CN117332298A - Object recognition method, device, terminal, storage medium and program product - Google Patents
Object recognition method, device, terminal, storage medium and program product Download PDFInfo
- Publication number
- CN117332298A CN117332298A CN202311316145.8A CN202311316145A CN117332298A CN 117332298 A CN117332298 A CN 117332298A CN 202311316145 A CN202311316145 A CN 202311316145A CN 117332298 A CN117332298 A CN 117332298A
- Authority
- CN
- China
- Prior art keywords
- account
- node
- switching
- information
- event
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 80
- 238000003860 storage Methods 0.000 title claims abstract description 26
- 230000008569 process Effects 0.000 claims abstract description 21
- 238000013475 authorization Methods 0.000 claims description 26
- 238000000605 extraction Methods 0.000 claims description 16
- 230000004927 fusion Effects 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 7
- 230000000977 initiatory effect Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000013473 artificial intelligence Methods 0.000 abstract description 21
- 230000008859 change Effects 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 28
- 230000003993 interaction Effects 0.000 description 19
- 230000006870 function Effects 0.000 description 16
- 238000012545 processing Methods 0.000 description 15
- 238000010586 diagram Methods 0.000 description 13
- 238000010801 machine learning Methods 0.000 description 7
- 238000012546 transfer Methods 0.000 description 7
- 230000002093 peripheral effect Effects 0.000 description 6
- 230000004044 response Effects 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 239000013598 vector Substances 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000012790 confirmation Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000002452 interceptive effect Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 238000005295 random walk Methods 0.000 description 2
- 238000009877 rendering Methods 0.000 description 2
- 238000012502 risk assessment Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000013526 transfer learning Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/042—Knowledge-based neural networks; Logical representations of neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4016—Transaction verification involving fraud or risk level assessment in transaction processing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/14—Network architectures or network communication protocols for network security for detecting or protecting against malicious traffic
- H04L63/1408—Network architectures or network communication protocols for network security for detecting or protecting against malicious traffic by monitoring network traffic
- H04L63/1416—Event detection, e.g. attack signature detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- Computer Security & Cryptography (AREA)
- Evolutionary Computation (AREA)
- Accounting & Taxation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Computer Hardware Design (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses an object identification method, an object identification device, a terminal, a storage medium and a program product, and relates to the field of artificial intelligence. The method comprises the following steps: acquiring account switching data generated by the change of an account interacted with a designated account; constructing a graph network structure based on account switching data; extracting features of the graph network structure to obtain node feature representation corresponding to the nodes; fusing the first node characteristic representation and a first information characteristic representation corresponding to the account information of the first account to obtain a first account characteristic representation; and identifying the characteristic representation of the first account to obtain an identification result corresponding to the first account. By constructing a graph network structure corresponding to the switching relation between the accounts, the hidden account risk condition is acquired from the account switching data, and when the account is risk-identified by combining the characteristics related to the account switching data and the account information, the accuracy of risk prediction is improved, and the safety coefficient in the application program use process is improved.
Description
Technical Field
Embodiments of the present application relate to the field of artificial intelligence, and in particular, to an object recognition method, device, terminal, storage medium, and program product.
Background
Application programs are important use ways in the use process of electronic equipment. The user uses various functions in the application by logging in an account in the application, such as: transaction functions, interaction functions, information query functions, etc. However, in the use of transaction functions, there are typically risk accounts that affect the security of the transaction functions in the application.
In the related art, when identifying a risk account, it is generally required to acquire data such as a resource transaction network environment and account history transaction information of the account after obtaining authorization, and analyze the acquired data, so as to determine whether the account belongs to the risk account, and further determine whether continuous transaction can be performed.
However, in the above manner of analyzing and identifying the risk account, some important details in the use process of the account are easily omitted, so that the identification accuracy is low, and the security of the transaction function in the application program is affected.
Disclosure of Invention
The embodiment of the application provides an object identification method, an object identification device, a terminal, a storage medium and a program product.
In one aspect, an embodiment of the present application provides an object identification method, where the method includes:
Under the condition of obtaining authorization, obtaining account switching data, wherein the account switching data is data generated by changing an account interacted with a designated account, and the account corresponds to account information;
constructing a graph network structure based on the account switching data, wherein the graph network structure comprises nodes and edges between the nodes, the nodes correspond to the accounts, and the edges between the nodes are used for representing switching relations between the accounts;
extracting features of the graph network structure to obtain node feature representations corresponding to the nodes, wherein first node representations corresponding to the first account are extracted to obtain first node feature representations;
fusing the first node characteristic representation with a first information characteristic representation corresponding to the account information of the first account to obtain a first account characteristic representation;
and identifying the characteristic representation of the first account to obtain an identification result corresponding to the first account, wherein the identification result is used for representing the identification type of the first account.
In another aspect, an embodiment of the present application provides an object identifying apparatus, including:
the system comprises an acquisition module, a storage module and a management module, wherein the acquisition module is used for acquiring account switching data under the condition of acquiring authorization, the account switching data is data generated by changing an account interacted with a designated account, and the account corresponds to account information;
The construction module is used for constructing a graph network structure based on the account switching data, wherein the graph network structure comprises nodes and edges between the nodes, the nodes correspond to the accounts, and the edges between the nodes are used for representing switching relations between the accounts;
the extraction module is used for extracting the characteristics of the graph network structure to obtain node characteristic representations corresponding to the nodes, wherein a first node corresponding to a first account is extracted to obtain first node characteristic representations;
the fusion module is used for fusing the first node characteristic representation with a first information characteristic representation corresponding to the account information of the first account to obtain a first account characteristic representation;
the identification module is used for identifying the first account characteristic representation to obtain an identification result corresponding to the first account, wherein the identification result is used for representing the identification type of the first account.
In an optional embodiment, the obtaining module is further configured to obtain account switching data of the first account and the second account in a historical period of time under the condition of obtaining authorization, where the account switching data includes a switching record of an account interacting with the specified account between the first account and the second account;
The construction module is further configured to construct an edge between the first node and a second node corresponding to the second account based on account switching data of the first account and the second account in the historical time period.
In an alternative embodiment, the building block comprises:
the determining unit is used for determining a target account switching event generated when the first account and the second account carry out resource exchange events with the appointed account in the historical time period, wherein the resource exchange event refers to an event generated between the appointed account and the purpose of resource exchange;
and the construction unit is used for determining the weight of the edge between the first node and the second node based on the target account switching event and constructing the edge between the first node and the second node.
In an optional embodiment, the determining unit is further configured to determine an event interval duration between the first account and the second account and the designated account for performing the resource exchange event respectively; and determining the weight of the edge between the first node and the second node based on the event interval duration, wherein the weight and the event interval duration are in a negative correlation relationship.
In an optional embodiment, the determining unit is further configured to, for an account switching event for switching from the first account to the second account, determine, in response to the initiation time of the resource exchange event of the first account being within a preset duration with the initiation time of the resource exchange event of the first account as a start time, the resource exchange event corresponding to the second account is initiated, and determine the account switching event as a target account switching event.
In an optional embodiment, the building module is further configured to build a directed edge between the first node and the second node based on an account switching relationship between the first account and the second account in the historical period, where a direction of the directed edge points from a switching source account to a switching target account.
