A method, device, computer equipment and medium for discovering problem clues

Abstract

The embodiments of the present invention disclose a method, device, computer equipment and medium for discovering problem clues, which relate to the field of artificial intelligence. The method includes: obtaining a user relationship knowledge graph; calculating the characteristic value of the corresponding node based on the association relationship between the nodes in the user relationship knowledge graph; determining hidden problem clues according to the characteristic values, and displaying the hidden problem clues. The embodiments of the present invention discover problem clues through the user relationship knowledge graph, improve the intelligence level of problem clue discovery, and realize comprehensive and accurate discovery of potential problem clues; by providing a visual interactive method, it displays and discovers clues to abnormal behavior problems of personnel, and meets the actual business needs of related inspection work.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of artificial intelligence, and in particular to a method, device, computer equipment and medium for discovering problem clues. Background Art

[0002] In user inspection work, problem clues are generally discovered through manual subjective impressions, letters and visits, and reports. The overall work efficiency is not high, and the risk loss caused by handling the problem after it is discovered is relatively difficult to control. Since the existing problem clue discovery scheme is based on the subjective consciousness of personnel, the information source is narrow, and it is difficult to discover information with complex relationships, and it is impossible to fully and effectively discover potential problem clues. Summary of the invention

[0003] The embodiments of the present invention provide a problem clue discovery method, device, computer equipment and medium, which can improve the intelligence level of problem clue discovery.

[0004] In a first aspect, an embodiment of the present invention provides a problem clue discovery method, comprising:

[0005] Obtain user relationship knowledge graph;

[0006] Calculate the feature value of the corresponding node based on the association relationship between nodes in the user relationship knowledge graph;

[0007] Hidden problem clues are determined based on the characteristic values, and the hidden problem clues are displayed.

[0008] In a second aspect, an embodiment of the present invention further provides a problem clue discovery device, the device comprising:

[0009] Knowledge graph acquisition module, used to obtain user relationship knowledge graph;

[0010] A feature value calculation module is used to calculate the feature value of the corresponding node based on the association relationship between nodes in the user relationship knowledge graph;

[0011] The problem clue display module is used to determine the hidden problem clues according to the characteristic values ​​and display the hidden problem clues.

[0012] In a third aspect, an embodiment of the present invention further provides a computer device, the computer device comprising:

[0013] one or more processors;

[0014] a memory for storing one or more programs,

[0015] When the one or more programs are executed by the one or more processors, the one or more processors implement the problem clue discovery method as described in any embodiment of the invention.

[0016] In a fourth aspect, an embodiment of the present invention further provides a storage medium comprising computer executable instructions, which, when executed by a computer processor, are used to execute the problem clue discovery method as described in any embodiment of the present invention.

[0017] In a fifth aspect, an embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the problem clue discovery method as described in any embodiment of the present invention.

[0018] The embodiment of the present invention provides a problem clue discovery solution, which calculates the characteristic value of the corresponding node based on the association relationship between the nodes in the user relationship knowledge graph, and then determines the hidden problem clue according to the characteristic value, and displays the hidden problem clue. The embodiment of the present invention discovers problem clues through the user relationship knowledge graph, improves the intelligence level of problem clue discovery, and realizes comprehensive and accurate discovery of potential problem clues; by providing a visual interactive method, it displays and discovers clues to abnormal behavior problems of personnel, meeting the actual business needs of related inspection work. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flowchart of a problem clue discovery method provided by an embodiment of the present invention;

[0020] Figure 2 A schematic diagram for defining an abnormal relationship and abnormal problem clues;

[0021] Figure 3 A flowchart of another problem clue discovery method provided by an embodiment of the present invention;

[0022] Figure 4 A schematic diagram of the structure of the entity and relationship wide table of relationships between people;

[0023] Figure 5 A schematic diagram of the structure of the entity and relationship wide tables representing the relationship between a person and other types of entities;

[0024] Figure 6 It is a schematic diagram of a user relationship knowledge graph;

[0025] Figure 7 A flowchart of another problem clue discovery method provided by an embodiment of the present invention;

[0026] Figure 8 This is an example diagram of a close centrality algorithm;

[0027] Fig. 9 This is an example diagram of a betweenness centrality algorithm;

[0028] Fig.10 A flowchart of problem clue discovery based on a user relationship knowledge graph provided by an embodiment of the present invention;

[0029] Fig.11 A structural block diagram of a problem clue discovery device provided by an embodiment of the present invention;

[0030] Fig.12 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0032] To facilitate understanding, the words that may appear below are explained.

[0033] A data lake is a repository or system that stores data in its original format. It stores data as is without structuring it in advance. A data lake can store structured data, semi-structured data, unstructured data, binary data, etc.

[0034] A data warehouse is a subject-oriented, integrated, relatively stable data collection that reflects historical changes and is used to support management decisions.

