Method and device for generating node authority based on authority data to determine specific tags

Abstract

The present application relates to a method, device, electronic device and computer-readable medium for generating node permissions based on permission data to determine specific labels. The method includes: determining keywords based on the category of the specific attribute label to be assigned; determining associated nodes based on the association relationship between the node and other nodes; screening the associated nodes by the keywords to extract multiple initial target nodes; generating multiple sets of model parameters through the permission certification data entered by the node and the multiple initial target nodes; inputting the multiple sets of model parameters into the target attribute label model to generate label probabilities corresponding to each of the initial target nodes; determining the target node from the multiple initial target nodes based on the label probability; and assigning specific permission levels to the target node and the node. The present application can accurately assign corresponding permissions to the current node and the target node, thereby improving system transaction security and data security.

Description

Technical Field

[0001] The present application relates to the field of computer information processing, and more specifically, to a method, device, electronic device, and computer-readable medium for generating node permissions based on permission data to determine specific tags. Background Art

[0002] Relationship networks are graphs that describe individuals and the relationships between them, and they are widely used across various industries. In existing technologies, network nodes and their neighboring nodes are analyzed using relationship graph networks, allowing for in-depth analysis of the nodes. In current relationship network mining, relationships are based on interaction records, but the strength of the relationships between nodes is not distinguished. Instead, they are treated as equal relationships, allowing for the construction of relationship networks, relationship extraction, and data feature mining. Such relationship networks are both redundant and weak. In existing technologies, this type of relationship network is often used for information mining. However, due to the high volume and redundancy of network data and the relatively low amount of information on each node, the conclusions drawn from such network analysis are often biased.

[0003] Therefore, a new method, device, electronic device and computer-readable medium for generating node permissions based on permission data to determine specific tags are needed.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the application and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] In view of this, the present application provides a method, device, electronic device and computer-readable medium for generating node permissions based on permission data to determine specific tags, which can accurately extract the target node corresponding to the current node from multiple nodes, thereby accurately assigning corresponding permissions to the current node and the target node, thereby improving data transmission security, ensuring privacy security, and improving system transaction security.

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

[0007] According to one aspect of the present application, a method for generating node permissions based on specific labels determined based on permission data is proposed, the method comprising: determining keywords according to the category of the specific attribute label to be assigned; determining associated nodes according to the association relationship between the node and other nodes; screening the associated nodes by the keywords to extract multiple initial target nodes; generating multiple groups of model parameters by the permission proof data entered by the node and the multiple initial target nodes; inputting the multiple groups of model parameters into the target attribute label model to generate label probabilities corresponding to each of the initial target nodes; determining the target node from the multiple initial target nodes according to the label probability; and assigning specific permission levels to the target node and the node.

[0008] Optionally, multiple groups of model parameters are input into the target attribute label model to generate multiple label probabilities corresponding to the initial target nodes, including: determining the target attribute label model from multiple attribute label models according to the specific attribute label to be assigned; inputting the multiple groups of model parameters into the target attribute label model in sequence, each group of model parameters corresponds to an initial target node; the target attribute label model generates multiple label probabilities corresponding to multiple initial target nodes based on the multiple groups of model parameters.

[0009] Optionally, multiple groups of model parameters are generated through the authority proof data entered by the node and multiple initial target nodes, including: generating multiple first groups of parameters based on the degree of information consistency between the node and multiple initial target nodes; generating multiple second groups of parameters based on the keyword matching degree of multiple initial target nodes; and generating multiple third groups of parameters based on the interaction frequency between the node and multiple initial target nodes.

[0010] Optionally, multiple first groups of parameters are generated based on the degree of information consistency between the node and multiple initial target nodes, including: performing multi-dimensional matching on the authority proof data of the node and multiple initial target nodes; generating multiple multi-dimensional consistency parameters based on the multi-dimensional matching results; and generating multiple first groups of parameters based on multiple multi-dimensional consistency parameters.