In an optional embodiment, the obtaining module is further configured to obtain a neighboring node feature representation corresponding to a neighboring node of the first node;
and the fusion module is further configured to fuse the first node feature representation, the neighboring node feature representation and the first information feature representation to obtain the first account feature representation.
In an alternative embodiment, the first information characteristic representation comprises at least one of the following characteristics:
Basic information characteristics corresponding to the basic information of the account are expressed, and the basic information of the account is information related to a use object of the first account;
the resource exchange information is represented by a resource exchange characteristic corresponding to the resource exchange information, and the resource exchange information is generated by the first account in a historical resource exchange scene;
social characteristics corresponding to social information are represented, wherein the social information is social related information between the first account and other accounts;
and the risk characteristics corresponding to the historical account risk information are represented, and the historical account risk information is used for representing the risk condition of the first account in the historical use process.
In an optional embodiment, the identification module is further configured to perform risk identification on the first account feature representation, so as to obtain a risk identification result corresponding to the first account, where the risk identification result is used to characterize a risk degree of the first account.
In another aspect, embodiments of the present application provide a computer device, where the computer device includes a processor and a memory, where at least one section of program is stored in the memory, and the at least one section of program is loaded and executed by the processor to implement the object recognition method according to the above aspect.
In another aspect, there is provided a computer readable storage medium having stored therein at least one program loaded and executed by a processor to implement the object recognition method as described in the above aspect.
In another aspect, embodiments of the present application provide a computer program product or computer program comprising computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions so that the computer device performs the object recognition method provided in the above aspect.
The technical scheme provided by the embodiment of the application at least comprises the following beneficial effects.
By acquiring the account switching data and constructing a graph network structure corresponding to the switching relation between accounts, hidden account risk conditions are acquired from the account switching data based on the graph network structure, and when the account is subjected to risk identification by combining the account switching data and the related characteristics of the account information, the risk reflected during account switching can be predicted, the risk represented by the account per se can also be predicted, the accuracy of risk prediction is improved, and the safety coefficient in the application program use process is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 illustrates an overall flow diagram provided by an exemplary embodiment of the present application;
FIG. 2 illustrates a schematic diagram of an implementation environment provided by an exemplary embodiment of the present application;
FIG. 3 illustrates a flow chart of an object recognition method provided by an exemplary embodiment of the present application;
FIG. 4 is a schematic diagram of a graph network structure corresponding to account switching data provided based on the embodiment shown in FIG. 3;
FIG. 5 illustrates a flowchart of an object recognition method provided by another exemplary embodiment of the present application;
FIG. 6 illustrates a flowchart of an object recognition method provided by yet another exemplary embodiment of the present application;
FIG. 7 illustrates a block diagram of an object recognition device provided in an exemplary embodiment of the present application;
FIG. 8 is a block diagram illustrating an object recognition apparatus according to another exemplary embodiment of the present application;
Fig. 9 shows a block diagram of a computer device according to an exemplary embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present application more apparent, the embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
Artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use the knowledge to obtain optimal results. In other words, artificial intelligence is an integrated technology of computer science that attempts to understand the essence of intelligence and to produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence, i.e. research on design principles and implementation methods of various intelligent machines, enables the machines to have functions of sensing, reasoning and decision. The artificial intelligence technology is a comprehensive subject, and relates to the technology with wide fields, namely the technology with a hardware level and the technology with a software level. Artificial intelligence infrastructure technologies generally include, for example, sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, pre-training model technologies, operation/interaction systems, mechatronics, and the like. The pre-training model is also called a large model and a basic model, and can be widely applied to all large-direction downstream tasks of artificial intelligence after fine adjustment. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.
Machine Learning (ML) is a multi-domain interdisciplinary, involving multiple disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, etc. It is specially studied how a computer simulates or implements learning behavior of a human to acquire new knowledge or skills, and reorganizes existing knowledge structures to continuously improve own performance. Machine learning is the core of artificial intelligence, a fundamental approach to letting computers have intelligence, which is applied throughout various areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, induction learning, teaching learning, and the like. The pre-training model is the latest development result of deep learning, and integrates the technology.
With the research and advancement of artificial intelligence technology, the research and application of artificial intelligence technology is developed in various fields, such as common smart home, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned, autopilot, unmanned, digital twin, virtual man, robot, artificial intelligence generation content (Artificial Intelligence Generated Content, AIGC), conversational interaction, smart medical treatment, smart customer service, game AI, etc., and it is believed that with the development of technology, the artificial intelligence technology will find application in more fields and play an increasingly important value.
The scheme provided by the embodiment of the application relates to artificial intelligence voice technology, natural language processing, machine learning and other technologies, and is specifically described by the following embodiment.
It should be noted that, before and during the process of collecting the relevant data of the user, the present application may display a prompt interface, a popup window or output a voice prompt message, where the prompt interface, popup window or voice prompt message is used to prompt the user to collect the relevant data currently, so that the present application only starts to execute the relevant step of obtaining the relevant data of the user after obtaining the confirmation operation of the user to the prompt interface or popup window, otherwise (i.e. when the confirmation operation of the user to the prompt interface or popup window is not obtained), the relevant step of obtaining the relevant data of the user is finished, i.e. the relevant data of the user is not obtained. In other words, all user data collected in the present application is collected with the user agreeing and authorized, and the collection, use and processing of relevant user data requires compliance with relevant legal regulations and standards.
The object recognition method provided by the embodiment of the application at least comprises the following application scenes:
1. Account risk identification
That is, by extracting account characteristic representation corresponding to the account, risk identification is carried out on the account, whether the account belongs to the risk account is judged, and the problem that the security of an application program is low in a resource exchange scene such as transfer, red package and the like caused by the account is avoided; when the account characteristic representation is extracted, the graph network structure generated by the account switching event is taken as a consideration factor, so that the accuracy of risk identification is improved.
2. Friend recommendation function
The candidate accounts which do not establish the association relationship are recommended to the accounts by extracting the account characteristics corresponding to the accounts, so that the tightness of establishing the association relationship between the accounts is improved. When the account characteristic representation is extracted, the graph network structure generated by the account switching event is taken as a consideration, so that the analysis accuracy of the hidden association relationship between accounts is improved.
It should be noted that the above application scenario is merely an illustrative example, and the embodiments of the present application are not limited thereto.
In the embodiment of the application, a method for identifying risk accounts based on an interactive account switching relationship is provided. Schematically, as shown in fig. 1, an overall flow diagram of account identification provided in an exemplary embodiment of the present application is shown. Under the condition of obtaining user authorization, account switching data 100 between accounts are obtained, wherein the account switching data 100 is used for representing switching events between the accounts in the account interaction process, and taking a payment scene as an example, the switching events comprise at least one of the following conditions: firstly, a payee switches and logs in from an account A to an account B through a switching operation on a login account of an application program on electronic equipment due to a failure in collection; second, the payer switches from paying account a to paying account B due to failure of the payee to collect the money. The account switching event is an interaction event of the same or similar account by the pointer, and the interaction object accounts are switched, for example: the account switching event is a switching of the account of the collection object or a switching of the account of the payment object by the pointer, and in some embodiments, the account switching event is a switching of the account of the collection object under the condition that the account of the payment object is unchanged; or, when the account number to be collected is unchanged, the account number to be paid is switched.
