Method and device for determining node authority of a relationship network optimized by matching specific tags

Abstract

The present application relates to a method, device, electronic device and computer-readable medium for determining node permissions of a relationship network optimized by matching specific labels. The method includes: extracting node data of multiple nodes in a node relationship network; obtaining category keywords and category labels corresponding to multiple categories; obtaining level keywords and level labels corresponding to category levels of each category in multiple categories; matching the multiple nodes through the node data according to the category keywords and level keywords, and assigning category labels and / or level labels to the successfully matched nodes; merging the nodes in the node relationship network and / or deleting the relationship edges between nodes according to the category labels and / or level labels to optimize the relationship network; determining the data access permission level of each node based on the optimized relationship network. The present application can quickly, accurately and reasonably optimize and simplify the relationship network, and then quickly determine the node permissions through the optimized relationship network, thereby increasing system security and transaction 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 determining node permissions in a relationship network optimized by matching specific tags. Background Art

[0002] A relational network is a graph that describes the relationships between individuals and has been widely used in various industries. In the existing technology, a network node and its neighboring nodes are analyzed with the help of a relational graph network, and then the node is deeply analyzed. Existing relational networks mainly rely on graph databases for data processing, and use the algorithms provided by the graph database for graph calculations. A graph is mainly composed of nodes (Node) and the relationships (relationships) between nodes, that is, edges. However, due to the inconvenience of graph data processing, when the existing technology uses a relational graph network to perform node analysis, for a relational network with billions of nodes, limited resources make it difficult to support multiple such large graphs. Different business scenarios use different point attributes and edge attributes, and the expansion of edge attributes is very inconvenient.

[0003] Therefore, a new method, device, electronic device and computer-readable medium for determining node authority of a network optimized for matching specific tags is 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 determining node permissions of a relationship network optimized by matching specific tags, which can quickly, accurately and reasonably optimize and simplify the relationship network, and then quickly determine node permissions through the optimized relationship network, thereby increasing system security and 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 determining node permissions of a relationship network optimized by matching specific labels is proposed, the method comprising: extracting node data of multiple nodes in the node relationship network; obtaining category keywords and category labels corresponding to multiple categories; obtaining level keywords and level labels corresponding to category levels of each category in the multiple categories; matching the multiple nodes through the node data according to the category keywords and level keywords, and assigning category labels and / or level labels to the successfully matched nodes; merging the nodes in the node relationship network and / or deleting the relationship edges between nodes according to the category labels and / or level labels to optimize the relationship network; and determining the data access permission level of each node based on the optimized relationship network.

[0008] Optionally, the method further includes: acquiring multiple sample nodes corresponding to multiple categories; analyzing the node data of the sample nodes of each category to extract category keywords corresponding to the category; and determining category labels of corresponding categories according to the category keywords.

[0009] Optionally, it also includes: obtaining sample nodes corresponding to each category; analyzing the node data of the sample nodes of each level in each category, extracting the level keywords corresponding to each category level in the category; and determining the level labels corresponding to each level keyword.

[0010] Optionally, the multiple nodes are matched through the node data according to the category keywords and the level keywords, and category labels and / or level labels are assigned to the successfully matched nodes, including: category matching the category keywords and the node data of the multiple nodes; after the category matching is successful, assigning category labels to the nodes with successful category matching; level matching the level keywords corresponding to the category and the nodes with successful category matching; after the level matching is successful, assigning level labels to the nodes with successful level matching.

[0011] Optionally, the nodes in the node relationship network are merged and / or the relationship edges between nodes are deleted according to the category labels and / or level labels to optimize the relationship network, including: determining the target category and the target category level; retaining the nodes in the node relationship network that contain the target category label and the target category level label to optimize the node relationship network.

