confidence between the account identification value and the device identification value of device 1 increases. When the account identification value changes to log in on device 2 and no longer logs in on device 1, the confidence between the account identification value and the device identification value of device 1 decreases, and the confidence between the account identification value and the device identification value of device 2 increases. In addition, when the number of login days of the account identification value increases on device 2, the confidence between the account identification value and the device identification value of device 2 increases. The more days the account identification value logs in on device 1, the slower the confidence between the account identification value and the device identification value of device 1 decreases. The more consecutive days the account identification value logs in on device 2, the faster the confidence between the account identification value and the device identification value of device 2 increases.

[0072] For example, the current confidence between the account identification value and the device identification value of device 1 is denoted as Y. For Y, if a positive event occurs, it can cause Y to increase, and the increase amount is R*H*L in the following formula (1). If a negative event occurs, it can cause Y to decrease, and the decrease amount is R*H*L in the following formula (2):

[0073] Y+=R*H*L (1)

[0074] In formula (1), R represents the growth space, R = 1 - Y. H represents the historical login basis, H = (X / (M + 1))∧(1 / 2), where X represents the total number of days the account identification value has logged in on device 1, and M represents the maximum value of the number of days the account identification value has logged in on device 1 and device 2 respectively. L represents the performance of the current consecutive login, L = (P / (Q + 1))∧(1 / 3), where P represents the number of consecutive days the account identification value has logged in on device 1, and Q represents the maximum value of the historical consecutive login days of the account identification value on device 1 and device 2 respectively.

[0075] Y-=R*H*L (2)

[0076] In formula (2), R represents the decline space, R = Y. H represents the historical login basis, H = 1 - X / (M + 1). L = N / (X + N), where N represents the total number of days the account identification value has logged in on other devices before the most recent login date on device 1.

[0077] For example, when server 21 calculates the confidence between the account identification value and any device identification value, the output table structure is as shown in Table 1 below:

[0078] Table 1

[0079]

[0080] Optionally, according to the association graph composed of the account identifier, device identifier, global wide area network identifier, mobile phone identifier, and combined identifier, determine the identifier values with an association relationship among the at least one piece of identifier data reported by each of the at least one terminal device, and calculate the confidence level between the identifier values with the association relationship, including the following steps as Figure 4 shown:

[0081] S401. For any two different device identifier values that appear in the at least one piece of identifier data reported by each of the at least one terminal device, calculate the combination of the media access control address and the international mobile equipment identity associated with each device identifier value.

[0082] For example, in actual situations, there may be a case where multiple device identifier values correspond to the same physical device. To avoid multiple device identifier values corresponding to the same physical device being determined as device identifier values of different users, the embodiments of the present disclosure can identify multiple device identifier values corresponding to the same physical device and associate the multiple device identifier values corresponding to the same physical device with a globally unique identifier value.

[0083] Specifically, after the server 21 receives the identifier data reported by different terminal devices, it determines whether each piece of identifier data includes a device identifier value. If it does, it extracts the media access control (MAC) address, international mobile equipment identity (IMEI), operating system identifier value (androidid), advertising identifier (IDFA), and other field information corresponding to the device identifier value. Further, any two pieces of field information among the multiple pieces of field information are combined to obtain multiple combinations (combos) associated with the device identifier value. Among the multiple combinations, different combinations have different priorities. For example, the combination composed of the MAC address and the IMEI has the highest priority.

[0084] S402. If the combinations corresponding to the any two different device identifier values are the same and the operating system identifier values corresponding to the any two different device identifier values do not conflict, then associate the any two different device identifier values with the globally unique identifier value.

[0085] For different identification data received by server 21, when server 21 extracts different device identification values from the different identification data, for any two different device identification values, for example, device identification value 1 and device identification value 2, the combination of the MAC address and IMEI corresponding to device identification value 1 is denoted as combination 1, and the combination of the MAC address and IMEI corresponding to device identification value 2 is denoted as combination 2. If combination 1 and combination 2 are the same, and the androidid corresponding to device identification value 1 and the androidid corresponding to device identification value 2 do not conflict, then device identification value 1 and device identification value 2 are respectively associated with a globally unique identifier (Global Identity, GID) value.