In some embodiments, taking the above-mentioned example that the payee switches the login accounts, an application on the electronic device logs in only one account at a time, so that the account B cannot be operated under the condition of logging in the account a, and needs to be switched from the account a to the account B; alternatively, the application Cheng Xushan on the electronic device logs in multiple accounts, such as: the account switching event is used for representing the event of switching from the account A to the account C, or the account switching event is used for representing the event of switching from using the logged account A to using the logged account B. The specific switching manner of the account switching event is not limited in this embodiment.
The graph network structure 110 is constructed on the basis that the account switching is the data 100, and the graph network structure 110 comprises nodes and edges, wherein the nodes correspond to the accounts, and the edges correspond to the account switching data between the accounts, namely the edges are used for representing the switching relation between the accounts. In some embodiments, the side weights correspond to the switching relationships between accounts, and in this exemplary embodiment, when the interactive account is switched from the first account to the second account within the preset duration of the resource transaction event, the weight between the first account and the second account is increased by one, where the resource transaction event is an event that designates that the account and the first account perform resource transaction.
After constructing the graph network structure 110, extracting features of the graph network structure 110 to obtain node feature representation 120 corresponding to the account, and fusing the node feature representation 120 with information feature representation 130 corresponding to the account to obtain account feature representation 140; the account feature representation 140 is identified to obtain an identification result corresponding to the account. Illustratively, risk identification is performed on the account feature representation 140 to obtain a risk identification result corresponding to the account.
The object identification method provided by the embodiment of the application can be executed by a server, or can be executed by the cooperation of the terminal and the server.
In this embodiment, a method for identifying an object by matching a terminal and a server is described as an example, please refer to fig. 2 below, which shows a schematic structural diagram of an implementation environment provided in an exemplary embodiment of the present application. As shown in fig. 2, the implementation environment mainly includes a terminal 210 and a server 220, where a communication connection is established between the terminal 210 and the server 220 through a communication network 230.
The terminal 210 is used for running an application program, and the server 220 is used for processing part of data of the terminal 210 when the application program is run, and feeding back to the terminal 210 for display.
In this embodiment, the terminal 210 runs the application program and logs in with the account, and when exchanging resources with other accounts through the resource exchange function in the application program, sends a resource exchange request to the server 220, so that the server 220 processes the resource exchange event. And the server 220 determines whether the resource exchange event can be performed based on the account risk situation of both sides of the exchange.
In this embodiment, when the resource exchange event cannot be performed, the terminal 210 may initiate an account switching event, switch from the currently logged account to another account, or switch the payer account from the currently collected Zhang Zhanghao to another account, thereby completing the resource exchange event. The server 220 identifies and records account switching events occurring between the accounts, and applies account switching data to the determination process of the risk account.
The server 220 constructs a graph network structure according to the account switching data corresponding to the identified and recorded account switching event, constructs a switching relationship between each account through the graph network structure, and analyzes whether the account belongs to the risk account based on the graph network structure and the information of the account.
In some embodiments, an application is installed and run in the terminal 210, and an account switching event is transmitted to the server 220 through the application. The application may be a stand-alone application or an applet that is hosted by the host application. Such as: the host application is a search engine program, a travel application, a life assistance application, an instant messaging application, a video program, a game program, etc., which is not limited in this embodiment.
The terminal may be a mobile phone, a tablet computer, an intelligent robot, a desktop computer, a portable notebook computer, an intelligent television, a vehicle-mounted terminal, an intelligent home device, or other terminal devices in various forms, which is not limited in this embodiment of the present application.
The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), basic cloud computing services such as big data and artificial intelligence platforms.
Cloud technology (Cloud technology) refers to a hosting technology that unifies serial resources such as hardware, software, networks and the like in a wide area network or a local area network to realize calculation, storage, processing and sharing of data. The cloud technology is based on the general names of network technology, information technology, integration technology, management platform technology, application technology and the like applied by the cloud computing business mode, can form a resource pool, and is flexible and convenient as required. Cloud computing technology will become an important support. Background services of technical networking systems require a large amount of computing, storage resources, such as video websites, picture-like websites, and more portals. Along with the high development and application of the internet industry, each article possibly has an own identification mark in the future, the identification mark needs to be transmitted to a background system for logic processing, data with different levels can be processed separately, and various industry data needs strong system rear shield support and can be realized only through cloud computing.
In some embodiments, the servers described above may also be implemented as nodes in a blockchain system.
In connection with the above description, the method for identifying an object provided in the present application is described, and the method may be executed by a server or a terminal, or may be executed by the server and the terminal together.
Step 301, under the condition of obtaining authorization, obtaining account switching data.
The account switching data refers to data generated by changing an account interacted with a designated account, and the account corresponds to account information.
In some embodiments, account authorization may first be acquired before account switch data is acquired. Optionally, an authorization request is sent to the account, and the account switching data and the purpose of using the account switching data are indicated in the authorization request, which is illustrative, the account switching data generated in the process of switching payment, switching login or other switching interaction are indicated in the authorization request, and the account switching data are used for analyzing the risk situation of the account. After the terminal for logging in the account displays the authorization request, the authorization can be acquired by acquiring the authorization to the account switching data of the account in response to receiving the confirmation operation of the user to the authorization request, and the authorization is applied to the account identification process. It should be noted that the above authorization manner of the account switching data is only an illustrative example, and the embodiment of the present application does not limit the manner of obtaining the authorization.
In some embodiments, the interaction with the designated account may be selected from the designated account side or may be switched from the interaction account side.
Taking the selection of the appointed account side as an example, the account switching data refers to the data generated by the change of the appointed account due to the switching of the appointed account for the interaction with the appointed account. Illustratively, account switching data refers to data generated by switching the collection account from a payment request initiated by a specified account, such as: the appointed account firstly initiates a payment request to the account A, and after the payment fails, initiates the payment request to the account B. In some embodiments, the server identifies payment events, and when it is identified that there is an association between a specified account and payment interactions of account a and account B, the switching event that switches account a to account B generates account switching data.
Taking the switching of the interactive account side as an example, the account switching data refers to the data generated by switching login between a logged account and an unregistered account and interacting with a designated account; or, the account switching data refers to data generated by switching among a plurality of logged accounts and interacting with a designated account, which is not limited in the embodiment of the present application. The account switching data is used for representing switching of the used main control account. In this embodiment, the account switching data generation includes at least one of the following cases:
Firstly, switching from a logged-in account to an unregistered account in an application program to interact with a designated account;
schematically, an account A is logged in an application program, and the account A is switched and logged in to the account B through an account switching operation; or, the application program is logged with the account A and the account B, and the account A is switched to the account C through the account switching operation, so that the application program is logged with the account C and the account B.
And secondly, switching the application among a plurality of logged accounts in the application program to interact with the appointed accounts.
Illustratively, an account A and an account B are logged in an application program, a function in the application program corresponding to the main control account A of the current user is used, and in response to receiving an account switching operation, the function in the application program is used by switching from the main control account A to the main control account B.