[0012] Optionally, the nodes in the node relationship network are merged and / or the relationship edges between nodes are deleted according to the category labels and / or level labels to optimize the relationship network, including: generating relationship importance according to the category labels and / or level labels corresponding to the nodes; merging the nodes in the node relationship network and / or deleting the relationship edges between nodes according to the relationship importance.

[0013] Optionally, the relationship importance is generated according to the category label and / or level label corresponding to the node, including: assigning importance coefficients to multiple category labels and level labels respectively; and generating the relationship importance of the node according to the sum of the importance coefficients corresponding to the category label and / or level label of the node.

[0014] Optionally, the nodes in the node relationship network are merged and / or the relationship edges between nodes are deleted according to the relationship importance, including: determining the relationship importance interval, and taking the nodes whose relationship importance is not in the relationship importance interval as the nodes to be optimized; eliminating the nodes to be optimized; and / or merging the nodes to be optimized into other nodes with the same association relationship and category label according to the category label of the nodes to be optimized; and / or if the node connected to the node to be optimized is connected to other nodes with the same category label as the node to be optimized, then deleting the relationship edge between the node to be optimized and the node to be optimized, so as to optimize the node relationship network.

[0015] Optionally, it also includes: generating a node portrait of the target node through the optimized node relationship network; and / or generating a service strategy for the target node through the optimized node relationship network.

[0016] According to one aspect of the present application, a device for determining node permissions of a relationship network optimized by matching specific labels is proposed, and the device includes: a data module for extracting node data of multiple nodes in the node relationship network; a category module for obtaining category keywords and category labels corresponding to multiple categories; a level module for obtaining level keywords and level labels corresponding to category levels of each category in the multiple categories; a matching module for matching the multiple nodes through the node data according to the category keywords and level keywords, and assigning category labels and / or level labels to the successfully matched nodes; an optimization module for merging nodes in the node relationship network and / or deleting relationship edges between nodes according to the category labels and / or level labels to optimize the relationship network; and a permission module for determining the data access permission level of each node based on the optimized relationship network.

[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 determination method, device, electronic device and computer-readable medium of the present application for matching specific tags to optimize the relationship network, the node data of multiple nodes in the node relationship network are extracted; category keywords and category labels corresponding to multiple categories are obtained; level keywords and level labels corresponding to category levels of each category in the multiple categories are obtained; according to the category keywords and level keywords, the multiple nodes are matched through the node data, and category labels and / or level labels are assigned to the successfully matched nodes; according to the category labels and / or level labels, the nodes in the node relationship network are merged and / or the relationship edges between nodes are deleted to optimize the relationship network; the method of determining the data access permission level of each node based on the optimized relationship network can quickly, accurately and reasonably optimize and simplify the relationship network, and then quickly determine the node authority through the optimized relationship network, thereby increasing system security and 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 determining node authority of a relationship network optimized by matching specific tags according to an exemplary embodiment.

[0023] Figure 2 The present invention is a flowchart showing a method for determining node authority of a relationship network optimized by matching specific tags according to another exemplary embodiment.

[0024] Figure 3 The present invention is a flowchart showing a method for determining node authority of a relationship network optimized by matching specific tags according to another exemplary embodiment.

[0025] Figure 4 The present invention is a block diagram showing a device for determining node authority of a relationship network optimized for matching specific tags 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 determining node authority of a relationship network optimized by matching specific tags according to an exemplary embodiment is shown. The method 10 for determining node authority of a relationship network optimized by matching specific tags includes at least steps S102 to S112.

[0032] like Figure 1 As shown, in S102, node data of multiple nodes in the node relationship network are extracted.

[0033] In one embodiment of the present application, the node relationship network may be an electronic product terminal relationship network, and the nodes may be individual electronic product terminals in the electronic product terminal relationship network. The node data of the electronic product terminal may include: data transmission information, data acquisition information, data download information, information on whether the data transmission is in violation of regulations, information on illegal acquisition of data in the terminal, data processing volume information, etc.