[0011] Optionally, generating multiple second sets of parameters according to the keyword matching degree of the multiple initial target nodes includes: performing similarity matching between the keywords and the remark information of the multiple initial target nodes; and generating multiple second sets of parameters according to the similarity matching results.

[0012] Optionally, it also includes: generating multiple sample sets for multiple specific attribute labels; determining labels for samples in each sample set based on the specific attribute labels corresponding to the sample sets; and training the machine learning model separately through multiple sample sets with labels to generate multiple attribute label models.

[0013] Optionally, the associated nodes are filtered by the keyword to extract multiple initial target nodes, including: extracting the remark information of the associated nodes of the node; filtering the remark information of the associated nodes according to the keyword; and using the associated nodes matched with the keyword as the initial target nodes.

[0014] Optionally, the target node is determined from a plurality of the initial target nodes according to the label probability, including: determining the number of target nodes according to the attributes of the characteristic attribute label to be assigned; arranging the initial nodes from large to small according to the values ​​of their corresponding label probabilities; and extracting the initial nodes in sequence according to the number as the target nodes.

[0015] Optionally, it also includes: providing Internet customized services for the node and the target node according to a specific authority level; generating node portraits of the node and the target node according to the specific authority level.

[0016] According to one aspect of the present application, a device for generating node permissions based on permission data to determine specific labels is proposed, and the device includes: a keyword module for determining keywords according to the category of the specific attribute label to be assigned; an association module for determining associated nodes according to the association relationship between the node and other nodes; a screening module for screening the associated nodes through the keywords and extracting multiple initial target nodes; a parameter module for generating multiple groups of model parameters through the node information of the node and multiple initial target nodes; a probability module for inputting multiple groups of model parameters into a target attribute label model to generate multiple label probabilities corresponding to multiple initial target nodes; an extraction module for determining a target node from multiple initial target nodes according to the label probability; and an allocation module for assigning specific permission levels to the target node and the node.

[0017] According to one aspect of the present application, an electronic device is proposed, which includes: one or more processors; a storage device for storing 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 as described above.

[0018] According to one aspect of the present application, a computer-readable medium is provided, on which a computer program is stored. When the program is executed by a processor, the method described above is implemented.

[0019] According to the node authority generation method, device, electronic device and computer-readable medium of the present application for determining specific labels based on authority data, keywords are determined according to the category of the specific attribute label to be assigned; associated nodes are determined according to the association relationship between the node and other nodes; the associated nodes are screened by the keywords to extract multiple initial target nodes; multiple groups of model parameters are generated by the authority proof data entered by the node and the multiple initial target nodes; the multiple groups of model parameters are input into the target attribute label model to generate label probabilities corresponding to each of the initial target nodes; the target node is determined from the multiple initial target nodes according to the label probability; a method of assigning specific authority levels to the target node and the node can accurately extract the target node corresponding to the current node from multiple nodes, thereby accurately assigning corresponding authorities to the current node and the target node, thereby improving data transmission security, ensuring privacy security, and improving system transaction security.

[0020] It should be understood that the foregoing general description and the following detailed description are merely illustrative and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and other objects, features, and advantages of the present application will become more apparent by describing in detail exemplary embodiments thereof with reference to the accompanying drawings. The drawings described below are merely some embodiments of the present application, and it is apparent to those skilled in the art that other drawings can be derived from these drawings without inventive effort.

[0022] Figure 1 The present invention is a flowchart showing a method for generating node permissions based on permission data to determine specific tags according to an exemplary embodiment.

[0023] Figure 2 The figure is a flowchart showing a method for generating node authority based on authority data to determine a specific tag according to another exemplary embodiment.

[0024] Figure 3 The figure is a flowchart showing a method for generating node authority based on authority data to determine a specific tag according to another exemplary embodiment.

[0025] Figure 4 The present invention is a block diagram showing a device for generating node authority based on authority data to determine a specific tag according to an exemplary embodiment.