The account information is information related to the user, which is acquired after the user authorization. Schematically, when registering the account, the prompt message is displayed, the prompt message includes the authorized collected information in the application program and the corresponding purpose of the collected information, and after receiving the agreement operation of the user to the prompt message, the account information of the user is collected.
Step 302, a graph network structure is constructed based on account switching data, the graph network structure including nodes and edges between nodes.
Wherein the nodes correspond to the accounts, and edges between the nodes are used for representing switching relations between the accounts. The graph network structure is at least one of a directed graph, an undirected graph, a directed weighted graph and an undirected weighted graph, and when the graph network structure is the directed weighted graph, the direction of the edge indicates the direction of the account switching, and the weight indicates the risk degree of the account switching; when the graph network structure is realized as an undirected weighted graph, the weights of the edges indicate the risk degrees of the two-way switching of the accounts, namely, the risk degrees respectively corresponding to the two-way switching are integrated and embodied on the undirected edges.
In this embodiment, a description will be given by taking a graph network structure implemented as a directed weighted graph as an example. Nodes in the graph network structure are respectively corresponding to the accounts, wherein a one-to-one correspondence exists between the accounts and the nodes, and when a switching relationship exists between the two accounts, edges are connected between the nodes respectively corresponding to the two accounts.
In some embodiments, under the condition of obtaining authorization, account switching data of the first account and the second account in a historical time period is obtained, the account switching data includes a switching record of the account interacted with the appointed account between the first account and the second account, and an edge between a first node corresponding to the first account and a second node corresponding to the second account is constructed based on the account switching data of the first account and the second account in the historical time period.
In some embodiments, a directed edge between the first node and the second node is constructed based on account switching relationships of the first account and the second account over a historical period of time, the direction of the directed edge pointing from the switching source account to the switching target account.
Schematically, as shown in fig. 4, which shows a schematic diagram of a graph network structure provided in an exemplary embodiment of the present application, the graph network structure 400 includes a first node 410 and a second node 420, where the first node 410 corresponds to a first account, the second node 420 corresponds to a second account, and a directed edge 430 between the first node 410 and the second node 420 characterizes that a switching situation exists when the first account and the second account interact with a designated account, that is, the interaction between the first account and the designated account is switched from the interaction between the first account and the designated account to the interaction between the second account and the designated account.
Wherein the historical time period is a preset time period range, and in some embodiments, the historical time period is a time period within a preset duration before the current time; alternatively, the historical time period is a time period determined based on the data storage period.
And 303, extracting features of the graph network structure to obtain node feature representations corresponding to the nodes.
And extracting the first node corresponding to the first account to obtain a first node characteristic representation.
Optionally, feature extraction is performed on the graph network structure through a feature extraction model, wherein a first-order relation and/or a second-order relation between nodes corresponding to each account are mapped into a multidimensional space through the feature extraction model, so that node feature representation corresponding to the nodes is extracted.
Wherein the feature extraction model comprises at least one of the following models: deep walk model deep, node embedding algorithm Node2vec, large-scale network coding model LINE, heterogeneous network characterization model metaath 2vec. Different feature extraction models each have a emphasis when preserving semantic information in the graph network structure:
(1) Deep walk captures relative position information between nodes by modeling the random walk sequence in the graph.
(2) The Node2vec introduces a parameterized random walk strategy on the basis of deep walk to balance the local and global structures between nodes.
(3) The LINE model focuses on preserving both first and second order neighborhood information, thereby better capturing higher order relationships in the graph.
(4) When various types of nodes or edges are introduced into the graph, a Metapath2vec model can be further considered, the model learns node representation in a heterogeneous network by utilizing a meta-path concept, and semantic information in the edges of different types can be effectively described by different meta-paths.
Through the feature extraction model, the nodes in the network structure of the account switching diagram are mapped into a low-dimensional space, and meaningful feature representation is provided for subsequent clustering, classification or other analysis tasks. The method is helpful for better learning the relevance among accounts and discovering potential risks and rules.
Optionally, the feature extraction model is a model for feature extraction of a graph network structure, which is obtained through pre-training.
Step 304, fusing the first node characteristic representation and the first information characteristic representation corresponding to the account information of the first account to obtain the first account characteristic representation.
Taking risk analysis of the first account as an example, the first node feature representation is used for representing the risk performance of the first account in an account switching event, and the account information of the first account represents the risk performance of the first account in the use process, so that the first account feature representation is obtained by combining the first node feature representation and the first information feature representation corresponding to the account information, and the overall risk performance of the first account is represented by the first account feature representation.
The fusion mode of the first node characteristic representation and the first information characteristic representation comprises at least one of the following modes:
1. Splicing the first node characteristic representation and the first information characteristic representation according to a preset splicing sequence to obtain a first account characteristic representation;
2. and carrying out operation fusion on the first node characteristic representation and the first information characteristic representation according to a preset fusion algorithm, for example: and carrying out weighted fusion on the elements in the first node characteristic representation and the elements in the first information characteristic representation, wherein the first node characteristic representation and the first information characteristic representation are embodied in the form of a vector matrix.
It should be noted that the above-mentioned merging manner of the first node feature representation and the first information feature representation is merely an illustrative example, which is not limited in this embodiment of the present application.
Wherein the account information of the first account includes at least one of the following information:
(1) Basic account information: the information related to the use object of the first account comprises basic label information such as registration time, identity authentication condition, region and the like of the account, and the characteristics are used for describing the basic condition of the account; the account basic information corresponds to basic information characteristic representation;
(2) The resource exchanges information: the method is information generated by the first account in a historical resource exchange scene, and comprises statistical characteristics of resource exchange frequency, resource exchange time distribution, resource exchange amount distribution, resource exchange objects, resource exchange scene distribution and the like in a collection or payment scene, so that an abnormal transaction mode of a risk account is effectively mined; the resource exchange information corresponds to a resource exchange characteristic representation;
(3) Social information: the social related information between the first account and other accounts comprises information such as the number of friends, interaction frequency, release content and the like of the first account on a social platform, and is helpful for describing social activity and potential risk states of the accounts; the social information corresponds to a social feature representation;
(4) Historical account risk information: the method is used for representing the risk condition of the first account in the history use process, mainly comprises the information such as the past risk state and the severity of the first account, and is beneficial to judging whether the current account has long-term risk or not, and the risk information of the history account corresponds to the risk characteristic representation.
It should be noted that the account information of the first account is merely an exemplary example, and the type and the specific content of the account information are not limited in the embodiments of the present application.
And 305, performing object recognition on the first account feature representation to obtain a recognition result corresponding to the first account.
The identification result is used for indicating the identification type of the first account.
In some embodiments, the identification result is used to represent a risk identification result of the first account, that is, risk identification is performed on the first account feature representation, so as to obtain a risk identification result corresponding to the first account, where the risk identification result is used to represent a risk degree of the first account.
In some embodiments, the identification result is used to characterize the identification result of the interest point of the first account, that is, the interest point of the first account is identified in a predictive manner in combination with the interest expression between the switching accounts, which is schematically indicated, and whether the first account is identified in a predictive manner with respect to the interest keyword a in combination with the interest expression between the switching accounts.