[0034] In one embodiment of the present application, the node relationship network may also be a user relationship network, the nodes may be user objects, and the node relationship network may be a relationship network generated by the relationships between users. More specifically, the users may be individual users or corporate users.

[0035] Take, for example, an actual user relationship network. Node data can be social network data, and a user relationship network can be established through interaction data between users on the social network. The user relationship network is built based on interaction data between users, more specifically, based on interaction data such as likes, follows, and comments between users on the social network. Node information can also include note data, i.e., notes users have made for other users on the social network, or category names or annotations users have made for other users on third-party instant messaging platforms.

[0036] In S104, category keywords and category labels corresponding to multiple categories are obtained. Category labels and category keywords corresponding to different categories can be pre-generated, and category matching is performed between node data and category keywords. More specifically, categories may include friend categories, online friends categories, colleague categories, etc. After matching a category keyword, the user is assigned a category label corresponding to the category.

[0037] In S106, the level keywords and level labels corresponding to the category levels of each category in the multiple categories are obtained. Category level labels and level keywords corresponding to each category can be pre-generated, and the node data and level keywords are matched to the categories. More specifically, the friend category can include three levels: first-level friends, second-level friends, and third-level friends, each corresponding to a different level keyword. After matching a keyword for a certain level, the user is assigned a category label corresponding to that level.

[0038] In S108, the multiple nodes are matched through the node data according to the category keywords and the level keywords, and category labels and / or level labels are assigned to the nodes that are successfully matched. The category keywords are matched with the node data of the multiple nodes; after the category matching is successful, the category labels are assigned to the nodes that are successfully matched; the level keywords corresponding to the category are matched with the nodes that are successfully matched; after the level matching is successful, the level labels are assigned to the nodes that are successfully matched. First, different category labels are determined for different nodes. In the category level matching, only the keywords of the multiple level labels corresponding to the category labels of the nodes are matched.

[0039] In S110 , nodes in the node relationship network are merged and / or relationship edges between nodes are deleted according to category labels and / or level labels to optimize the relationship network.

[0040] In one embodiment, a target category and a target category level may be determined; and nodes in the node relationship network containing target category labels and target category level labels are retained to optimize the node relationship network.

[0041] For example, in a specific application, when analyzing a user, the focus is on the user's colleague relationship network. All neighboring nodes with the colleague label of the user are retained, and other nodes with the colleague label of all neighboring nodes are retained to simplify the user relationship network. During the simplification process, edges between nodes are deleted or merged according to the target category and target category level.

[0042] In one embodiment, the importance of a node relationship can be generated based on the category label and / or level label corresponding to the node; the nodes in the node relationship network can be merged and / or the relationship edges between nodes can be deleted based on the importance of the relationship. For example, different weights can be assigned to different labels to calculate the importance of a node, thereby simplifying the node relationship network based on the importance. The details will be discussed in detail. Figure 3 The corresponding embodiments are described in detail.

[0043] In S112, the data access permission level of each node is determined based on the optimized relationship network. The security level of the user to be analyzed can be analyzed based on the optimized relationship network. More specifically, the security level of the user can be determined based on the attributes of the user's neighboring users. The user and their neighboring users can also be input into a trained security analysis model to calculate the user's risk level. This application is not limited to this.

[0044] In one embodiment, the method further includes: generating a node profile of the target node through the optimized node relationship network; and / or generating a service strategy for the target node through the optimized node relationship network. A user profile or service strategy may also be generated through the optimized node relationship network.

[0045] According to the node authority determination method of the matching specific label optimization relationship network of the present application, the node data of multiple nodes of the node relationship network is extracted; category keywords and category labels corresponding to multiple categories are obtained; level keywords and level labels corresponding to category levels of each category in the multiple categories are obtained; according to the category keywords and level keywords, the multiple nodes are matched through the node data, and category labels and / or level labels are assigned to the successfully matched nodes; according to the category labels and / or level labels, the nodes in the node relationship network are merged and / or the relationship edges between nodes are deleted to optimize the relationship network; based on the method of determining the data access permission level of each node in the optimized relationship network, the relationship network can be optimized and simplified quickly, accurately and reasonably, and then the node authority can be quickly determined through the optimized relationship network, thereby increasing system security and transaction security.