[0035] A data mart, also known as a data market, is a data mart that meets the needs of specific departments or users and is stored in a multi-dimensional manner, including defining dimensions, indicators that need to be calculated, dimensional levels, etc., to generate data cubes for decision-making analysis needs.

[0036] Data lineage refers to the origin and development of data, mainly including the source of data, data processing methods, mapping relationships and data exports. In the actual application of data, the original data will be processed in multiple steps according to actual needs, and finally new data will be generated. In this process, many data tables will be generated. The link relationship between these data tables is called data lineage.

[0037] The knowledge graph is a semantic network that reveals the relationship between entities. The knowledge graph can be divided into two levels in terms of logical structure: the model layer and the data layer. The data layer is mainly composed of a series of knowledge, and the knowledge is stored in units of facts. If a triple such as (entity 1, relationship, entity 2) and (entity, attribute, attribute value) is used to express facts, a graph database can be selected as the storage medium. The model layer is built on the data layer, mainly through the ontology library to standardize a series of fact expressions in the data layer. Ontology is a conceptual template for a structured knowledge base that lists the types of entities. The relationships that connect them and the limitations on how entities and relationships are combined. It can also be understood that, to some extent, the ontology defines the rules for how entities are connected in the world.

[0038] The general process of problem clue discovery is as follows:

[0039] 1) Personnel related to the inspection work shall verify and authenticate the petitions and reports.

[0040] 2) Identify clues to problems based on the subjective impressions of the subjects by those involved in the investigation.

[0041] 3) Data collection and acquisition: perform data ETL (data extraction, transformation, and loading) processing on the collected data, and uniformly load and store it in the enterprise data lake and data warehouse.

[0042] 4) Establish a personnel problem clue analysis library through predefined rules, and then use it to evaluate whether the problem clues are valid.

[0043] The above problem clue discovery method is based on the subjective consciousness of personnel to identify problem clues. The information source is narrow. It is difficult to discover hidden problem clues for information with complex relationships. The intelligence level is low and the accuracy is not high.

[0044] In the case where it is difficult to discover hidden problem clues using predefined rules, an embodiment of the present invention constructs and uses a user relationship knowledge graph to assist in mining hidden problem clues.

[0045] Among them, the user relationship knowledge graph is a knowledge graph for discovering clues to abnormal behavior problems of personnel, which is established by extracting relevant data and management requirements from actual business processes and various rules and regulations. Since user inspection management is a management process centered on personnel, and personnel often involve various complex relationships, the knowledge graph can be well integrated into user inspection management due to its relationship expression ability. According to regulatory requirements, the existing data is summarized and abstracted, and entities and relationships that can express problem clues are extracted, and a user relationship knowledge graph for discovering clues to personnel problems is formed based on entities and relationships.

[0046] Specifically, based on the collection, acquisition, processing, analysis and mining of multi-dimensional key data in the data mart, such as financial reimbursement, human resources, travel, credit, deposits, transaction flows, procurement, real estate, non-performing assets, risk scoring, air travel, industrial and commercial enterprises, internal control compliance, anti-money laundering, anti-fraud, inspections, etc., a more intelligent user relationship knowledge graph is constructed to help improve the digital, scientific and intelligent management of problem clue discovery.

[0047] It should be noted that the collection, storage, use and processing of user personal information in the technical solution of the present invention comply with relevant national laws and regulations on personal information security.

[0048] Figure 1 This is a flowchart of a problem clue discovery method provided by an embodiment of the present invention. This embodiment is applicable to the situation of monitoring abnormal behavior problems of personnel. The method can be executed by a problem clue discovery device, which can be implemented by software and / or hardware and configured in a computer device. The method includes:

[0049] Step 110: Obtain the user relationship knowledge graph.

[0050] Exemplarily, when it is necessary to monitor clues of abnormal personnel problems, a user relationship knowledge graph pre-constructed based on data in a data mart is obtained.

[0051] Step 120: Calculate the feature value of the corresponding node based on the association relationship between the nodes in the user relationship knowledge graph.

[0052] The association relationship may be the degree of influence of a node on other nodes. For example, the association relationship includes the difficulty of a node reaching other nodes and the proportion of times a node serves as a bridge for the shortest path between two other nodes.

[0053] Exemplarily, the reciprocal of the average distance between each node and other nodes in the user relationship knowledge graph is calculated, and the reciprocal is used as the closeness centrality value of the corresponding node; the number of times any node in the user relationship knowledge graph serves as a bridge for the shortest path between two other nodes and the number of shortest paths between the two corresponding nodes are determined, and the ratio of the number of times to the number of shortest paths is calculated and the calculation result is used as the betweenness centrality value of the corresponding node; the node identifier, closeness centrality value and betweenness centrality value of the associated storage node are stored.