In summary, in the object recognition method provided in this embodiment, by acquiring the account switching data, a graph network structure corresponding to the switching relationship between accounts is constructed, so that an implicit account risk condition is acquired from the account switching data based on the graph network structure, and when risk recognition is performed on the account by combining the account switching data and the characteristics related to the account information, the risk reflected during account switching can be predicted, the risk reflected during account switching can also be predicted, the accuracy of risk prediction is improved, and the safety factor in the application using process is improved.
According to the method provided by the embodiment, the edges between the first node and the second node in the graph network structure are constructed aiming at the account switching events of the first account and the second account in the historical time period, the account switching records between the first account and the second account are intuitively represented through the graph network structure, and the data processing efficiency is improved.
In an alternative embodiment, the weights of edges in the graph network structure are determined based on resource exchange events and account switching events. Fig. 5 is a flowchart of an object recognition method according to another exemplary embodiment of the present application, and the method is applied to a server, for example, as shown in fig. 5, where step 302 may be implemented as the following steps 3021 to 3022.
Step 3021, determining a target account switching event generated when the first account and the second account perform a resource exchange event with the designated account in the historical time period.
The resource exchange event refers to an event generated between the purpose of resource exchange and a designated account. The first account is a fund receiving party, namely a payee, in the resource exchange event, or the first account is a fund paying party, namely a payor, in the resource exchange event. In this embodiment, the first account is taken as an example of a fund receiving party, and the resource exchange scenario includes, but is not limited to, a face-to-face two-dimensional code scanning payment scenario, a personal transfer scenario, a red packet scenario, a commercial payment scenario, and the like.
The association relation between the resource exchange event and the account switching event comprises at least one of the following relations:
Firstly, a switching source account corresponding to an account switching event and a switching target account both have resource switching events within a preset time window range, and the resource switching conditions of the resource switching events are the same or similar;
schematically, after the first account and the designated account perform the resource exchange event, the first account is switched to the second account and the designated account perform the resource exchange event within the preset time window, and the resource exchange conditions of the resource exchange events corresponding to the two accounts are the same or similar, for example: the resource exchange event between the first account and the appointed account is that the appointed account transfers 10 yuan to the first account, and the appointed account transfers failure; the resource exchange event between the second account and the appointed account is that the appointed account transfers 9 yuan to the second account, the payment objects are the same, and the payment amount is similar, so the account switching event between the first account and the second account is generated. The similarity of the resource exchange conditions is determined through information such as the resource exchange objects, the resource exchange quantity and the like. Such as: the resource exchange object and the resource exchange number are converted into feature vectors, and the similarity of the resource exchange conditions is determined based on the vector distances between the feature vectors.
Secondly, when the account switching event is realized as an account switching login event of a resource receiver, the account switching event exists in a preset time window range where the resource switching event starts or ends;
illustratively, within 5 minutes after the resource exchange event fails and ends, there is an account switching event, and the participating accounts of the resource exchange event are the same as the account that initiated the account switching event. Such as: the account A is switched from the login account A to the login account B within 5 minutes after the collection failure; or, the appointed account number is switched from paying the account number A to paying the account number B within 5 minutes after the payment failure.
Third, when the account switching event is implemented as an account switching login event of the resource receiving party, an account switching event exists in a preset time window range where the resource switching event starts or ends, another resource switching event exists in a preset time window range after the account switching event, and the resource switching condition of the resource switching event and the resource switching condition of the other resource switching event are the same or similar.
Illustratively, an account switching event exists within 5 minutes after the end of the resource switching event, and the account of the resource switching event is the same as the account from which the account switching event originated, and the resource switching event is also initiated within 5 minutes after the account switching event. Such as: the account A is switched from the account A to the account B within 5 minutes after 10 yuan of collection fails, and the account B initiates a 10 yuan of collection resource exchange event within 5 minutes after the account B is logged in.
In some embodiments, for an account switching event switched from a first account to a second account, in response to a resource exchange event corresponding to the second account being initiated within a preset duration with an initiation time of the resource exchange event of the first account as a start time, the account switching event is determined to be a target account switching event. In the above example, the starting time of the resource exchange event is taken as the starting time, and the starting time may also be implemented as the ending time of the resource exchange event, for example: the failure termination time of the resource exchange event.
It should be noted that the above-mentioned association relationship is only an illustrative example, and the present embodiment is not limited thereto.
In some embodiments, the resource exchange event corresponding to the first account is an event of resource exchange failure, and in some embodiments, the resource exchange event is an event of resource exchange failure caused by the resource exchange being limited by the risk intervention policy.
When the account switching event accords with the association relation, the account switching event is determined to be a target account switching event.
And 3022, determining the weight of the edge between the first node and the second node based on the target account switching event, and constructing the edge between the first node and the second node to obtain the graph network structure.
In some embodiments, the weight of the edge between the first node and the second node is determined based on the number of unidirectional target account switching events between the first account and the second account, wherein the number of unidirectional target account switching events is proportional to the weight of the corresponding directed edge. Schematically, when a target account switching event occurs that the first account is switched to the second account, the edge of the first node pointing to the second node increases the weight 1.
Or in other embodiments, determining the time interval duration of the resource exchange event carried out by the first account and the second account respectively, and determining the weight of the edge between the first node and the second node based on the time interval duration, wherein the weight and the time interval duration are in a negative correlation. In some embodiments, a single increasing weight of the edge between the first node and the second node is determined based on the event interval duration. Schematically, when the time length of the event interval of the resource exchange event carried out by the first account and the second account is 1 minute, the risk association relationship between the two accounts is stronger, so that 1 is added to the weight of the edge between the first node and the second node; when the time length of the event interval of the resource exchange event carried out by the first account and the second account is 4 minutes, the risk association relationship between the two accounts is weak, so that the weight of the directed edge of the first node pointing to the second node is added with 0.25.
Illustratively, for the first account, traversing transaction behaviors within a specific time range, when a policy of intervention of a specific malicious type appears, at a certain time, for example: if the behavior of switching the collection number occurs within 5 minutes, the connection edge exists between the two collection number nodes, and the weight of the edge is added by one. And (3) carrying out the steps on all the collection numbers to obtain an account switching diagram G= (V, E) of the risk collection number under a specific malicious type. Wherein V represents the node set of the collection account, each edge E E represents a pair of ordered collection number node pairs (u, V), w (u, V) is the weight of the corresponding edge, and represents the occurrence frequency of the collection number node pairs. In addition, since the graph is a directed weighted graph, there is
Optionally, the graph network structure includes a homogeneous graph network structure or a heterogeneous graph network structure, the homogeneous graph network structure has only one type of point type and edge type, and the heterogeneous graph network structure has multiple types of point type and edge type. When the graph network structure is realized as a isomorphic graph network structure, account switching events associated with different resource switching events are corresponding to weight setting in the same isomorphic graph network structure; when the graph network structure is implemented as a heterogeneous graph network structure, account switching events associated with different resource exchange events are embodied as different edges in the graph network structure, such as: the account switching event associated with the red package corresponds to a first type edge in the graph network structure, the account switching event associated with the transfer corresponds to a second type edge in the graph network structure, and feature extraction of the graph network structure is performed based on the edges of different types.