[0046] According to the node authority determination method of the present application for matching specific tags to optimize the relationship network, the intimacy relationship data of the node to be identified and its related neighbors in the social network can be output only based on the interaction data on the social network. For example, the intimacy between users who interact frequently will definitely be higher, and users with different relationships will also have different language patterns during the interaction. If other users are notified when commenting, there must be a certain relationship between the notified user and the notified user. It mainly aims to evaluate the intimacy of the social network relationship after obtaining the interaction data of the user on the social network, and give the intimacy relationship from strong to weak; this type of social relationship with strong intimacy not only provides the weight evaluation of intimacy in the existing relationship network, but also provides the dependency relationship of mutual user portraits for the Internet service business, so as to better provide personalized services for users.

[0047] In a specific embodiment, the node relationship network can be an electronic product terminal relationship network. In the relationship network, for data terminals or data servers that provide data services, it is necessary to determine the security of the terminal that obtains the service, or in other words, it is necessary to determine whether the terminal that obtains the data is a black market terminal. At this time, since the number of terminals (user terminals) that have an associated relationship with the data terminal is very large, the constructed electronic product terminal relationship network is also very complicated, and it is necessary to improve the recognition efficiency of each terminal. Through the scheme disclosed in the present application, different nodes are matched according to category keywords and level keywords, and category labels and / or level labels are assigned to the successfully matched nodes. The relationship network is optimized through category labels or level labels. On the one hand, the terminal relationship network is simplified. On the other hand, merging the terminal nodes means that the merged terminal nodes have the same characteristics. For example, the registration information of the accounts in the merged terminal nodes overlaps. Specifically, the identity information used for registration overlaps, the registration location overlaps, the data access habits are consistent, etc. These data are used to determine whether there is a relationship between the terminals. Since the black industry generally obtains data through massive data access, the terminals controlled by the black industry are generally automatically accessed by computers. Therefore, at this time, merging the terminal nodes in the relationship network can be used to indicate whether the black industry has been identified, and ultimately determine the data access rights of each node based on the optimized relationship network.

[0048] 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.

[0049] Figure 2 The present invention is a flowchart showing a method for determining node authority of a relationship network optimized by matching specific tags according to another exemplary embodiment. Figure 2 The process 20 shown is Figure 1 Supplementary description of the process shown.

[0050] like Figure 2 As shown, in S202, category keywords corresponding to multiple categories are generated. Multiple sample nodes corresponding to multiple categories can be obtained, the node data of the sample nodes of each category are analyzed, the category keywords corresponding to the category are extracted, and the category labels of the corresponding categories are determined according to the category keywords.

[0051] In S204, the level keywords corresponding to each category are generated. Sample nodes corresponding to each category are obtained; the node data of the sample nodes of each level in each category are analyzed to extract the level keywords corresponding to each category level in the category; and the level labels corresponding to each level keyword are determined.

[0052] In S206 , the category keywords and the level keywords corresponding to each category are stored.

[0053] In a practical scenario, taking the relationship network of electronic product terminals as an example, for example, when an official website provides data services to users, it needs to be wary of black industries crawling the data stored in the website. At the same time, when black industries crawl the stored data, they will obtain data at a high frequency and a high number of downloads, causing congestion on the website. In addition, in order to circumvent the system's restrictions on data download volume, black industries will also register multiple accounts and use multiple terminals to obtain data. At this time, it is necessary to identify the black industry terminals to improve data security; the category keywords corresponding to the terminal identity tag can be the terminal identification code, the terminal's location, the terminal's access frequency, and the terminal's data access volume. The relationship between different terminals can be determined based on the category keywords to improve data security. In this embodiment, the relationship network of electronic product terminals is optimized through category keywords and level keywords. Specifically, the relationship between terminals can be determined based on the account registration information submitted by the terminals. For example, if the same certification documents are submitted during account registration, and if the same certification documents are used to register different accounts, these accounts can be considered to have certain risks. The terminals corresponding to these accounts are merged in the relationship network so that terminals in the same node can only obtain a certain amount of data, which can protect data security and make it easier to restrict black industries.