[0086] When server 21 calculates the confidence level between device identification value 1 and this globally unique identifier value, as well as the confidence level between device identification value 2 and this globally unique identifier value, the output table structure is shown in Table 2 below:

[0087] Table 2

[0088]

[0089] Optionally, according to the association graph composed of the account identifier, device identifier, global wide area network identifier, mobile phone identifier, and combined identifier, determine the identifier values with an association relationship in the at least one piece of identification data respectively reported by the at least one terminal device, and calculate the confidence level between the identifier values with the association relationship, including the following steps as shown in Figure 5 shown below:

[0090] S501. For the multiple global wide area network identifier values and device identifier values that appear in the at least one piece of identification data respectively reported by the at least one terminal device, calculate the key value of each global wide area network identifier value among the multiple global wide area network identifier values, and the key value of the device identifier value.

[0091] For example, when a user browses car information through a small program that provides car information, the identification data includes a global wide area network identifier value and a combined identifier value. When a user browses car information through an APP that provides car information, the identification data includes an account identifier value and a device identifier value. When the same user browses car information through the small program and the APP respectively, since the identification data reported by the device is different, therefore, it is necessary to associate the identification data reported by the device in the two cases, so as to identify the identification data reported by the device in the two cases as the identification data of the same user.

[0092] Specifically, the server 21 can extract multiple global wide area network identification values and a device identification value from different identification data on the same day. For example, the server 21 extracts the global wide area network identification value 1 from the identification data 1, extracts the global wide area network identification value 2 from the identification data 2, and extracts the device identification value from the identification data 3. Further, calculate the key value of each global wide area network identification value and the key value of the device identification value. Among them, the key value can be generated according to information such as the system version, mobile phone brand, and mobile phone model.

[0093] S502. If the key value of each global wide area network identification value among the multiple global wide area network identification values is the same as the key value of the device identification value, determine that the multiple global wide area network identification values are respectively associated with the device identification value, and calculate the confidence level between each of the multiple global wide area network identification values and the device identification value, and retain the association relationship between the global wide area network identification value with the highest confidence level and the device identification value.

[0094] For example, the server 21 compares the key value of the global wide area network identification value 1 with the key value of the device identification value, and compares the key value of the global wide area network identification value 2 with the key value of the device identification value. If the key value of the global wide area network identification value 1 is the same as the key value of the device identification value, and the key value of the global wide area network identification value 2 is the same as the key value of the device identification value, it means that one device identification value matches multiple global wide area network identification values. At this time, it is necessary to calculate the confidence level between each global wide area network identification value and the device identification value, and retain the association relationship between the global wide area network identification value with the highest confidence level and the device identification value. Among them, the information for calculating the confidence level is other information except the information required for calculating the key value, such as cookie, Internet Protocol (IP) address, etc.

[0095] Optionally, the method further includes: establishing an association relationship between the combined identification value and the device identification value according to the association relationship between the global wide area network identification value and the device identification value, and the corresponding relationship between the global wide area network identification value and the combined identification value.

[0096] For example, when the user browses car information through a small program that provides car information, the identification data includes a global wide area network identification value and a combined identification value, that is, the global wide area network identification value and the combined identification value are in a binding relationship. Therefore, while retaining the association relationship between the global wide area network identification value with the highest confidence level and the device identification value, according to the association relationship between the global wide area network identification value and the device identification value, and the binding relationship between the global wide area network identification value and the combined identification value, establish the association relationship between the combined identification value and the device identification value.