[0026] Figure 5 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0027] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be embodied in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art. Like reference numerals in the drawings represent like or similar parts, and thus repetitive description thereof will be omitted.

[0028] The technical abbreviations involved in this application are explained as follows:

[0029] Word frequency is a commonly used weighting technique in information retrieval and text mining, used to assess the repetition rate of a word within a document or a collection of documents within a specific field within a corpus. Word frequency statistics offer new methods and perspectives for academic research.

[0030] TFIDF: The TFIDF method assumes that the lower the frequency of a word in a text, the greater its ability to distinguish different types of text. Therefore, the concept of inverse text frequency (IDF) is introduced. The product of TF and IDF is used as the value measurement of the feature space coordinate system and is used to adjust the weight TF. The purpose of adjusting the weight is to highlight important words and suppress less important words.

[0031] Figure 1 The flowchart of a method for generating node authority based on authority data and specific tags is shown according to an exemplary embodiment. The method 10 for generating node authority based on authority data and specific tags includes at least steps S102 to S114.

[0032] like Figure 1 As shown, in S102, keywords are determined according to the category of the specific attribute tag to be assigned. Different specific attribute tags can be determined in advance according to different business contents or node objects to be analyzed.

[0033] In the embodiments of the present application, nodes may be electronic product terminals. Specific attribute tags between electronic product terminals may include: tags for terminals in the same cluster, tags for terminals belonging to the same user, tags for upper and lower level authority control, and so on. Different electronic product category attribute tags correspond to different keywords, which can be determined in advance through massive data analysis. Keyword tags for terminals in the same cluster may correspond to, for example, "XX cluster XX number," "XX company," and so on.

[0034] In the embodiments of the present application, nodes may also be user objects, and user-specific attribute tags may include: husband-wife tag, father-son tag, mother-son tag, father-daughter tag, mother-daughter tag, brother tag, sister tag, etc. Different user-specific attribute tags correspond to different keywords, and the keywords corresponding to different tags can be obtained in advance through massive data analysis.

[0035] In S104, an associated node is determined according to the association relationship between the node and other nodes.

[0036] In one embodiment, when the nodes are electronic product terminals, the association relationship between the electronic product terminals may include: data transmission relationship, data acquisition relationship, data download relationship, control relationship between nodes, etc.

[0037] In one embodiment, when the node is a user object, the relationship between users may be an interactive relationship between users, which may include: exchanging information, notification information, accessing information, following information, logging in with electronic products together on social networks, etc. More specifically, the user may be an individual user or an enterprise user.

[0038] Without loss of generality, in the following text of this application, the association relationship between actual user objects is taken as an example to provide a detailed description of the technology of this application.

[0039] In S106, the associated nodes are filtered using the keywords to extract multiple initial target nodes. The remark data of the associated nodes of the node can be extracted; the remark data of the associated nodes is filtered based on the keywords; and the associated nodes that match the keywords are used as the initial target nodes. More specifically, the remark data can be information such as attention and notes between different users on a social APP, or data such as group tags of users on an instant messaging platform. For example, remark data between terminal nodes, or identification information of the data volume and data interaction status between terminals in a terminal cluster can also be extracted.

[0040] In S108, multiple sets of model parameters are generated based on the authority certification data entered by the node and the multiple initial target nodes. Multiple first sets of parameters can be generated based on the degree of information consistency between the node and the multiple initial target nodes; multiple second sets of parameters can be generated based on the degree of keyword matching between the multiple initial target nodes; and multiple third sets of parameters can be generated based on the frequency of interaction between the node and the multiple initial target nodes.

[0041] The content of "generating multiple sets of model parameters through the authority proof data entered by the node and multiple initial target nodes" will be Figure 2 The corresponding embodiments are described in detail.