In summary, in the object recognition method provided in this embodiment, by acquiring the account switching data, a graph network structure corresponding to the switching relationship between accounts is constructed, so that an implicit account risk condition is acquired from the account switching data based on the graph network structure, and when risk recognition is performed on the account by combining the account switching data and the characteristics related to the account information, the risk reflected during account switching can be predicted, the risk reflected during account switching can also be predicted, the accuracy of risk prediction is improved, and the safety factor in the application using process is improved.
According to the method provided by the embodiment of the application, the weight of the edge between each node in the graph network structure is determined through the association relation between the resource exchange event and the account switching event, so that the risk degree of switching between each account is reflected through the weight, the accuracy of graph network structure expression is improved, and the accuracy of risk prediction of the account is further improved.
According to the method provided by the embodiment of the application, the weight of the edge between the nodes is determined based on the event interval duration between the resource exchange event and the account switching event, and the edge is assigned with the weight which is in negative correlation with the event interval duration, so that the accuracy of the degree of risk reflected by the weight is improved, and the accuracy of the structural expression of the graph network structure is improved.
According to the method provided by the embodiment of the application, the account switching relation among all accounts is embodied through the directed weighted graph, the switching source account and the switching target account of account switching are improved, the data expression capability of the graph network structure is improved, and the feature extraction accuracy of the nodes is improved.
In an alternative embodiment, the neighbor node feature representation may also be required as a consideration in generating the account feature representation. Fig. 6 is a flowchart of an object recognition method according to another exemplary embodiment of the present application, and the method is applied to a server, for example, and as shown in fig. 6, the method includes the following steps.
In step 601, under the condition of obtaining authorization, account switching data is obtained.
In some embodiments, the interaction with the designated account may be selected from the designated account side or may be switched from the interaction account side.
Taking the selection of the appointed account side as an example, the account switching data refers to the data generated by the change of the appointed account due to the switching of the appointed account for the interaction with the appointed account. Illustratively, account switching data refers to data generated by switching the collection account from a payment request initiated by a specified account, such as: the appointed account firstly initiates a payment request to the account A, and after the payment fails, initiates the payment request to the account B. In some embodiments, the server identifies payment events, and when it is identified that there is an association between a specified account and payment interactions of account a and account B, the switching event that switches account a to account B generates account switching data.
Step 602, based on account switching data of the first account and the second account in the historical time period, constructing an edge between the first node and the second node corresponding to the second account, and obtaining a graph network structure.
In some embodiments, a directed edge between the first node and the second node is constructed based on account switching relationships of the first account and the second account over a historical period of time, the direction of the directed edge pointing from the switching source account to the switching target account.
And 603, extracting features of the graph network structure to obtain node feature representation corresponding to the nodes.
And extracting the first node corresponding to the first account to obtain a first node characteristic representation.
Optionally, feature extraction is performed on the graph network structure through a feature extraction model, wherein a first-order relation and/or a second-order relation between nodes corresponding to each account are mapped into a multidimensional space through the feature extraction model, so that node feature representation corresponding to the nodes is extracted.
The first-order relationship is a similarity relationship between nodes in the graph network structure, and the second-order relationship is a similarity relationship between neighboring nodes in the graph network structure.
Step 604, obtaining a neighboring node characteristic representation corresponding to a neighboring node of the first node.
The adjacent nodes are nodes with adjacent relation with the first node in the graph network structure, the adjacent nodes are nodes with a designated number of connecting edges with the first node, and the adjacent nodes are nodes with 1 or 2 edges with the first node. It should be noted that the specified number is a preset number, which is not limited in the embodiment of the present application.
The neighboring node feature representation is a feature representation corresponding to the neighboring node in the node feature representations extracted in the step 603.
Step 605, fusing the first node feature representation, the neighboring node feature representation and the first information feature representation corresponding to the account information of the first account to obtain a first account feature representation.
Taking risk analysis for the first account as an example, the first node feature representation is used for representing the risk performance of the first account in an account switching event, the account information of the first account represents the risk performance of the first account in the use process, the adjacent node feature representation is used for representing the risk performance of the adjacent node in the account switching event, and the adjacent node is an account having a switching relation with the first account, and the account switching risk of the first account is represented from the side, so the first account feature representation is obtained by combining the first node feature representation, the adjacent node feature representation and the first information feature representation corresponding to the account information, and the integral risk performance of the first account is represented by the first account feature representation.
Step 606, the first account feature representation is identified, and an identification result corresponding to the first account is obtained.
The identification result is used for indicating the identification type of the first account.
In some embodiments, the identification result is used to represent a risk identification result of the first account, that is, risk identification is performed on the first account feature representation, so as to obtain a risk identification result corresponding to the first account, where the risk identification result is used to represent a risk degree of the first account.
Optionally, the recognition model obtained by training in advance performs recognition prediction on the first account feature representation corresponding to the first account to obtain a recognition result, where the recognition model includes a logistic regression model, a support vector machine model, a decision tree model, a random forest model or a deep neural network model, and the embodiment of the application is not limited to this.
In summary, in the object recognition method provided in this embodiment, by acquiring the account switching data, a graph network structure corresponding to the switching relationship between accounts is constructed, so that an implicit account risk condition is acquired from the account switching data based on the graph network structure, and when risk recognition is performed on the account by combining the account switching data and the characteristics related to the account information, the risk reflected during account switching can be predicted, the risk reflected during account switching can also be predicted, the accuracy of risk prediction is improved, and the safety factor in the application using process is improved.
According to the method provided by the embodiment of the invention, the first account characteristic representation is obtained by fusion based on the first account characteristic representation, the adjacent node characteristic representation of the adjacent node and the first information characteristic representation of the first account, so that risk prediction is carried out on the first account from three angles of a switching event of the first account, a switching event of the adjacent node and information of the first account, and the accuracy rate of risk prediction is improved.
Fig. 7 is a block diagram of an object recognition apparatus according to an exemplary embodiment of the present application, and as shown in fig. 7, the apparatus includes:
the obtaining module 710 is configured to obtain account switching data under the condition of obtaining authorization, where the account switching data is data generated by changing an account interacting with a specified account, and the account corresponds to account information;
a construction module 720, configured to construct a graph network structure based on the account switching data, where the graph network structure includes nodes and edges between nodes, the nodes correspond to the accounts, and the edges between the nodes are used to characterize a switching relationship between the accounts;
an extracting module 730, configured to perform feature extraction on the graph network structure to obtain a node feature representation corresponding to the node, where a first node corresponding to the first account is extracted to obtain a first node feature representation;
The fusion module 740 is configured to fuse the first node feature representation with a first information feature representation corresponding to account information of the first account, to obtain a first account feature representation;
and the identification module 750 is used for identifying the first account feature representation to obtain an identification result corresponding to the first account, wherein the identification result is used for representing the identification type of the first account.
In an optional embodiment, the obtaining module 710 is further configured to obtain account switching data of the first account and the second account in a historical period of time, where the account switching data includes a switching record of an account interacting with the specified account between the first account and the second account, where the account switching data is used for obtaining authorization;
the construction module 720 is further configured to construct an edge between the first node and a second node corresponding to the second account based on account switching data of the first account and the second account in the historical time period.