[0054] Optionally, the node identifier, the closeness centrality value and the betweenness centrality value of the corresponding node can be associated and stored through a data table. The closeness centrality value and the betweenness centrality value can also be added to the triple as attributes to generate a user relationship knowledge graph through the updated triple to display the corresponding closeness centrality value and betweenness centrality value in the node of the user relationship knowledge graph.

[0055] Step 130: Determine hidden problem clues according to the characteristic values, and display the hidden problem clues.

[0056] Exemplarily, the characteristic value of each node is compared with a preset threshold value to determine the target node whose characteristic value is greater than the preset threshold value; the target node and other nodes connected to the target node by edges are used as clues to hidden problems, and the hidden problem clues are displayed. Since there is more than one characteristic value, the close centrality value of each node can be compared with the preset close centrality threshold to obtain a first comparison result; the betweenness centrality value of each node is compared with the preset betweenness centrality threshold to obtain a second comparison result. If at least one of the first comparison result and the second comparison result is greater than the corresponding threshold, the node is determined to be a target node with a characteristic value greater than the preset threshold value. The target node and other nodes connected to the target node by edges are used as clues to hidden problems, that is, it is necessary to examine other nodes that are connected to the target node to determine whether there are abnormal behavior problems between entities.

[0057] Optionally, when displaying the user relationship knowledge graph, the user can be reminded by displaying the edge between the target node and other nodes in a preset color. The user can also be reminded by displaying the borders of the target node and other nodes in a preset color when displaying the user relationship knowledge graph. Alternatively, the user can be reminded by filling the borders of the target node and other nodes with a preset color when displaying the user relationship knowledge graph.

[0058] Optionally, before comparing the characteristic value of each node with the preset threshold, the method further includes: querying a preset data table according to the node identifier to obtain the characteristic value of the corresponding node, wherein the preset data table stores the node identifier and the characteristic value in an associated manner. For example, querying the preset data table according to the node identifier to obtain the closeness centrality value and the betweenness centrality value of the corresponding node.

[0059] Optionally, before comparing the feature value of each node with a preset threshold, the method further includes: obtaining the feature value corresponding to each node in the user relationship knowledge graph, wherein the feature value is displayed in the node of the user relationship knowledge graph. For example, the triple data of the corresponding node is queried through the node identifier to obtain the closeness centrality value and the betweenness centrality value.

[0060] The technical solution of this embodiment calculates the characteristic value of the corresponding node based on the association relationship between the nodes in the user relationship knowledge graph, and then determines the hidden problem clues according to the characteristic values, and displays the hidden problem clues. The embodiment of the present invention discovers problem clues through the user relationship knowledge graph, improves the intelligent level of problem clue discovery, and realizes comprehensive and accurate discovery of potential problem clues; by providing a visual interactive method, it displays and discovers clues to abnormal behavior problems of personnel, meeting the actual business needs of related inspection work.

[0061] On the basis of the above technical solution, after determining the hidden problem clues according to the characteristic values, the following additional technical features may also be included: using abnormal relationship rules to verify the validity of the hidden problem clues. Among them, the abnormal relationship rules are the judgment rules for abnormal behavior problems summarized from the existing data according to regulatory requirements. The abnormal relationship rules include but are not limited to: using one's own account or control account to have financial transactions with corporate credit customers and their affiliated companies, or having financial transactions with the personal accounts of senior executives of corporate credit customers and their affiliated companies; assisting enterprises or individuals to obtain bank loans and then re-lending them to others at high interest rates; raising loan repayment funds for credit customers for "recovery and re-lending"; whether there are financial transactions with customers.

[0062] Figure 2 A schematic diagram for defining an abnormal relationship and abnormal problem clues. Figure 2 The definition of abnormal relationships between people and abnormal problem clues based on abnormal relationship rules is presented in the form of a mind map. For employee-related violation events, the person himself is the associated object of the abnormal event. For example, employee-related violation events include collusion, abnormal punishment, non-performing loans, and suspended promotion, etc. For violation associations, there is a normal relationship between the person and the core object of the abnormal event. For example, the person is a relative, colleague, or alumni of the core object of the collusion bidding. For abnormal identical mobile phone numbers, there should be no relationship in the normal behavior reasoning of the person. For example, leaving one's own mobile phone number in other people's credit card and other account information. For abnormal fund transactions, frequent large-value transactions with the same counterparty in a short period of time are suspected to be abnormal fund transactions.

[0063] Since the hidden abnormal behavior problem clues discovered through the user relationship knowledge graph may sometimes be invalid clues, after the embodiment of the present invention determines the hidden problem clues according to the characteristic values, it uses pre-defined abnormal relationship rules to verify the validity of the hidden problem clues, which can eliminate invalid hidden problem clues, reduce the amount of data of hidden problem clues that need to be verified by users, and improve the verification efficiency of hidden problem clues.

[0064] On the basis of the above technical solution, before obtaining the user relationship knowledge graph, it also includes additional technical features for constructing the user relationship knowledge graph.