[0097] such asFigure 6 As shown, the association relationships between different identifiers in box 61 are not limited to the association relationships in the association graph as Figure 3 shown, but also include derived association relationships. For example, the association relationship between GID and DID, and the association relationship between Union-ID and DID. According to the association relationships between different identifiers in box 61, determine the identifier values with association relationships, and then calculate the confidence levels between the identifier values with association relationships. Further, based on the identifier values with association relationships and the confidence levels between the identifier values with association relationships, generate at least one connected graph, such as Figure 6 the 3 connected graphs shown. If each of the 3 connected graphs is not a large connected graph, assign a natural person identifier value to each connected graph, such as P1, P2, P3. In the connected graph, the fields included in the data structure of each edge are as shown in Table 3 below, and the fields included in the data structure of each node are as shown in Table 4 below:

[0098] Table 3

[0099]

[0100] Table 4

[0101]

[0102] Optionally, the method further includes the following steps as Figure 7 shown:

[0103] S701. If the number of nodes in the connected graph is greater than the preset number, split the connected graph to obtain multiple subgraphs.

[0104] For example, when the number of nodes in any connected graph is greater than the preset number, it indicates that the connected graph is a large connected graph. At this time, it is necessary to split the large connected graph to obtain multiple subgraphs.

[0105] S702. For each of the multiple subgraphs, generate a natural person identifier value corresponding to the subgraph.

[0106] For example, assign a natural person identifier value to each subgraph.

[0107] Optionally, splitting the connected graph to obtain multiple subgraphs includes the following steps as Figure 8 shown:

[0108] S801. Using any mobile phone identifier value in the connected graph as the root node, determine multiple target nodes connected to the root node.

[0109] For example, the large connected graph includes 100 nodes, and the 100 nodes include multiple types of nodes, with at least one node of each type. For example, there are multiple nodes corresponding to mobile phone identifiers, globally unique identifiers, account identifiers, and device identifiers respectively, that is, in the large connected graph, there are multiple mobile phone identifier values, globally unique identifier values, account identifier values, and device identifier values respectively. When splitting the large connected graph, any mobile phone identifier value can be used as the root node. Further, determine multiple target nodes connected to the root node. For example, the root node is connected to 4 target nodes, and the numbers of the 4 target nodes are 1, 2, 3, and 4 in sequence.

[0110] S802. Calculate the maximum confidence and average confidence between each target node and its downstream nodes respectively according to the confidence between each target node in the multiple target nodes and its downstream nodes.

[0111] For example, for each of the 4 target nodes, the target node is connected to multiple downstream nodes, and there is a confidence between the target node and each downstream node. For example, the downstream nodes of target node 1 include node 5, node 6, and node 7. Calculate the maximum confidence and average confidence between target node 1 and its downstream nodes according to the confidence between target node 1 and node 5, node 6, and node 7 respectively. Similarly, calculate the maximum confidence and average confidence between other target nodes and their downstream nodes. For example, the number of each target node, the maximum confidence and average confidence between each target node and its downstream nodes form an array, and the 4 target nodes can correspond to 4 arrays, which are (1, 0.9, 0.85), (2, 0.88, 0.80), (3, 0.88, 0.79), and (4, 0.88, 0.79) respectively.

[0112] S803. Sort the multiple target nodes according to the maximum confidence and average confidence between each target node and its downstream nodes respectively.

[0113] For example, sort the 4 target nodes according to the above 4 arrays. The sorting basis is to first sort in descending order according to the maximum confidence, and when the maximum confidence is the same, sort in descending order according to the average confidence. The sorting result is target node 1, target node 2, target node 4, and target node 3.

[0114] S804. Traverse each sorted target node. If the target node meets the preset conditions, retain the target node. When the traversal ends, generate a subgraph according to the root node, the retained target nodes, and the downstream nodes of the retained target nodes.

[0115] For example, traverse each target node after sorting, that is, traverse target node 1, target node 2, target node 4, and target node 3 in sequence. For example, when the currently traversed target node is target node 1, further determine whether the target node 1 meets the following preset conditions:

[0116] 1. Whether the average confidence level meets the threshold requirement (the threshold is 0.5 or 0.8, and the specific value mainly depends on the scale of the large connected graph).