[0054] Limit the access rights of terminals merged into the same node to improve data security. For example, set the maximum data acquisition amount for terminals in the same node so that when the data acquisition amount in the node reaches a certain level, data cannot be acquired. You can also set terminals in nodes with more serious data acquisition situations to be unable to access data, so as to prevent black industries from obtaining data.

[0055] In a real-world scenario, taking a user relationship network as an example, the category keyword corresponding to the friend tag can be "friend," and the friend category can correspond to three category levels:

[0056] a. First-degree friendship: Keywords that can usually be used to identify friendships, such as "classmate," "[older / younger / older]+[surname]," "[surname]+[brother / sister / brother / uncle]," or some reduplicated titles.

[0057] b. Second-level friend relationship: For example, it can be a name or nickname, the structure is "[surname]+[nickname]", and you can choose a surname from the Hundred Family Surnames as the keyword, such as "Li Dapang" and so on.

[0058] c. Third-level friend relationship: You can directly use the name as the keyword. The structure is "[surname]+[given name]". You can choose a surname from the Hundred Family Surnames as the [surname] keyword and a common name as the [given name] keyword.

[0059] In the current relationship network mining, the basis of the relationship is the interactive information on the social network, but it does not distinguish the strength of the relationship between people. Instead, it regards the relationship as equal, and constructs the relationship network, extracts the relationship, and mines data features. Such relationships are redundant and weak. When building a network for information mining, relatively rough character relationships are built into it, which easily constructs a large relationship network with little information. In this solution, the relationship network will be optimized through the category and level of the relationship network, so that nodes of the same type and level are merged, and multiple information is stored through a single node, making the simplified relationship network more concise and efficient in recording various data, facilitating rapid data query and improving data processing efficiency. At the same time, the simplified relationship network can more efficiently determine the data access permission level between nodes. For example, in a social network, a node publicly sets data that can only be accessed by close relationships. Compared with the traditional relationship network, the simplified relationship network can quickly determine whether the node accessing the data has permission, thereby improving data security.

[0060] The relationship network optimization method introduced in this application can simplify the user relationship network, combine the relationship notes and counter-note information, optimize and reduce the network relationship, maintain an effective and highly intimate relationship network, and thus quickly and accurately analyze the node data.

[0061] Figure 3 The present invention is a flowchart showing a method for determining node authority of a relationship network optimized by matching specific tags according to another exemplary embodiment. Figure 3 The process 30 shown is Figure 1The detailed description of S110 “merging nodes in the node relationship network and / or deleting relationship edges between nodes according to category labels and / or level labels to optimize the relationship network” in the process shown is provided.

[0062] like Figure 3 As shown, in S302 , importance coefficients are respectively assigned to a plurality of category labels and level labels.

[0063] In a practical application, friend keywords can be classified into first-level friend relationships, second-level friend relationships, and third-level friend relationships, with the closeness of the friend relationships decreasing in sequence, and importance coefficients are assigned to each.

[0064] In S304, the relationship importance of the node is generated based on the sum of the importance coefficients corresponding to the category label and / or level label of the node. All labels included in a node can be added together to generate the relationship importance coefficient of the node.

[0065] In S306 , the nodes in the node relationship network are merged and / or the relationship edges between nodes are deleted according to the relationship importance to optimize the relationship network.