[0065] Figure 3 A flowchart of another problem clue discovery method provided by an embodiment of the present invention, such as Figure 3 As shown, the method includes:

[0066] Step 310: Obtain the model building data table and relationship detail table.

[0067] Among them, the model construction data table is used to store the entity information for building the knowledge graph, and the relationship details table is used to store the relationship between entities.

[0068] For example, business data that can be used to construct a knowledge graph can be obtained and processed from existing data lakes and data warehouses based on the analysis of data sources and data lineage. Optionally, business data can be filtered out personnel-centric data and integrated into a model-constructed data table representing entity information as shown in Table 1.

[0069] Table 1 is a model construction data table.

[0070] Serial number Table name 1 Label table for collusion between relatives of personnel and suppliers 2 Personnel Relatives Avoidance Label Table 3 Personnel Debit Card Information Form 4 Basic information table of personnel 5 Suspected Collusion Quotation Label Traceability Table …… ……

[0071] Among them, around the expression of personnel, entity types include personnel, events, organizations, mobile phone numbers, accounts, equipment, etc. More specifically, personnel include personnel, customers, relatives, etc. Events mainly include illegal events involving various types of personnel in information records. For example, suspended promotion, bad loans, involvement in cases, etc., personnel can have relationships through events. Organizations mainly involve organizations in which different personnel have relationships, such as work organizations, schools, etc. That is, people in the same work organization may be colleagues. People in the same school may be classmates. Mobile phone numbers are private items of users. In the inspection business, mobile phone numbers are necessary information and are retained in various events. Private mobile phones may be registered under the names of other people as auxiliary tools, resulting in abnormal associations, so they are also represented as entities. Accounts can include unique identifiers, card numbers, account numbers, and customer numbers. It should be noted that entity types are not limited to the above-mentioned types. If new types of entities are generated during the actual business operation, the entity information of the new types of entities can be added to the model construction data table.

[0072] By mining the relationships between various entities, a relationship table showing the relationship between entities is constructed as shown in Table 2.

[0073]

[0074]

[0075] Among them, the relationships between entities include relationships between people, relationships between people and accounts, relationships between people and mobile phone numbers, relationships between people and events, relationships between people and organizations, relationships between accounts, relationships between accounts and mobile phone numbers, relationships between accounts and events, relationships between accounts and organizations, relationships between mobile phone numbers and organizations, and relationships between events and organizations. It should be noted that the relationships between entities are not limited to the relationships listed above. If new relationship types are generated during the actual business operation process, the new relationship types can be included in the relationship details table of the above-mentioned entity-entity relationships.

[0076] Step 320: Use a knowledge graph generation tool to construct a user relationship knowledge graph based on the model to build a data table and a relationship detail table.

[0077] Exemplarily, the entities included in the model construction data table are determined; the relationships between the entities are determined based on the relationship details table; the triples used to construct the knowledge graph are determined based on the entities and relationships, and the user relationship knowledge graph is constructed based on the triples.

[0078] Specifically, before constructing a user relationship knowledge graph through a model-based data table and a relationship detail table, it is also necessary to design entity and relationship wide tables based on the above two types of data tables to construct triples that form the knowledge graph. Figure 4 Schematic diagram of the structure of the entity and relationship wide tables for relationships between people. Figure 5 The schematic diagram of the structure of the entity and relationship wide table representing the relationship between a person and other types of entities. Optionally, before generating the user relationship knowledge graph, it also includes defining entity types and relationship types.

[0079] Exemplarily, multiple entity and relationship wide tables are generated based on the model construction data table and the relationship detail table, so that the data of the model construction data table and the relationship detail table of entity and entity relationship meet the processing requirements of the graph database. Among them, the above-mentioned entity and relationship wide tables may include entity and relationship triple relationships. The entity and relationship wide tables are input into the graph database to build a user relationship knowledge graph through the graph database. Figure 6 A schematic diagram of a user relationship knowledge graph. Figure 6 It shows the relationship between people, organizations, accounts, contact information, debit card accounts, and violations. The entity type definition and relationship type definition are shown in the following table.

[0080] Table 3 is the entity type definition table.

[0081]

[0082] Table 4 is a relationship type definition table.

[0083] Relationship Name Relationship Type Start Node End Node relatives RELATIVES Person Person Fellow townsman COUNTRYMEN Person Person Alumni ALUMNUS Person Person colleague COLLEAGUE Person Person guarantee GUARANTEE Person Person Account Affiliation ACCOUNT_BELONG_TO Account Person Mobile phone number PHONENUMBER_BELONG_TO PhoneNumber Person Mobile phone number PHONENUMBER_BELONG_TO PhoneNumber Organization Abnormal event correlation RELATE_TO_EVENT Organization Event Abnormal event correlation RELATE_TO_EVENT Person Event director DIRECTOR Person External institutions Supervisor SUPERVISOR Person External institutions Executives MANAGEMENT Person External institutions Affiliated Institutions BELONG_TO_ORG Person InternalOrganization trade TRANSACTION Account Account

[0084] Step 330: Obtain the user relationship knowledge graph.