[0117] 2. Whether the number of globally unique identification values among the 5 nodes including the root node, target node 1, target node 2, target node 4, and target node 3 meets the requirement of at most 2, and whether the number of mobile phone identification values meets the requirement of at most 4.

[0118] 3. If target node 1 is an account identification value, whether there is a connection between target node 1 and the mobile phone identification value.

[0119] 4. If target node 1 is a globally unique identification value, whether there is a connection between target node 1 and the device identification value.

[0120] 5. If target node 1 is an account identification value and there are multiple device identification values among the downstream nodes of target node 1, and when the confidence level between target node 1 and any device identification value is lower than the average confidence level and target node 1 needs to be separated from the any device identification value, whether target node 1 remains connected to the device identification value with a higher confidence level.

[0121] If target node 1 meets the above several conditions, retain the connection between target node 1 and the root node; otherwise, delete the connection between target node 1 and the root node, and traverse the next target node.

[0122] After traversing the above 4 target nodes, generate a subgraph based on the root node, the retained target nodes, and the downstream nodes of the retained target nodes. For example, starting from the root node, passing through the retained target nodes and the downstream nodes of the retained target nodes and gradually spreading to the leaf nodes, and forming a subgraph with all the nodes from the root node to the leaf nodes. The subgraph may cover other mobile phone identification values except the root node. Therefore, when selecting the next mobile phone identification value from the large connected graph, select the mobile phone identification value not covered by the subgraph, and cycle the above processing process with this mobile phone identification value as the root node. After all the mobile phone identification values in the large connected graph are included in the subgraph, assign a natural person identification value to each subgraph.

[0123] Figure 9The flowchart of a user identity recognition method in an embodiment of the present disclosure. Specifically, gid represents the globally unique identifier value, mid represents the mobile phone identifier value, uid represents the account identifier value, did represents the device identifier value, web_id represents the global wide area network identifier value, and union_id represents the union identifier value. Further, an association relationship between different identifier values is constructed. For example, gid-did represents the association relationship between the globally unique identifier value and the device identifier value, uid-mid represents the association relationship between the account identifier value and the mobile phone identifier value, uid-did represents the association relationship between the account identifier value and the device identifier value, and the confidence level of uid-did needs to be calculated. did-web_id represents the association relationship between the device identifier value and the global wide area network identifier value. web_id-union_id represents the association relationship between the global wide area network identifier value and the union identifier value. According to did-web_id and web_id-union_id, the association relationship between did-union_id can also be constructed. Further, a connectivity graph is constructed according to each association relationship. If a large connectivity graph appears, it is split. Further, a natural person identifier value is assigned to each connectivity graph or each subgraph obtained after splitting.

[0124] In an embodiment of the present disclosure, each terminal device can use the Hypertext Transfer Protocol Secure (HTTPS) to encrypt and transmit the collected user behavior information to the server 21, thereby ensuring the security of data during transmission. For sensitive data, such as the account identifier value, device identifier value, etc., after being collected at the front end, it can also be encrypted using the Advanced Encryption Standard (AES) encryption algorithm to ensure that the data is not stolen during transmission. User data is stored in an encrypted database, and the AES encryption algorithm is used to encrypt and store the data. At the same time, regular security audits are performed on the database to ensure the security of data storage.

[0125] In addition, strict access permissions are set for the data storage system and the database, and only authorized systems and personnel can access user data. For example, the Role-Based Access Control (RBAC) mechanism is adopted, and different access permissions are assigned according to the user's role (such as administrator, data analyst, etc.) to ensure the legality of data access. The multi-factor identity authentication mechanism (such as username + password + verification code) is adopted to ensure the legality of the user's identity. At the same time, the operation behavior of the user is recorded and audited to ensure the security of data access.