[0066] In one embodiment, a relationship importance interval can be determined, and nodes whose relationship importance is not in the relationship importance interval can be used as nodes to be optimized; the nodes to be optimized are eliminated; and / or according to the category label of the node to be optimized, the node to be optimized is merged into other nodes with the same association relationship and category label; and / or if the node connected to the node to be optimized is connected to other nodes with the same category label as the node to be optimized, the relationship edge between the node to be optimized and the node is deleted to optimize the node relationship network.

[0067] The node authority determination method of the relationship network optimized by matching specific tags in this application introduces the relationship network simplification technology. Compared with the traditional relationship network, the relationships with equal weights are separated into relationships with different weights, and a certain degree of intimacy is given to different relationships. Based on such relationships, the currently cumbersome relationship network can be optimized, the computing cost can be reduced, the effect can be guaranteed to remain unchanged, and the computing time efficiency can be improved.

[0068] 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.

[0069] 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.

[0070] Figure 4 FIG. 1 is a block diagram showing a device for determining node authority of a matching specific tag optimization relationship network according to an exemplary embodiment. Figure 4 As shown, the node authority determination device 40 for matching a specific tag optimized relationship network includes: a data module 402 , a category module 404 , a level module 406 , a matching module 408 , an optimization module 410 , and a authority module 412 .

[0071] The data module 402 is used to extract node data of multiple nodes in the node relationship network;

[0072] The category module 404 is used to obtain category keywords and category labels corresponding to multiple categories;

[0073] The level module 406 is used to obtain level keywords and level labels corresponding to the category level of each category in the multiple categories;

[0074] The matching module 408 is configured to match the plurality of nodes using the node data according to the category keywords and the level keywords, and assign category labels and / or level labels to the successfully matched nodes;

[0075] The optimization module 410 is configured to merge nodes in the node relationship network and / or delete relationship edges between nodes according to category labels and / or level labels to optimize the relationship network;

[0076] The permission module 412 is used to determine the data access permission level of each node based on the optimized relationship network.

[0077] According to the node authority determination device for matching specific tags to optimize the relationship network of the present application, the node data of multiple nodes in the node relationship network are extracted; category keywords and category labels corresponding to multiple categories are obtained; level keywords and level labels corresponding to category levels of each category in the multiple categories are obtained; the multiple nodes are matched through the node data according to the category keywords and level keywords, and category labels and / or level labels are assigned to the successfully matched nodes; the nodes in the node relationship network are merged and / or the relationship edges between nodes are deleted according to the category labels and / or level labels to optimize the relationship network; the method of determining the data access permission level of each node based on the optimized relationship network can quickly, accurately and reasonably optimize and simplify the relationship network, and then quickly determine the node authority through the optimized relationship network, thereby increasing system security and transaction security.

[0078] 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;

[0079] Memory 530, for storing computer programs;

[0080] 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 .

[0081] The communication interface 520 is used for communication between the electronic device and other devices.

[0082] 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 .

[0083] An embodiment of the present invention 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 determination method for matching a specific label optimization relationship network according to any of the above embodiments. For example, the node data of multiple nodes in the node relationship network can be extracted; category keywords and category labels corresponding to multiple categories are obtained; level keywords and level labels corresponding to category levels of each category in the multiple categories are obtained; according to the category keywords and level keywords, the multiple nodes are matched through the node data, and category labels and / or level labels are assigned to the successfully matched nodes; according to the category labels and / or level labels, the nodes in the node relationship network are merged and / or the relationship edges between the nodes are deleted to optimize the relationship network; based on the optimized relationship network, the data access permission level of each node is determined.

[0084] While the exemplary embodiments of the present application have been 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 encompass various modifications and equivalent configurations within the spirit and scope of the appended claims.