[0085] Step 340: Calculate the feature value of the corresponding node based on the association relationship between the nodes in the user relationship knowledge graph.

[0086] Step 350: Determine hidden problem clues according to the characteristic values, and display the hidden problem clues.

[0087] The technical solution of the embodiment of the present invention constructs a user relationship knowledge graph by extracting relevant data and management requirements from the actual business processes and various rules and regulations, thereby improving the intelligence level of problem clue discovery.

[0088] On the basis of the above technical solution, after obtaining the user relationship knowledge graph, it also includes the following additional technical features: cropping the user relationship knowledge graph in response to the user's query information, and displaying the cropped user relationship knowledge graph.

[0089] Figure 7 A flowchart of another problem clue discovery method provided by an embodiment of the present invention is as follows: Figure 7 As shown, the method includes:

[0090] Step 710: Obtain the user relationship knowledge graph.

[0091] Step 720: Calculate the feature value of the corresponding node based on the association relationship between the nodes in the user relationship knowledge graph.

[0092] For example, the eigenvalues ​​include closeness centrality and betweenness centrality. Calculate the difficulty of a node to reach other nodes, that is, calculate the inverse of the average distance from a node to all other nodes. It can be expressed as: raw closeness centrality (node) = 1 / sum (distance from node to all other nodes)

[0093] Figure 8 is an example diagram of a close centrality algorithm. Figure 8 The relationship between the nodes in the network can infer that node C has the shortest average distance to other nodes, followed by node B and node D, and the longest average distance to other nodes is node A and node E. The calculation results of the close centrality calculated using the above formula can verify the above conclusion.

[0094] Betweenness centrality is a method to detect the degree of influence of a node on the information flow in a graph. All shortest paths between any two nodes in the network are calculated. If many of these paths pass through a certain node, then the betweenness centrality of this node is considered high. The specific calculation formula is as follows:

[0095]

[0096] Among them, σ st represents the number of shortest paths from node s to node t; σ st (v) represents the number of nodes v passed through in the shortest path from node s to node t.

[0097] Fig. 9is an example diagram of the betweenness centrality algorithm. Fig. 9 From the middle nodes and relationships, we can find that node A has the most associations, and we can infer that this node has the greatest impact on the information flow side, that is, the node has the highest betweenness centrality, followed by node C. The calculation results of the betweenness centrality calculated using the calculation formula can verify the above conclusion.

[0098] Step 730: Determine hidden problem clues based on the feature values.

[0099] Optionally, after determining the hidden problem clues according to the feature values, the preset abnormal relationship rules can be used to verify whether the hidden problem clues are valid. Alternatively, the abnormal behavior problem features trained by the machine learning model can be used to verify whether the hidden problem clues are valid.

[0100] Step 740: crop the user relationship knowledge graph in response to the user's query information, and display hidden problem clues in the cropped user relationship knowledge graph.

[0101] Specifically, obtain the employee ID input by the user; select the corresponding first node from the user relationship knowledge graph according to the employee ID, delete the second nodes in the user relationship knowledge graph that are not connected to the first node, and delete the edges between the second nodes; and display hidden problem clues based on the deleted user relationship knowledge graph.

[0102] Optionally, the user relationship knowledge graph is cropped in response to the user's query information, and hidden problem clues are displayed in the cropped user relationship knowledge graph, including: obtaining the employee identifications of at least two persons to be queried input by the user; determining at least two corresponding fourth nodes from the user relationship knowledge graph according to the employee identifications, deleting the fifth node of the user relationship knowledge graph except the fourth node, and deleting the edges between the fifth nodes; and displaying hidden problem clues based on the deleted user relationship knowledge graph.

[0103] It should be noted that there are many ways to display hidden problem clues in the user relationship knowledge graph, and the embodiments of the present invention do not specifically limit them. For example, the hidden problem clues can be displayed in the user relationship knowledge graph by adjusting the display attributes of the edges corresponding to the hidden problem clues. The hidden problem clues can be displayed in the user relationship knowledge graph by adjusting the display attributes of the nodes corresponding to the hidden problem clues. The display attributes of the edges and nodes corresponding to the hidden problem clues can also be adjusted at the same time.

[0104] Step 750: In response to the clue push request input by the user, the hidden problem clue is pushed to the superior user of the user to instruct the superior user to determine whether the hidden problem clue is valid.

[0105] Step 760: In response to the superior user's cancel operation on the invalid hidden problem clue, modify the display attribute of the edge or node corresponding to the invalid hidden problem clue.

[0106] The technical solution of the embodiment of the present invention provides a visual interactive mode, provides strong support for business and meets the actual business needs of related inspection work.