[0126] In addition, during the data collection and storage process, user data can also be anonymized to ensure user privacy. For example, hash the account identification value and replace the user's real identity information with an anonymous identifier. Only collect and store the user data necessary to implement the functions, and avoid over-collecting user information. At the same time, regularly clean up the data and delete the user data that is no longer needed. Before the user uses the service, clearly inform the user of the data collection and usage specifications and obtain the user's explicit consent. At the same time, provide a user data management function that allows users to view, modify, or delete their own data.

[0127] Optionally, the method further includes the following steps as Figure 10 shown:

[0128] S1001. Obtain the user behavior information associated with the natural person identification value according to the same natural person identification value.

[0129] For example Figure 6 as shown, any natural person identification value corresponds to a connected graph, which includes multiple types of identification values, and even there are multiple identification values of each type. Therefore, when using the multiple identification values in the connected graph as the identification values of the same user, since each identification value corresponds to corresponding user behavior information, a large amount of user behavior information can be obtained according to the connected graph.

[0130] S1002. Generate user tags according to the user behavior information.

[0131] Generate user tags according to a large amount of user behavior information. For example, user activity level, user payment ability, user search preference, user coupon preference, user preference for different vehicle models, user preference for different vehicle series, user viewing content preference, car wash preference, maintenance preference, etc. In addition, user tags can also be managed. For example, maintain the entire tag system, define the generation rules of tags through Structured Query Language (SQL), rule engines, etc., and be responsible for the creation, modification, deletion, query, grouping, etc. of tags. In addition, the usage situation of tags can also be monitored and optimized to detect and optimize problem tags in a timely manner.

[0132] S1003. Generate a message push method according to the user tags and the conversion target.

[0133] For example, according to user tags and conversion goals such as maximizing click-through rate (CTR), maximizing user value (Cost Per Mille, CPM or Cost Per Sales, CPS), a message push method is generated. For example, under the constraint of the conversion goal, the optimal reach strategy, operation language, picture materials, copywriting materials, rewards (coupons, cashback, points), etc. are automatically matched. Among them, the reach strategy includes reach time points, reach user groups, reach forms (such as push notifications, text messages, in-site messages, pop-ups, intelligent outbound calls, etc.), reach materials (such as pictures, copywriting, etc.), reach rewards, etc. The operation language includes welcome copywriting, activity guiding language, keyword responses, etc. The message push method includes picture format, picture size, picture link, copywriting word count, main title, subtitle, etc.

[0134] S1004. Analyze user needs based on the user behavior information and the user tags.

[0135] For example, analyze user needs according to user behavior information and user tags. In addition, user needs can also be analyzed by conducting research on users. For example, questionnaire design is carried out based on a research outline, including: setting of questions, question stems, and options. In addition, the question types can also be set, such as including single-choice, multiple-choice, fill-in-the-blank, question-and-answer, etc.

[0136] S1005. Push messages matching the user needs to the user in the message push method.

[0137] For example, after determining the message push method for the user and the user's needs, push messages matching the user's needs to the user in the message push method. For example, feedback coupons to the user in the form of pictures, or feedback coupon prompt information to the user in the form of copywriting, etc.

[0138] In addition, this embodiment can also analyze the conversion effect, such as the user conversion rate, and correct the user tags according to the conversion effect, so that the user tags are more accurate and more in line with user preferences.

[0139] Figure 11 This is the overall architecture diagram provided by the embodiments of the present disclosure. The overall architecture includes user identity recognition, generating user tags, identifying and analyzing needs, intelligent operation, effect analysis and effect feedback. Among them, intelligent operation refers to generating a corresponding message push method according to user tags. Other parts can refer to the content described in the above embodiments.