[0042] In S110, multiple sets of model parameters are input into a target attribute label model to generate label probabilities corresponding to each of the initial target nodes. The target attribute label model can be determined from multiple attribute label models based on the specific attribute label to be assigned; the multiple sets of model parameters are sequentially input into the target attribute label model, with each set of model parameters corresponding to an initial target node; and the target attribute label model generates multiple label probabilities corresponding to multiple initial target nodes based on the multiple sets of model parameters.

[0043] Attribute label models for different attribute labels can be trained in advance, for example, a couple attribute label model, which is used to determine whether the input initial target node and the current node to be analyzed are in a husband-and-wife relationship, or a same-cluster terminal recognition model, which is used to determine whether the input initial target node and the current node to be analyzed are terminals in the same cluster.

[0044] According to the calculation above, each initial node corresponds to a set of model parameters. The model parameters corresponding to multiple initial nodes are input into the target attribute model in turn to generate the corresponding label probability of each initial node.

[0045] In one embodiment, it also includes: generating multiple sample sets for multiple specific attribute labels; determining labels for samples in each sample set based on the specific attribute labels corresponding to the sample sets; and training the machine learning model separately through multiple sample sets with labels to generate multiple attribute label models.

[0046] The details of "Generating Multiple Attribute Label Models" will be Figure 3 The corresponding embodiments are described in detail.

[0047] In S112, a target node is determined from the plurality of initial target nodes based on the label probability. The number of target nodes may be determined based on the attributes of the characteristic attribute labels to be assigned; the initial nodes are arranged from largest to smallest according to their corresponding label probability values; and the initial nodes are sequentially extracted as the target nodes based on the number.

[0048] It can be understood that different target attribute label models correspond to different numbers of target nodes. The number of target nodes corresponding to the husband-wife target attribute label model is 1, that is, the husband-wife relationship is a one-to-one relationship. The number of target nodes corresponding to the cluster terminal recognition model can be multiple, such as 3 or 5, which can be set according to the analysis of big data.

[0049] In a practical application, keywords filter out three initial target nodes, and the model parameters corresponding to the three initial target nodes are input into the couple attribute label model to generate three label probabilities, which can be, for example, 0.6; 0.3; 0.8, representing the probabilities that the initial target node and the current node are a couple, respectively.

[0050] The number of target nodes corresponding to the couple attribute label model is 1, so the initial target node corresponding to the maximum probability is extracted as the target node.

[0051] Similarly, in the same cluster terminal identification model, the number of identification target nodes can be set to 1 or multiple to filter whether they belong to the same cluster.

[0052] In S114 , a specific authority level is assigned to the target node and the node.

[0053] In one embodiment, after the same cluster terminal identification model is calculated, specific attribute labels can be assigned to the current node and the target node according to the calculation results. For example, the terminal identifier in the auxiliary identification information can be used to set the same cluster corresponding identifier for the current node and the target node.

[0054] In one embodiment, after the parent-child attribute label model is calculated, the age in the user information can also be used to assign specific attribute labels to the current node and the target node, such as assigning a "parent" label to the elderly and a "child" label to other nodes.

[0055] In one embodiment, Internet customized services can be provided to the node and the target node according to specific authority levels. Data resource sharing services or network resource sharing services can be allocated to users with the couple label.

[0056] In one embodiment, node portraits of the node and the target node may be generated based on a specific permission level. User portraits of users may be assisted in generating based on a specific permission level and specific attribute tags.

[0057] According to the node authority generation method for determining specific labels based on authority data of the present application, keywords are determined according to the category of the specific attribute label to be assigned; associated nodes are determined according to the association relationship between the node and other nodes; the associated nodes are screened by the keywords to extract multiple initial target nodes; multiple groups of model parameters are generated by the authority proof data entered by the node and the multiple initial target nodes; the multiple groups of model parameters are input into the target attribute label model to generate label probabilities corresponding to each of the initial target nodes; the target node is determined from the multiple initial target nodes according to the label probability; a method of assigning specific authority levels to the target node and the node can accurately extract the target node corresponding to the current node from multiple nodes, thereby accurately assigning corresponding authorities to the current node and the target node, thereby improving data transmission security, ensuring privacy security, and improving system transaction security.