[0107] Fig.10 A flowchart for problem clue discovery based on a user relationship knowledge graph provided in an embodiment of the present invention. Fig.10 The steps described have already been Figure 1 For greater clarity, Fig.10 The key steps are highlighted. Fig.10 As shown, the main process of problem clue discovery is presented. First, access the data source, and then uniformly process the data from the data source. The standardized data after uniform processing enters the data warehouse, and the data is processed in the data warehouse. Subsequently, a user relationship knowledge graph is constructed through the processed data. The close centrality algorithm and the betweenness centrality algorithm are used to calculate the close centrality value and the betweenness centrality value of each node in the user relationship knowledge graph. Hidden problem clues are discovered based on the close centrality value and the betweenness centrality value. After the current user completes the validity judgment based on the discovered problem clues, the problem clues are pushed to a user with a higher level of authority to request the user with a higher level of authority to judge whether the hidden problem clues are valid. Optionally, after discovering the hidden problem clues, the validity of the discovered hidden problem clues can also be verified based on abnormal relationship rules and / or abnormal behavior features trained by machine learning algorithms.

[0108] The use of the present invention can adapt to ever-changing business needs, and use the user relationship knowledge graph to improve the digitization, intelligence and scientific nature of problem clue discovery; by implementing the embodiments of the present invention, the effectiveness and accuracy of problem clue discovery can be improved, and the relevant needs of problem clue discovery in various businesses can be met to a certain extent; a more friendly application and display interface is provided for users, making the user operation simple, convenient and easy to use.

[0109] Fig.11 This is a structural block diagram of a problem clue discovery device provided by an embodiment of the present invention. The device can execute the problem clue discovery method of any embodiment of the present invention, and realize hidden problem clue discovery by executing the above method. The device can be implemented by software and / or hardware and configured in a computer device. Fig.11 As shown, the device comprises:

[0110] The knowledge graph acquisition module 1110 is used to acquire the user relationship knowledge graph;

[0111] The feature value calculation module 1120 is used to calculate the feature value of the corresponding node based on the association relationship between the nodes in the user relationship knowledge graph;

[0112] The problem clue display module 1130 is used to determine the hidden problem clues according to the characteristic values ​​and display the hidden problem clues.

[0113] An embodiment of the present invention provides a problem clue discovery device, which discovers problem clues through a user relationship knowledge graph, improves the intelligence level of problem clue discovery, and realizes comprehensive and accurate discovery of potential problem clues; by providing a visual interactive method, it displays and discovers clues to abnormal behavior problems of personnel, thereby meeting the actual business needs of related inspection work.

[0114] Optionally, the device further comprises:

[0115] The clue verification module is used to verify the validity of the hidden problem clue by using the abnormal relationship rule after determining the hidden problem clue according to the characteristic value.

[0116] Optionally, the device further comprises:

[0117] The graph construction module is used to obtain the model construction data table and the relationship detail table before obtaining the user relationship knowledge graph, wherein the model construction data table is used to store the entity information for constructing the knowledge graph, and the relationship detail table is used to store the relationship between entities; the knowledge graph generation tool is used to construct the user relationship knowledge graph based on the model construction data table and the relationship detail table.

[0118] Optionally, the graph modeling module is further specifically used for:

[0119] Determine the entities included in the model building data table;

[0120] Determine the relationship between entities according to the preset relationship details table;

[0121] Triples used to construct a knowledge graph are determined based on entities and relationships, and a user relationship knowledge graph is constructed based on the triples.

[0122] Optionally, the eigenvalue calculation module 1120 is specifically used for:

[0123] Calculate the reciprocal of the average distance between each node and other nodes in the user relationship knowledge graph, and use the reciprocal as the close centrality value of the corresponding node;

[0124] According to the number of times any node in the user relationship knowledge graph serves as a bridge for the shortest path between two other nodes and the number of shortest paths between the two corresponding nodes, the betweenness centrality value of the corresponding node is calculated;

[0125] The association stores the node identity, closeness centrality value, and betweenness centrality value of the node.

[0126] Optionally, the problem clue display module 1130 is specifically used for:

[0127] Compare the characteristic value of each node with the preset threshold, and determine the target node whose characteristic value is greater than the preset threshold;

[0128] The target node and other nodes connected to the target node through edges are taken as hidden problem clues, and the hidden problem clues are displayed.

[0129] Optionally, the device further comprises:

[0130] The characteristic value determination module is used to query a preset data table according to the node identifier to obtain the characteristic value of the corresponding node before comparing the characteristic value of each node with a preset threshold, wherein the preset data table stores the node identifier and the characteristic value in an associated manner.

[0131] Optionally, the feature value determination module is also used to obtain the feature value corresponding to each node in the user relationship knowledge graph before comparing the feature value of each node with a preset threshold, wherein the feature value is displayed in the node of the user relationship knowledge graph.

[0132] Optionally, the problem clue display module is further specifically used for:

[0133] The user relationship knowledge graph is cropped in response to the user's query information, and hidden problem clues are displayed in the cropped user relationship knowledge graph.