[0140] Figure 12 This is the structural schematic diagram of a user identity recognition device in the embodiments of the present disclosure. The device provided by the embodiments of the present disclosure can be configured in a server. AsFigure 12 As shown, the user identity recognition device 120 specifically includes:

[0141] An acquisition module 121, configured to acquire at least one piece of identification data reported by at least one terminal device respectively, and each piece of identification data includes at least one identification value;

[0142] A determination module 122, configured to determine, according to an association graph formed by an account identifier, a device identifier, a global wide area network identifier, a mobile phone identifier, and a combined identifier, identification values with an association relationship in at least one piece of identification data reported by at least one terminal device respectively, and calculate a confidence level between the identification values with the association relationship;

[0143] A first generation module 123, configured to generate at least one connected graph according to the identification values with the association relationship and the confidence level between the identification values with the association relationship, where nodes in the connected graph represent the identification values, and edges in the connected graph represent the confidence level;

[0144] A second generation module 124, configured to, for each connected graph in the at least one connected graph, if the number of nodes in the connected graph is less than or equal to a preset number, generate a natural person identification value corresponding to the connected graph.

[0145] Optionally, when the determination module 122 determines, according to an association graph formed by an account identifier, a device identifier, a global wide area network identifier, a mobile phone identifier, and a combined identifier, identification values with an association relationship in at least one piece of identification data reported by at least one terminal device respectively, and calculates a confidence level between the identification values with the association relationship, it is specifically configured to:

[0146] According to the association relationship between the account identifier and the device identifier in the association graph, when it is determined that there are multiple device identifier values corresponding to the same account identifier value in at least one piece of identification data reported by at least one terminal device respectively, calculate the confidence level between the same account identifier value and each device identifier value in the multiple device identifier values respectively.

[0147] Optionally, when the determination module 122 determines, according to an association graph formed by an account identifier, a device identifier, a global wide area network identifier, a mobile phone identifier, and a combined identifier, identification values with an association relationship in at least one piece of identification data reported by at least one terminal device respectively, and calculates a confidence level between the identification values with the association relationship, it is specifically configured to:

[0148] For any two different device identifier values that appear in at least one piece of identification data reported by at least one terminal device respectively, calculate a combination of a media access control address and an international mobile equipment identity associated with each device identifier value;

[0149] If the combinations corresponding to any two different device identification values are the same, and the operating system identification values corresponding to any two different device identification values do not conflict, then associate the any two different device identification values with the globally unique identification value respectively.

[0150] Optionally, when determining module 122 determines the identification values with an association relationship in the at least one piece of identification data respectively reported by the at least one terminal device according to the association graph composed of the account identification, device identification, global wide area network identification, mobile phone identification, and combined identification, and calculates the confidence level between the identification values with the association relationship, it is specifically used for:

[0151] For the multiple global wide area network identification values and device identification values that appear in the at least one piece of identification data respectively reported by the at least one terminal device, calculate the key value of each global wide area network identification value among the multiple global wide area network identification values and the key value of the device identification value;

[0152] If the key value of each global wide area network identification value among the multiple global wide area network identification values is the same as the key value of the device identification value, then determine that the multiple global wide area network identification values are respectively associated with the device identification value, calculate the confidence level between the multiple global wide area network identification values and the device identification value respectively, and retain the association relationship between the global wide area network identification value with the largest confidence level and the device identification value.

[0153] Optionally, the user identity recognition device 120 further includes:

[0154] A splitting module 125, configured to, if the number of nodes in the connected graph is greater than the preset number, split the connected graph to obtain multiple subgraphs;

[0155] The second generation module 124 is further configured to: for each subgraph among the multiple subgraphs, generate a natural person identification value corresponding to the subgraph.

[0156] Optionally, when the splitting module 125 splits the connected graph to obtain multiple subgraphs, it is specifically used for:

[0157] Taking any mobile phone identification value in the connected graph as the root node, determine multiple target nodes connected to the root node;

[0158] According to the confidence level between each target node and its downstream node among the multiple target nodes, calculate the maximum confidence level and average confidence level between each target node and its downstream node;

[0159] Sort the multiple target nodes according to the maximum confidence level and average confidence level between each target node and its downstream node;

[0160] Traverse each target node after sorting. If the target node meets the preset condition, retain the target node. When the traversal ends, generate a subgraph based on the root node, the retained target nodes, and the downstream nodes of the retained target nodes.