[0058] In a practical application, in the current relationship network mined, the basis of the relationship is the interaction record between users, but it does not distinguish the strength of the relationship between each person. Instead, it regards the relationship as equal, and constructs the relationship network and extracts the relationship, and mines data features. Such relationships are redundant and weak. Simply using the existing interactive relationships to build a network for information mining, and including relatively rough character relationships, it is easy to build a huge relationship network with little information. The node authority generation method based on authority data to determine specific tags in this application can simplify the existing relationship network, such as only retaining nodes with the same relationship, or only retaining nodes with close relationships. The simplified relationship network can more quickly analyze the user portrait of the current node.

[0059] In another practical application, some exclusive services include clauses, such as requiring that private information be shared only between immediate family members. This policy relies heavily on the special relationships between users, and it's easy to make mistakes if judgments are made based solely on the family information provided by the customer during the application process. This method, which uses permission data to determine specific tags for node permissions, can cover 90% of specific attribute relationship tags, providing user service providers with additional basis for judgment and making it more conducive to providing exclusive services to users.

[0060] In another practical application, in some data services, the crawled data accessed by multiple terminals of the black market needs to be strictly screened to prevent the data from being illegally obtained. In this solution, when it is determined that the terminals associated with the server providing the service belong to the same cluster, they are marked to quickly screen out terminals that may be controlled by the black market, thereby improving data security.

[0061] It should be clearly understood that this application describes how to form and use specific examples, but the principles of this application are not limited to any details of these examples. On the contrary, based on the teaching of the content disclosed in this application, these principles can be applied to many other embodiments.

[0062] Figure 2 The figure is a flowchart showing a method for generating node authority based on authority data to determine a specific tag according to another exemplary embodiment. Figure 2 The process 20 shown is Figure 1 A detailed description of S108 "generating multiple sets of model parameters through the authority certification data entered by the node and multiple initial target nodes" in the process shown.

[0063] like Figure 2 As shown, in S202, multiple first groups of parameters are generated according to the degree of consistency of information between the node and multiple initial target nodes.

[0064] In one embodiment, the permission certification data of the node and the multiple initial target nodes are multi-dimensionally matched; multiple multi-dimensional consistency parameters are generated based on the multi-dimensional matching results; and multiple first group parameters are generated based on the multiple multi-dimensional consistency parameters.

[0065] Among them, the authority proof data can be the node attributes in the node information, node description data, operation behavior data, the password information of the node user in the social APP, login information, IP address information, operation information information in the APP, open social information, registration identity information, etc., or, in the terminal cluster, it can be the account information logged in to the terminal, the terminal's operation behavior data, the terminal's illegal operation records, etc.

[0066] Nodes corresponding to different node attribute labels are likely to have overlapping content or very similar content in different permission proof data.

[0067] In one embodiment, the authority certification data of the initial target node and the authority certification data of the current node are matched in multiple dimensions. For user nodes, age, address, and place of origin can be matched. For terminal nodes, IP address and node identifier names can be matched.

[0068] In S204, multiple second sets of parameters are generated based on the keyword matching degrees of the multiple initial target nodes. The number and frequency of words matching the keywords of the initial target nodes and the target attributes are different, and the second sets of parameters are generated based on different matching degrees.

[0069] In one embodiment, similarity matching is performed between the keyword and the remark information of the plurality of initial target nodes, and the plurality of second sets of parameters are generated based on the similarity matching results. The similarity matching can be performed using TFIDF technology.

[0070] In S206, multiple third sets of parameters are generated based on the interaction frequencies between the node and the multiple initial target nodes. The number of data transmissions and information exchanges between the node and the initial node can be used as the interaction frequencies to generate the third set of parameters.