Claims

1. A method for determining node authority in a relationship network optimized by matching specific tags, characterized in that: include: Get multiple sample nodes corresponding to multiple categories; Analyze the node data of sample nodes in each category and extract the category keywords corresponding to the category; Determine a category label for the corresponding category based on the category keyword; Get the sample nodes corresponding to each category; Analyze the node data of the sample nodes of each level in each category and extract the level keywords corresponding to each category level in the category; Determine the level labels corresponding to the keywords at each level; Extract node data of multiple nodes in the node relationship network; Get category keywords and category labels corresponding to multiple categories; Get the level keywords and level labels corresponding to the category levels of each category in multiple categories; Matching the plurality of nodes using the node data according to the category keywords and the level keywords, and assigning category labels and / or level labels to the successfully matched nodes; Merging nodes in the node relationship network and / or deleting relationship edges between nodes according to category labels and / or level labels to optimize the relationship network; The data access permission level of each node is determined based on the optimized relationship network.

2. The method according to claim 1, wherein According to the category keyword and the level keyword, the plurality of nodes are matched using the node data, and category labels and / or level labels are assigned to the successfully matched nodes, including: Performing category matching on the category keywords and the node data of the plurality of nodes; After the category matching is successful, the category label is assigned to the node with successful category matching; Perform level matching on the level keywords corresponding to the category and the nodes with successful category matching; After the level matching is successful, a level label is assigned to the node with the successful level matching.

3. The method according to claim 1, wherein Merging nodes in the node relationship network and / or deleting relationship edges between nodes according to category labels and / or level labels to optimize the relationship network includes: Determine target categories and target category levels; Nodes containing target category labels and target category level labels in the node relationship network are retained to optimize the node relationship network.

4. The method according to claim 1, wherein Merging nodes in the node relationship network and / or deleting relationship edges between nodes according to category labels and / or level labels to optimize the relationship network includes: Generate relationship importance based on the category label and / or level label corresponding to the node; Nodes in the node relationship network are merged and / or relationship edges between nodes are deleted according to relationship importance to optimize the relationship network.

5. The method according to claim 4, wherein Generate relationship importance based on the category label and / or level label corresponding to the node, including: Assign importance coefficients to multiple category labels and level labels respectively; The relationship importance of the node is generated according to the sum of the importance coefficients corresponding to the category label and / or level label of the node.

6. The method according to claim 4, wherein Merging nodes in the node relationship network and / or deleting relationship edges between nodes according to relationship importance to optimize the relationship network includes: Determine a relationship importance interval, and select nodes whose relationship importance is not within the relationship importance interval as nodes to be optimized; Eliminate the node to be optimized; and / or According to the category label of the node to be optimized, the node to be optimized is merged into other nodes with the same association relationship and category label; and / or If the node connected to the node to be optimized is connected to other nodes with the same category label as the node to be optimized, the relationship edge between the node to be optimized and the node is deleted to optimize the node relationship network.

7. The method according to claim 1, wherein Also includes: Generate a node profile of the target node through the optimized node relationship network; and / or Generate a service strategy for the target node through the optimized node relationship network.

8. A device for determining node authority of a matching specific tag optimization relationship network, characterized in that: include: The data module is used to obtain multiple sample nodes corresponding to multiple categories; Analyze the node data of sample nodes in each category and extract the category keywords corresponding to the category; Determine the category label of the corresponding category based on the category keyword; obtain sample nodes corresponding to each category; analyze the node data of the sample nodes of each level in each category, extract the level keywords corresponding to each category level in the category; determine the level label corresponding to each level keyword; Extract node data of multiple nodes in the node relationship network; Category module, used to obtain category keywords and category labels corresponding to multiple categories; A level module is used to obtain level keywords and level labels corresponding to the category level of each category in multiple categories; a matching module, configured to match the plurality of nodes using the node data according to the category keywords and the level keywords, and assign category labels and / or level labels to the successfully matched nodes; an optimization module, configured to merge nodes in the node relationship network and / or delete relationship edges between nodes according to category labels and / or level labels to optimize the relationship network; The permission module is used to determine the data access permission level of each node based on the optimized relationship network.

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