In an alternative embodiment, as shown in fig. 8, the constructing module 720 includes:
a determining unit 721, configured to determine a target account switching event generated when the first account and the second account perform a resource exchange event with the specified account in the history period, where the resource exchange event refers to an event generated between the specified account and the first account for the purpose of resource exchange;
A construction unit 722, configured to determine a weight of an edge between the first node and the second node based on the target account switching event, and construct the edge between the first node and the second node.
In an optional embodiment, the determining unit 721 is further configured to determine an event interval duration between the first account and the second account and the designated account for performing the resource exchange event respectively; and determining the weight of the edge between the first node and the second node based on the event interval duration, wherein the weight and the event interval duration are in a negative correlation relationship.
In an optional embodiment, the determining unit 721 is further configured to, for an account switching event for switching from the first account to the second account, determine, in response to the initiation time of the resource exchange event of the first account being within a preset duration with the initiation time of the resource exchange event of the second account as a start time, the resource exchange event corresponding to the second account to be initiated, as a target account switching event.
In an optional embodiment, the building module 720 is further configured to build a directed edge between the first node and the second node based on an account switching relationship between the first account and the second account in the historical period, where a direction of the directed edge points from a switching source account to a switching target account.
In an optional embodiment, the obtaining module 710 is further configured to obtain a neighboring node feature representation corresponding to a neighboring node of the first node;
the fusion module 740 is further configured to fuse the first node feature representation, the neighboring node feature representation, and the first information feature representation to obtain the first account feature representation.
In an alternative embodiment, the first information characteristic representation comprises at least one of the following characteristics:
basic information characteristics corresponding to the basic information of the account are expressed, and the basic information of the account is information related to a use object of the first account;
the resource exchange information is represented by a resource exchange characteristic corresponding to the resource exchange information, and the resource exchange information is generated by the first account in a historical resource exchange scene;
social characteristics corresponding to social information are represented, wherein the social information is social related information between the first account and other accounts;
and the risk characteristics corresponding to the historical account risk information are represented, and the historical account risk information is used for representing the risk condition of the first account in the historical use process.
In an optional embodiment, the identification module 750 is further configured to perform risk identification on the first account feature representation, so as to obtain a risk identification result corresponding to the first account, where the risk identification result is used to characterize a risk degree of the first account.
In summary, the object recognition device provided in this embodiment constructs a graph network structure corresponding to a switching relationship between accounts by acquiring account switching data, so that an implicit account risk condition is acquired from the account switching data based on the graph network structure, and when risk recognition is performed on the account by combining the account switching data and the characteristics related to account information, the risk reflected during account switching can be predicted, the risk reflected during account switching can also be predicted, the accuracy of risk prediction is improved, and the safety factor in the application using process is improved.
It should be noted that: the apparatus provided in the above embodiment is only exemplified by the division of the above functional modules, and in practical application, the above functional allocation may be performed by different functional modules according to needs, that is, the internal structure of the apparatus is divided into different functional modules, so as to perform all or part of the functions described above. In addition, the apparatus and the method embodiments provided in the foregoing embodiments belong to the same concept, and detailed implementation processes of the method embodiments are described in the method embodiments, which are not repeated herein.
Referring to FIG. 9, a block diagram of a computer device 900 provided in an exemplary embodiment of the present application is shown. The computer device 900 may be a portable mobile computer device such as: smart phones, tablet computers, dynamic video expert compression standard audio layer 3 (Moving Picture Experts Group Audio Layer III, MP 3) players, dynamic video expert compression standard audio layer 4 (Moving Picture Experts Group Audio Layer IV, MP 4) players. Computer device 900 may also be referred to by other names of user devices, portable computer devices, etc.
In general, the computer device 900 includes: a processor 901 and a memory 902.
Processor 901 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and the like. The processor 901 may be implemented in hardware in at least one of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), programmable logic array (Programmable Logic Array, PLA). Processor 901 may also include a main processor, which is a processor for processing data in an awake state, also referred to as a central processor (Central Processing Unit, CPU), and a coprocessor; a coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 901 may integrate with an image processor (Graphics Processing Unit, GPU) for rendering and rendering of content required to be displayed by the display screen. In some embodiments, the processor 901 may also include an artificial intelligence (Artificial Intelligence, AI) processor for processing computing operations related to machine learning.
The memory 902 may include one or more computer-readable storage media, which may be tangible and non-transitory. The memory 902 may also include high-speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In some embodiments, a non-transitory computer readable storage medium in memory 902 is used to store at least one instruction for execution by processor 901 to implement the object recognition methods provided by embodiments of the present application.
In some embodiments, the computer device 900 may also optionally include: a peripheral interface 903, and at least one peripheral.
The peripheral interface 903 may be used to connect at least one Input/Output (I/O) related peripheral to the processor 901 and the memory 902.
In some embodiments, the processor 901, memory 902, and peripheral interface 903 are integrated on the same chip or circuit board; in some other embodiments, either or both of the processor 901, the memory 902, and the peripheral interface 903 may be implemented on separate chips or circuit boards, which is not limited in this embodiment.
Those skilled in the art will appreciate that the architecture shown in fig. 9 is not limiting of the computer device 900, and may include more or fewer components than shown, or may combine certain components, or employ a different arrangement of components.
Embodiments of the present application also provide a computer readable storage medium storing at least one program loaded and executed by a processor to implement the object recognition method described in the above embodiments.
According to one aspect of the present application, a computer program product is provided that includes computer instructions stored in a computer-readable storage medium. The processor of the terminal reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions so that the terminal performs the object recognition method provided in various alternative implementations of the above aspect.
Those skilled in the art will appreciate that in one or more of the examples described above, the functions described in the embodiments of the present application may be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, these functions may be stored on or transmitted over as one or more instructions or code on a computer-readable storage medium. Computer-readable storage media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.
The foregoing description of the preferred embodiments is merely exemplary in nature and is in no way intended to limit the invention, since it is intended that all modifications, equivalents, improvements, etc. that fall within the spirit and scope of the invention.
Claims (13)
1. An object recognition method, the method comprising:
under the condition of obtaining authorization, obtaining account switching data, wherein the account switching data is data generated by changing an account interacted with a designated account, and the account corresponds to account information;
constructing a graph network structure based on the account switching data, wherein the graph network structure comprises nodes and edges between the nodes, the nodes correspond to the accounts, and the edges between the nodes are used for representing switching relations between the accounts;
extracting features of the graph network structure to obtain node feature representations corresponding to the nodes, wherein first node representations corresponding to the first account are extracted to obtain first node feature representations;
fusing the first node characteristic representation with a first information characteristic representation corresponding to the account information of the first account to obtain a first account characteristic representation;
and carrying out object recognition on the first account characteristic representation to obtain a recognition result corresponding to the first account, wherein the recognition result is used for representing the recognition type of the first account.