[0134] Optionally, the problem clue display module is further used for:

[0135] Get the employee ID entered by the user;

[0136] Select a corresponding first node from the user relationship knowledge graph according to the employee ID, delete a second node in the user relationship knowledge graph that is not connected to the first node, and delete the edge between the second nodes;

[0137] Display hidden problem clues based on the deleted user relationship knowledge graph.

[0138] Optionally, the problem clue display module is further used for:

[0139] Obtain at least two employee IDs of the persons to be queried input by the user;

[0140] Determine at least two corresponding fourth nodes from the user relationship knowledge graph according to the employee identifier, delete the fifth nodes in the user relationship knowledge graph except the fourth nodes, and delete the edges between the fifth nodes;

[0141] Display hidden problem clues based on the deleted user relationship knowledge graph.

[0142] Optionally, the device further comprises:

[0143] The clue judgment module is used to determine the hidden problem clue according to the characteristic value, display the hidden problem clue, and then, in response to the clue push request input by the user, push the hidden problem clue to the user's superior user to instruct the superior user to judge whether the hidden problem clue is valid; in response to the superior user's cancel clue operation on the invalid hidden problem clue, modify the display attributes of the edge or node corresponding to the invalid hidden problem clue.

[0144] The problem clue discovery device provided in the embodiment of the present invention can execute the problem clue discovery method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0145] Fig.12 A schematic diagram of the structure of a computer device provided by an embodiment of the present invention is shown in FIG. Fig.12 As shown, the computer device includes a processor 1210, a memory 1220, an input device 1230, and an output device 1240; the number of processors 1210 in the computer device can be one or more. Fig.12 A processor 1210 is taken as an example; the processor 1210, the memory 1220, the input device 1230 and the output device 1240 in the computer device can be connected via a bus or other means. Fig.12 The example of connecting through bus is taken in the following.

[0146] The memory 1220, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the problem clue discovery method in the embodiment of the present invention (for example, the knowledge graph acquisition module 1110, the feature value calculation module 1120 and the problem clue display module 1130 in the problem clue discovery device). The processor 1210 executes various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 1220, that is, realizes the above-mentioned problem clue discovery method.

[0147] The memory 1220 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 1220 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 1220 may further include a memory remotely arranged relative to the processor 1210, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0148] The input device 1230 may be used to receive input digital or character information and generate key signal input related to user settings and function control of the computer device. The output device 1240 may include a display device such as a display screen. The knowledge graph and hidden question clues are displayed through the output device 1240.

[0149] The embodiment of the present invention further provides a storage medium containing computer executable instructions, and the computer executable instructions are used to execute a problem clue discovery method when executed by a computer processor, the method comprising:

[0150] Obtain user relationship knowledge graph;

[0151] Calculate the feature value of the corresponding node based on the association relationship between nodes in the user relationship knowledge graph;

[0152] Hidden problem clues are determined based on the characteristic values, and the hidden problem clues are displayed.

[0153] Of course, the storage medium containing computer executable instructions provided by an embodiment of the present invention, whose computer executable instructions are not limited to the method operations described above, can also execute related operations in the problem clue discovery method provided by any embodiment of the present invention.

[0154] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0155] It is worth noting that in the embodiment of the above-mentioned problem clue discovery device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0156] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A problem clue discovery method, characterized in that: include: Obtain user relationship knowledge graph; Calculate the reciprocal of the average distance between each node and other nodes in the user relationship knowledge graph, and use the reciprocal as the closeness centrality value of the corresponding node; calculate the betweenness centrality value of the corresponding node according to the number of times any node in the user relationship knowledge graph serves as a bridge for the shortest path between two other nodes and the number of shortest paths between the two corresponding nodes; associate and store the node identifier, closeness centrality value and betweenness centrality value of the node; Adding the closeness centrality value and the betweenness centrality value as attributes to the triplet to generate a user relationship knowledge graph through the updated triplet, and displaying the corresponding closeness centrality value and betweenness centrality value in the nodes of the newly generated user relationship knowledge graph; Determining hidden problem clues according to the closeness centrality value and the betweenness centrality value, and displaying the hidden problem clues, including: Compare the closeness centrality value of each node with the preset closeness centrality threshold to obtain a first comparison result; compare the betweenness centrality value of each node with the preset betweenness centrality threshold to obtain a second comparison result; if the first comparison result and the second comparison result are greater than the corresponding threshold, determine that the node is a target node; and use the target node and other nodes that have edge connections with the target node as clues to hidden problems; The validity of the hidden problem clues is verified by using abnormal relationship rules, wherein the abnormal relationship rules are the judgment rules for abnormal behavior problems summarized from existing data according to regulatory requirements; In response to the user's query information, the user relationship knowledge graph is cropped, and hidden problem clues are displayed in the cropped user relationship knowledge graph, wherein displaying the hidden problem clues includes adjusting display properties of edges or nodes corresponding to the hidden problem clues; In response to a clue push request input by a user, the hidden problem clue is pushed to a superior user of the user to instruct the superior user to determine whether the hidden problem clue is valid; In response to a superior user's cancelling operation on an invalid hidden problem clue, a display attribute of an edge or a node corresponding to the invalid hidden problem clue is modified.