[0161] Optionally, the obtaining module 121 is further configured to: obtain user behavior information associated with the natural person identification value according to the same natural person identification value;

[0162] The user identity recognition device 120 further includes:

[0163] A third generation module 126, configured to generate user tags according to the user behavior information; generate a message push method according to the user tags and the conversion target;

[0164] An analysis module 127, configured to analyze user needs according to the user behavior information and the user tags;

[0165] A push module 128, configured to push messages matching the user needs to the user in the message push method.

[0166] The device provided in the embodiments of the present disclosure can execute the method steps provided in the method embodiments of the present disclosure, and the beneficial effects thereof will not be elaborated here.

[0167] Figure 13 This is a schematic structural diagram of an electronic device in the embodiments of the present disclosure. Specifically refer to the following Figure 13 , which shows a schematic structural diagram of the electronic device 1300 suitable for implementing the embodiments of the present disclosure. The electronic device 1300 in the embodiments of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), wearable electronic devices, etc., and fixed terminals such as digital TVs, desktop computers, smart home devices, etc. Figure 13 The electronic device shown is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present disclosure.

[0168] Such as Figure 13As shown, the electronic device 1300 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 1301, which may perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1302 or a program loaded from a storage device 1308 into a random access memory (RAM) 1303 to implement the method of the embodiments as described in this disclosure. In the RAM 1303, various programs and data required for the operation of the electronic device 1300 are also stored. The processing device 1301, the ROM 1302, and the RAM 1303 are connected to each other through a bus 1304. An input / output (I / O) interface 1305 is also connected to the bus 1304.

[0169] Generally, the following devices may be connected to the I / O interface 1305: an input device 1306 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 1307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1309. The communication device 1309 may allow the electronic device 1300 to communicate with other devices wirelessly or wiredly to exchange data. Although Figure 13 the electronic device 1300 with various devices is shown, it should be understood that it is not required to implement or have all the shown devices. Instead, more or fewer devices may be implemented or had.

[0170] Specifically, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, the embodiments of the present disclosure include a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts, thereby implementing the methods as described above. In such an embodiment, the computer program may be downloaded and installed from a network through the communication device 1309, or installed from the storage device 1308, or installed from the ROM 1302. When the computer program is executed by the processing device 1301, the above-mentioned functions defined in the methods of the embodiments of the present disclosure are executed.

[0171] The present disclosure provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the method as provided in any one of the present disclosure is implemented.

[0172] The embodiments of the present disclosure also provide a computer program product, which includes a computer program or instruction, and when the computer program or instruction is executed by a processor, the method as described above is implemented.

[0173] The above description is only a preferred embodiment of the present disclosure and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) disclosed in the present disclosure that have similar functions.

Claims

1. A user identity recognition method, characterized in that, The method includes: Obtaining at least one piece of identification data respectively reported by at least one terminal device, where each piece of identification data includes at least one identification value; Determining, according to an association graph composed of an account identification, a device identification, a global wide area network identification, a mobile phone identification, and a combined identification, the identification values with an association relationship in the at least one piece of identification data respectively reported by the at least one terminal device, and calculating the confidence level between the identification values with the association relationship; Generating at least one connected graph according to the identification values with the association relationship and the confidence level between the identification values with the association relationship, where the nodes in the connected graph represent the identification values, and the edges in the connected graph represent the confidence level; For each of the at least one connected graph, if the number of nodes in the connected graph is less than or equal to a preset number, generating a natural person identification value corresponding to the connected graph.

2. The method according to claim 1, wherein Determining, according to an association graph composed of an account identification, a device identification, a global wide area network identification, a mobile phone identification, and a combined identification, the identification values with an association relationship in the at least one piece of identification data respectively reported by the at least one terminal device, and calculating the confidence level between the identification values with the association relationship, includes: According to the association relationship between the account identification and the device identification in the association graph, when it is determined that there are multiple device identification values corresponding to the same account identification value in the at least one piece of identification data respectively reported by the at least one terminal device, calculating the confidence level between the same account identification value and each of the multiple device identification values respectively.