[0071] In S208, model parameters are generated using the first, second, and third sets of parameters. More specifically, the first set of parameters can be used to describe the consistency between the node's own information and the initial node, such as age, province, age difference between two people, whether they are in the same province or city, whether they are in the same county, and whether their mobile phone numbers are in the same province. The second set of parameters can be used to describe the similarity of additional information, which can be the number of matched keywords, the length of the note minus the keyword length, the TF-IDF of the matched keywords, etc. The third set of parameters can be used to describe the frequency of interaction between two people, such as the frequency of data sharing between the two people, the number of online chats, and the number of online information exchanges.

[0072] Figure 3 The figure is a flowchart showing a method for generating node authority based on authority data to determine a specific tag according to another exemplary embodiment. Figure 3 The process 30 shown is Figure 1 In the process shown, S110 “inputting multiple groups of the model parameters into the target attribute label model to generate label probabilities corresponding to the initial target nodes” is described in detail.

[0073] like Figure 3 As shown, in S302, multiple sample sets are generated for multiple specific attribute tags. For example, different electronic product category attribute tags correspond to different keywords. The keywords corresponding to different tags can be obtained in advance through massive data analysis. The keywords corresponding to the terminal tags of the same cluster can be "XX cluster XX number", "XX company", etc.

[0074] In S304, labels are determined for samples in each sample set based on the specific attribute labels corresponding to the sample set. The label assignment strategy for positive samples can be: keyword match + company / cluster information in the authorization data + the company / cluster is the same as the target node; negative samples: no keyword match, no company / cluster information in the authorization data, or the company / cluster is different from the target node.

[0075] In S306, the machine learning model is trained using multiple sample sets with labels to generate multiple attribute label models. The machine learning model is trained using different samples to generate attribute label models for different attribute labels.

[0076] In this application's node authority generation method for determining specific tags based on authority data, an attribute labeling model is proposed that can analyze the associated nodes of the current node to obtain the specific relationship nodes of the node. In a specific user relationship network, the attribute labeling model in this application can be used to obtain relationships such as the cluster relationship to which the terminal belongs, whether the terminal operation status is the same, and so on. This not only provides a weighted assessment of intimacy in the existing relationship network, but also provides more strongly associated interdependence information for portraits, etc.

[0077] Similarly, for individual users or corporate users, the above method can also be used to determine whether there is a relationship between user nodes, such as family relationships, friendships, colleagues, etc. among individual users, or whether there is a relationship between corporate controllers, corporate superior-subordinate relationships, corporate transaction relationships, etc. among corporate users.

[0078] Furthermore, it should be noted that the aforementioned figures are merely illustrative of the processes included in the methods according to exemplary embodiments of the present application and are not intended to be limiting. It is readily understood that the processes illustrated in the aforementioned figures do not indicate or limit the temporal order of these processes. Furthermore, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0079] The following are device embodiments of the present application, which can be used to implement the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0080] Figure 4 FIG. 1 is a block diagram showing a device for generating node authority based on authority data to determine a specific tag according to an exemplary embodiment. Figure 4 As shown, the node authority generation device 40 for determining a specific tag based on authority data includes: a keyword module 402 , an association module 404 , a screening module 406 , a parameter module 408 , a probability module 410 , an extraction module 412 , and an allocation module 414 .

[0081] The keyword module 402 is used to determine keywords according to the category of the specific attribute tag to be assigned;

[0082] The association module 404 is used to determine an associated node based on the association relationship between the node and other nodes;

[0083] The screening module 406 is used to screen the associated nodes using the keywords and extract multiple initial target nodes; the screening module 406 is also used to extract the remark information of the associated nodes of the node; screen the remark information of the associated nodes according to the keywords; and use the associated nodes that match the keywords as the initial target nodes.

[0084] The parameter module 408 is used to generate multiple groups of model parameters based on the node information of the node and multiple initial target nodes; the parameter module 408 is also used to generate multiple first groups of parameters based on the degree of information consistency between the node and multiple initial target nodes; generate multiple second groups of parameters based on the keyword matching degree of multiple initial target nodes; and generate multiple third groups of parameters based on the interaction frequency between the node and multiple initial target nodes.

[0085] The probability module 410 is used to input multiple groups of model parameters into the target attribute label model to generate multiple label probabilities corresponding to the multiple initial target nodes; the probability module 410 is also used to determine the target attribute label model from multiple attribute label models based on the specific attribute label to be assigned; input the multiple groups of model parameters into the target attribute label model in sequence, each group of model parameters corresponds to an initial target node; the target attribute label model generates multiple label probabilities corresponding to the multiple initial target nodes based on the multiple groups of model parameters.

[0086] The extraction module 412 is used to determine the target node from the multiple initial target nodes according to the label probability; the extraction module 412 is also used to determine the number of target nodes according to the attributes of the characteristic attribute label to be assigned; arrange the initial nodes from large to small according to the values ​​of their corresponding label probabilities; and extract the initial nodes in order according to the number as the target nodes.

[0087] The allocating module 414 is configured to allocate specific permission levels to the target node and the node.

[0088] According to the node authority generation device for determining specific labels based on authority data of the present application, keywords are determined according to the category of the specific attribute label to be assigned; associated nodes are determined according to the association relationship between the node and other nodes; the associated nodes are screened by the keywords to extract multiple initial target nodes; multiple groups of model parameters are generated by the authority proof data entered by the node and the multiple initial target nodes; the multiple groups of model parameters are input into the target attribute label model to generate label probabilities corresponding to each of the initial target nodes; the target node is determined from the multiple initial target nodes according to the label probability; a method of assigning specific authority levels to the target node and the node can accurately extract the target node corresponding to the current node from multiple nodes, thereby accurately assigning corresponding authorities to the current node and the target node, thereby improving data transmission security, ensuring privacy security, and improving system transaction security.

[0089] like Figure 5 As shown, an embodiment of the present invention provides an electronic device, including a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540;

[0090] Memory 530, for storing computer programs;

[0091] The processor 510 is configured to implement the data allocation authority adjustment method based on video facial expressions of any of the above embodiments when executing the program stored in the memory 530 .

[0092] In an electronic device provided by an embodiment of the present invention, the processor 510 obtains the initial data allocation permissions and access information of a target by executing a program stored in the memory 530; determines the video text content through the access information; establishes a real-time video link with the target, and displays the video text content based on the video connection to generate video data; recognizes the user's facial expressions and actions in the video data to determine the corresponding permission adjustment coefficient; and adjusts the user's data allocation permissions based on the initial permissions and the permission adjustment coefficient.

[0093] The communication interface 520 is used for communication between the electronic device and other devices.

[0094] The memory 530 may include a random access memory 530 (Random Access Memory, RAM for short), and may also include a non-volatile memory 530 (non-volatile memory), such as at least one disk memory 530 .

[0095] The embodiment of the present application provides a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the node authority generation method based on authority data to determine a specific label in any of the above embodiments. For example, the method can determine keywords according to the category of the specific attribute label to be assigned; determine associated nodes according to the association relationship between the node and other nodes; filter the associated nodes by the keywords to extract multiple initial target nodes; generate multiple groups of model parameters by the authority certification data entered by the node and the multiple initial target nodes; input the multiple groups of model parameters into the target attribute label model to generate label probabilities corresponding to each of the initial target nodes; determine the target node from the multiple initial target nodes according to the label probability; and assign specific authority levels to the target node and the node.

[0096] While the exemplary embodiments of the present application are specifically illustrated and described above, it should be understood that the present application is not limited to the detailed structures, configurations, or implementations described herein; rather, the present application is intended to cover various modifications and equivalent configurations within the spirit and scope of the appended claims.