2. The method according to claim 1, characterized in that: Before obtaining the user relationship knowledge graph, it also includes: Obtain a model construction data table and a relationship detail table, wherein the model construction data table is used to store entity information for constructing a knowledge graph, and the relationship detail table is used to store the relationship between entities; A knowledge graph generation tool is used to construct a data table and a relationship detail table based on the model to construct a user relationship knowledge graph.

3. The method according to claim 2, characterized in that The method of using a knowledge graph generation tool to construct a data table and a relationship detail table based on the model to construct a user relationship knowledge graph includes: Determine the entities included in the model building data table; Determine the relationship between the entities according to the relationship list; Triples for constructing a knowledge graph are determined according to the entities and relationships, and a user relationship knowledge graph is constructed based on the triples.

4. The method according to claim 1, characterized in that Before comparing the closeness centrality value of each node with the preset closeness centrality threshold to obtain the first comparison result; and comparing the betweenness centrality value of each node with the preset betweenness centrality threshold to obtain the second comparison result, the method further includes: The preset data table is queried according to the node identifier to obtain the closeness centrality value and the betweenness centrality value of the corresponding node, wherein the node identifier, the closeness centrality value and the betweenness centrality value are stored in the preset data table in an associated manner.

5. The method according to claim 1, characterized in that The step of cropping the user relationship knowledge graph in response to the user's query information and displaying hidden problem clues in the cropped user relationship knowledge graph includes: Get the employee ID entered by the user; Selecting a corresponding first node from the user relationship knowledge graph according to the employee identifier, deleting a second node in the user relationship knowledge graph that is not connected to the first node, and deleting an edge between the second nodes; Display hidden problem clues based on the deleted user relationship knowledge graph.

6. The method according to claim 1, characterized in that The step of cropping the user relationship knowledge graph in response to the user's query information and displaying hidden problem clues in the cropped user relationship knowledge graph includes: Obtain at least two employee IDs of the persons to be queried input by the user; Determine at least two corresponding fourth nodes from the user relationship knowledge graph according to the employee identifier, delete the fifth node of the user relationship knowledge graph except the fourth node, and delete the edges between the fifth nodes; Display hidden problem clues based on the deleted user relationship knowledge graph.

7. A problem clue discovery device, characterized in that: include: Knowledge graph acquisition module, used to obtain user relationship knowledge graph; A feature value calculation module is used to calculate the reciprocal of the average distance between each node and other nodes in the user relationship knowledge graph, and use the reciprocal as the closeness centrality value of the corresponding node; calculate the betweenness centrality value of the corresponding node according to the number of times any node in the user relationship knowledge graph serves as a bridge for the shortest path between two other nodes and the number of shortest paths between two corresponding nodes; and store the node identifier, closeness centrality value and betweenness centrality value of the node in association; Adding the closeness centrality value and the betweenness centrality value as attributes to the triplet to generate a user relationship knowledge graph through the updated triplet, and displaying the corresponding closeness centrality value and betweenness centrality value in the nodes of the newly generated user relationship knowledge graph; A problem clue display module, used to determine hidden problem clues according to the closeness centrality value and the betweenness centrality value, and display the hidden problem clues, including: Compare the closeness centrality value of each node with the preset closeness centrality threshold to obtain a first comparison result; compare the betweenness centrality value of each node with the preset betweenness centrality threshold to obtain a second comparison result; if the first comparison result and the second comparison result are greater than the corresponding threshold, determine that the node is a target node; and use the target node and other nodes that have edge connections with the target node as clues to hidden problems; The validity of the hidden problem clues is verified by using abnormal relationship rules, wherein the abnormal relationship rules are the judgment rules for abnormal behavior problems summarized from existing data according to regulatory requirements; In response to the user's query information, the user relationship knowledge graph is cropped, and hidden problem clues are displayed in the cropped user relationship knowledge graph, wherein displaying the hidden problem clues includes adjusting display properties of edges or nodes corresponding to the hidden problem clues; In response to a clue push request input by a user, the hidden problem clue is pushed to a superior user of the user to instruct the superior user to determine whether the hidden problem clue is valid; In response to a superior user's cancelling operation on an invalid hidden problem clue, a display attribute of an edge or a node corresponding to the invalid hidden problem clue is modified.

8. A computer device, characterized in that: The computer device comprises: one or more processors; a memory 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 problem clue discovery method as described in any one of claims 1-6.

9. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions are used to execute the problem clue discovery method as described in any one of claims 1-6 when executed by a computer processor.

Images