3. The method according to claim 1, characterized in that, Determining, according to an association graph composed of an account identification, a device identification, a global wide area network identification, a mobile phone identification, and a combined identification, the identification values with an association relationship in the at least one piece of identification data respectively reported by the at least one terminal device, and calculating the confidence level between the identification values with the association relationship, includes: Calculating, for any two different device identification values that appear in the at least one piece of identification data respectively reported by the at least one terminal device, the combination of the media access control address and the international mobile equipment identity code respectively associated with each device identification value; If the combinations respectively corresponding to the any two different device identification values are the same, and the operating system identification values respectively corresponding to the any two different device identification values do not conflict, then associating the any two different device identification values with a globally unique identification value respectively.

4. The method according to claim 1, wherein Determining, according to an association graph composed of an account identification, a device identification, a global wide area network identification, a mobile phone identification, and a combined identification, the identification values with an association relationship in the at least one piece of identification data respectively reported by the at least one terminal device, and calculating the confidence level between the identification values with the association relationship, includes: Calculating, for multiple global wide area network identification values and device identification values that appear in the at least one piece of identification data respectively reported by the at least one terminal device, the key value of each global wide area network identification value among the multiple global wide area network identification values, and the key value of the device identification value. If the key value of each global wide area network identification value among the multiple global wide area network identification values is the same as the key value of the device identification value, determine that the multiple global wide area network identification values are respectively associated with the device identification value, calculate the confidence levels between the multiple global wide area network identification values and the device identification value respectively, and retain the association relationship between the global wide area network identification value with the highest confidence level and the device identification value.

5. The method according to claim 1, wherein The method further includes: If the number of nodes in the connected graph is greater than the preset number, split the connected graph to obtain multiple subgraphs; For each of the multiple subgraphs, generate a natural person identification value corresponding to the subgraph.

6. The method according to claim 5, characterized in that Splitting the connected graph to obtain multiple subgraphs includes: Taking any mobile phone identification value in the connected graph as the root node, determine multiple target nodes connected to the root node; According to the confidence levels between each target node among the multiple target nodes and its downstream nodes respectively, calculate the maximum confidence level and the average confidence level between each target node and its downstream nodes respectively; Sort the multiple target nodes according to the maximum confidence level and the average confidence level between each target node and its downstream nodes respectively; Traverse each sorted target node. If the target node meets the preset condition, retain the target node. When the traversal ends, generate a subgraph according to the root node, the retained target nodes, and the downstream nodes of the retained target nodes.

7. The method according to claim 1 or 5, characterized in that, The method further includes: According to the same natural person identification value, obtain user behavior information associated with the natural person identification value; Generate user tags according to the user behavior information; Generate a message push method according to the user tags and the conversion target; Analyze user needs according to the user behavior information and the user tags; Push a message matching the user needs to the user in the message push method.

8. A user identity recognition device, characterized in that, It includes: An acquisition module, configured to acquire at least one piece of identification data respectively reported by at least one terminal device, and each piece of identification data includes at least one identification value; A determination module, configured to determine, according to an association graph composed of an account identification, a device identification, a global wide area network identification, a mobile phone identification, and a combined identification, the identification values having an association relationship in the at least one piece of identification data respectively reported by the at least one terminal device, and calculate the confidence levels between the identification values having the association relationship; A first generation module, configured to generate at least one connected graph according to the identification values having the association relationship and the confidence levels between the identification values having the association relationship, where the nodes in the connected graph represent the identification values, and the edges in the connected graph represent the confidence levels; A second generation module, configured to, for each of the at least one connected graph, if the number of nodes in the connected graph is less than or equal to the preset number, generate a natural person identification value corresponding to the connected graph.

9. An electronic device, characterized in that, The electronic device includes: One or more processors; A storage device, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, the method according to any one of claims 1-7 is implemented.

Images