Claims

1. A method for generating node permissions based on permission data to determine specific tags, characterized in that: include: Determining keywords based on the category of the specific attribute label to be assigned; Determine an associated node based on an association relationship between the node and other nodes; Extracting the remark information of the node associated with the node; Filter the remark information of the associated nodes according to the keywords; The associated node that matches the keyword is used as the initial target node; Generate multiple sets of model parameters using the permission proof data entered by the node and the multiple initial target nodes; Determine a target attribute label model from a plurality of attribute label models according to a specific attribute label to be assigned; Inputting the multiple sets of model parameters into the target attribute label model in sequence, each set of model parameters corresponding to an initial target node; The target attribute label model generates multiple label probabilities corresponding to multiple initial target nodes based on multiple groups of model parameters; Determine a target node from the multiple initial target nodes according to label probability; The target node and the node are assigned specific authority levels.

2. The method according to claim 1, wherein Generate multiple sets of model parameters using the authority proof data entered by the node and the multiple initial target nodes, including: Generate multiple first groups of parameters based on the degree of consistency of information between the node and the multiple initial target nodes; generating a plurality of second sets of parameters according to the keyword matching degrees of the plurality of initial target nodes; A plurality of third groups of parameters are generated according to the interaction frequencies between the node and the plurality of initial target nodes.

3. The method according to claim 2, wherein A plurality of first groups of parameters are generated according to the degree of consistency of information between the node and the plurality of initial target nodes, including: Performing multi-dimensional matching on the authority proof data of the node and the plurality of initial target nodes; Generate multiple multi-dimensional consistency parameters according to the multi-dimensional matching results; A plurality of the first groups of parameters are generated according to a plurality of the multi-dimensional consistency parameters.

4. The method according to claim 2, wherein A plurality of second sets of parameters are generated according to the keyword matching degrees of the plurality of initial target nodes, including: Performing similarity matching between the keyword and the remark information of the plurality of initial target nodes; A plurality of the second groups of parameters are generated according to the similarity matching results.

5. The method according to claim 1, wherein Also includes: Generate multiple sample sets for multiple specific attribute labels; Determine labels for samples in each sample set based on specific attribute labels corresponding to the sample set; The machine learning model is trained separately using multiple sample sets with labels to generate multiple attribute label models.

6. The method according to claim 1, wherein Determining a target node from the plurality of initial target nodes according to label probability includes: Determine the number of target nodes according to the attributes of the characteristic attribute labels to be assigned; Arrange the initial nodes from large to small according to the values ​​of their corresponding label probabilities; Initial nodes are extracted in sequence according to the number as the target nodes.

7. The method according to claim 1, wherein Also includes: providing Internet customized services for the node and the target node according to a specific authority level; Generate node profiles of the node and the target node according to a specific authority level.

8. A node authority generation device for determining a specific tag based on authority data, characterized in that: include: A keyword module, used to determine keywords based on the category of the specific attribute label to be assigned; An association module, configured to determine an associated node based on an association relationship between the node and other nodes; A screening module, used to extract the remark information of the associated nodes of the node; Filter the remark information of the associated nodes according to the keyword; the associated nodes matching the keyword are used as the initial target nodes; A parameter module, configured to generate multiple sets of model parameters using node information of the node and multiple initial target nodes; a probability module, configured to determine a target attribute label model from a plurality of attribute label models according to a specific attribute label to be assigned; and sequentially input the plurality of sets of model parameters into the target attribute label model, wherein each set of model parameters corresponds to an initial target node; The target attribute label model generates multiple label probabilities corresponding to multiple initial target nodes based on multiple groups of model parameters; An extraction module, configured to determine a target node from the plurality of initial target nodes according to label probability; An allocation module is used to allocate specific permission levels to the target node and the node.

9. An electronic device, characterized in that: include: one or more processors; a storage device for storing 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 to 7.

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

Images