2. The method according to claim 1, wherein, in the case of obtaining authorization, obtaining account switching data includes:
under the condition of obtaining authorization, obtaining account switching data of the first account and the second account in a historical time period, wherein the account switching data comprises switching records of accounts interacted with the appointed account between the first account and the second account;
the constructing a graph network structure based on the account switching data comprises the following steps:
and constructing an edge between the first node and a second node corresponding to the second account based on the account switching data of the first account and the second account in the historical time period.
3. The method according to claim 2, wherein the constructing an edge between the first node and a second node corresponding to the second account based on account switching data of the first account and the second account in the history period includes:
determining a target account switching event generated when the first account and the second account carry out resource exchange events with the appointed account in the historical time period, wherein the resource exchange event refers to an event generated between the appointed account aiming at resource exchange;
And determining the weight of the edge between the first node and the second node based on the target account switching event, and constructing the edge between the first node and the second node.
4. The method of claim 3, wherein the determining weights for edges between the first node and the second node based on the target account switching event comprises:
determining the time interval between the first account and the second account and the resource exchange event carried out by the appointed account respectively;
and determining the weight of the edge between the first node and the second node based on the event interval duration, wherein the weight and the event interval duration are in a negative correlation relationship.
5. The method of claim 3, wherein said determining a target account switch event generated when said first account and said second account are in a resource exchange event with said designated account over said historical period of time comprises:
for an account switching event from the first account to the second account, responding to the fact that the resource switching event corresponding to the second account is initiated within a preset duration with the initiation time of the resource switching event of the first account as the initiation time, and determining the account switching event as a target account switching event.
6. The method of claim 2, wherein constructing an edge between the first node and a second node corresponding to the second account based on account switching data of the first account and the second account over the historical period of time comprises:
and constructing a directed edge between the first node and the second node based on the account switching relation of the first account and the second account in the historical time period, wherein the direction of the directed edge points to a switching target account from a switching source account.
7. The method according to any one of claims 1 to 6, wherein the fusing the first node feature representation with a first information feature representation corresponding to account information of the first account to obtain a first account feature representation includes:
acquiring an adjacent node characteristic representation corresponding to an adjacent node of the first node;
and fusing the first node characteristic representation, the adjacent node characteristic representation and the first information characteristic representation to obtain the first account characteristic representation.
8. The method according to any one of claims 1 to 6, wherein the first information characteristic representation comprises at least one of the following characteristics:
Basic information characteristics corresponding to the basic information of the account are expressed, and the basic information of the account is information related to a use object of the first account;
the resource exchange information is represented by a resource exchange characteristic corresponding to the resource exchange information, and the resource exchange information is generated by the first account in a historical resource exchange scene;
social characteristics corresponding to social information are represented, wherein the social information is social related information between the first account and other accounts;
and the risk characteristics corresponding to the historical account risk information are represented, and the historical account risk information is used for representing the risk condition of the first account in the historical use process.
9. The method according to any one of claims 1 to 6, wherein the performing object recognition on the first account feature representation to obtain a recognition result corresponding to the first account includes:
and performing risk identification on the first account characteristic representation to obtain a risk identification result corresponding to the first account, wherein the risk identification result is used for representing the risk degree of the first account.
10. An object recognition apparatus, the apparatus comprising:
the system comprises an acquisition module, a storage module and a management module, wherein the acquisition module is used for acquiring account switching data under the condition of acquiring authorization, the account switching data is data generated by changing an account interacted with a designated account, and the account corresponds to account information;
The construction module is used for constructing a graph network structure based on the account switching data, wherein the graph network structure comprises nodes and edges between the nodes, the nodes correspond to the accounts, and the edges between the nodes are used for representing switching relations between the accounts;
the extraction module is used for extracting the characteristics of the graph network structure to obtain node characteristic representations corresponding to the nodes, wherein a first node corresponding to a first account is extracted to obtain first node characteristic representations;
the fusion module is used for fusing the first node characteristic representation with a first information characteristic representation corresponding to the account information of the first account to obtain a first account characteristic representation;
the identification module is used for identifying the first account characteristic representation to obtain an identification result corresponding to the first account, wherein the identification result is used for representing the identification type of the first account.
11. A computer device comprising a processor and a memory, wherein the memory has stored therein at least one program that is loaded and executed by the processor to implement the object recognition method of any one of claims 1 to 9.
12. A computer-readable storage medium, characterized in that at least one program is stored in the storage medium, the at least one program being loaded and executed by a processor to implement the object recognition method according to any one of claims 1 to 9.
13. A computer program product comprising a computer program or instructions which, when executed by a processor, implements the object recognition method according to any one of claims 1 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311316145.8A CN117332298A (en) | 2023-10-10 | 2023-10-10 | Object recognition method, device, terminal, storage medium and program product |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311316145.8A CN117332298A (en) | 2023-10-10 | 2023-10-10 | Object recognition method, device, terminal, storage medium and program product |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117332298A true CN117332298A (en) | 2024-01-02 |
Family
ID=89294937
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311316145.8A Pending CN117332298A (en) | 2023-10-10 | 2023-10-10 | Object recognition method, device, terminal, storage medium and program product |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117332298A (en) |
-
2023
- 2023-10-10 CN CN202311316145.8A patent/CN117332298A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Chen et al. | An edge traffic flow detection scheme based on deep learning in an intelligent transportation system | |
CN111371767B (en) | Malicious account identification method, malicious account identification device, medium and electronic device | |
CN114282681A (en) | Multitask processing and model training method, device, medium and equipment | |
CN112231570B (en) | Recommendation system support attack detection method, device, equipment and storage medium | |
EP4409466A1 (en) | Techniques for input classification and responses using generative neural networks | |
CN112819024B (en) | Model processing method, user data processing method and device and computer equipment | |
CN114692007B (en) | Method, device, equipment and storage medium for determining representation information | |
CN114219971A (en) | Data processing method, data processing equipment and computer readable storage medium | |
CN116823428A (en) | Anti-fraud detection method, device, equipment and storage medium | |
CN112308093B (en) | Air quality perception method based on image recognition, model training method and system | |
CN111966920B (en) | Method, device and equipment for predicting stable condition of public opinion propagation | |
CN110889467A (en) | Company name matching method and device, terminal equipment and storage medium | |
CN116048944A (en) | Load trend prediction method and system for big data platform | |
CN116975706A (en) | Data processing method, apparatus, device, readable storage medium, and program product | |
CN117332298A (en) | Object recognition method, device, terminal, storage medium and program product | |
CN113935738B (en) | Transaction data processing method, device, storage medium and equipment | |
CN113010772A (en) | Data processing method, related equipment and computer readable storage medium | |
Aynehband et al. | Accuracy and availability modeling of social networks for Internet of Things event detection applications | |
CN115578100A (en) | Payment verification mode identification method and device, electronic equipment and storage medium | |
CN117171562A (en) | Training method and device of intent prediction model, electronic equipment and storage medium | |
CN116976902A (en) | Data processing method, device, equipment and readable storage medium | |
CN116976901A (en) | Object recognition method, device and computer readable storage medium | |
Xu et al. | Generative Model for Models: Rapid DNN Customization for Diverse Tasks and Resource Constraints | |
Yang et al. | Visual privacy behaviour recognition for social robots based on an improved generative adversarial network | |
CN116127033A (en) | Network illegal behavior user identification method and corresponding device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication |