CN116663556A - Crowd portrayal construction method, device, equipment and medium based on graph neural network - Google Patents
Crowd portrayal construction method, device, equipment and medium based on graph neural network Download PDFInfo
- Publication number
- CN116663556A CN116663556A CN202310694316.4A CN202310694316A CN116663556A CN 116663556 A CN116663556 A CN 116663556A CN 202310694316 A CN202310694316 A CN 202310694316A CN 116663556 A CN116663556 A CN 116663556A
- Authority
- CN
- China
- Prior art keywords
- user
- node
- features
- feature
- aggregation
- 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
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 42
- 238000010276 construction Methods 0.000 title claims abstract description 35
- 230000002776 aggregation Effects 0.000 claims abstract description 61
- 238000004220 aggregation Methods 0.000 claims abstract description 61
- 238000012937 correction Methods 0.000 claims abstract description 47
- 238000004364 calculation method Methods 0.000 claims abstract description 46
- 238000010586 diagram Methods 0.000 claims abstract description 14
- 238000010606 normalization Methods 0.000 claims abstract description 12
- 239000013598 vector Substances 0.000 claims description 22
- 230000011218 segmentation Effects 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 10
- 238000000034 method Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 9
- 239000004973 liquid crystal related substance Substances 0.000 claims description 6
- 238000006116 polymerization reaction Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 6
- 230000006870 function Effects 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 238000007726 management method Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/353—Clustering; Classification into predefined classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- 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
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- 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
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The application relates to a user portrait technology, and discloses a crowd portrait construction method based on a graphic neural network, which comprises the following steps: acquiring user data, constructing a user association diagram according to the user data, and obtaining user nodes; extracting the characteristics of the user nodes to obtain node characteristics; feature aggregation is carried out on the node features of the adjacent users, so that aggregated features are obtained; weighting calculation is carried out on the aggregation characteristics and node characteristics corresponding to the aggregation characteristics, so as to obtain correction characteristics; and representing the correction characteristic as a user label of the node, and carrying out normalization calculation on the user label to obtain a user portrait. The application also provides a crowd portrayal construction device, electronic equipment and a storage medium based on the graph neural network. The application can improve the comprehensiveness of the feature information collection of the financial users.
Description
Technical Field
The application relates to the technical field of user portraits, in particular to a crowd portrayal construction method, a device, electronic equipment and a computer readable storage medium based on a graph neural network.
Background
The crowd portraits are basic technologies in the financial business field, and with the rise of crowd portraits technologies, effective crowd portraits can support mining of potential attributes of users, and are widely applied to financial management scenes such as commodity recommendation, banking user retention, user differentiation and the like. The crowd image technology adopts individual information of financial users to construct descriptive characteristics, and analyzes and models based on the characteristics, so that labels of corresponding images of the financial users are given. The collection completeness of the existing crowd image technology on the descriptive characteristics of the financial users is low, and the accuracy of the image labels of the financial users inferred based on the algorithm is also reduced due to the lack of the descriptive characteristics of the financial users, so that incorrect user labels are obtained, and the application effect of the downstream economic scene is affected.
In summary, the existing crowd portrayal construction technology has the problem that feature information of financial users is not fully collected.
Disclosure of Invention
The application provides a crowd portrayal construction method, a device, electronic equipment and a computer readable storage medium based on a graph neural network, and mainly aims to solve the problem that characteristic information of a user is not fully collected.
In order to achieve the above purpose, the crowd portrayal construction method based on the graph neural network provided by the application comprises the following steps:
acquiring user data, constructing a user association diagram according to the user data, and obtaining user nodes;
extracting the characteristics of the user nodes to obtain node characteristics;
feature aggregation is carried out on the node features of the adjacent users, so that aggregated features are obtained;
weighting calculation is carried out on the aggregation characteristics and node characteristics corresponding to the aggregation characteristics, so as to obtain correction characteristics;
and representing the correction characteristic as a user label of the node, and carrying out normalization calculation on the user label to obtain a user portrait.
Optionally, the extracting the characteristics of the user node to obtain node characteristics includes:
acquiring text information of each user node, and performing word segmentation on the text information to obtain a text identification sequence corresponding to each node;
inputting the text identification sequence into a preset semantic identification model to obtain initial node characteristics;
and carrying out convergence calculation on the initial node characteristics to obtain the node characteristics.
Optionally, the constructing a user association graph according to the user data to obtain a user node includes:
acquiring a data relationship between the user data, and taking the data relationship as an edge of the user association graph;
and inserting the users corresponding to the data relationship between the edges of the user association graph to obtain the nodes of the user association graph.
Optionally, the feature aggregation is performed on the node features of the adjacent users to obtain an aggregate feature, which includes:
combining the adjacent node characteristics to obtain combined characteristics;
performing polymerization iteration on the combined features to obtain iterative features;
and splicing the iteration features and the node features to obtain an aggregation feature.
Optionally, the performing aggregation iteration on the combined feature to obtain the iterative feature includes:
vector splicing is carried out on the combined features and the node features corresponding to the combined features, so that spliced vectors are obtained;
carrying out average value calculation on the spliced vectors to obtain average value vectors;
and inputting the mean vector into a preset convolution network, and performing aggregation calculation in the convolution layer to obtain aggregation characteristics.
Optionally, the weighting calculation is performed on the aggregate feature and the node feature corresponding to the aggregate feature to obtain a correction feature, which includes:
performing correlation calculation on the aggregation features and node features corresponding to the aggregation features to obtain correlation coefficients;
performing attention calculation on the node characteristics by using the correlation coefficient to obtain an attention coefficient;
and carrying out weighted summation on node characteristics corresponding to the aggregation characteristics according to the attention coefficient to obtain correction characteristics.
Optionally, the representing the correction feature as the user label of the node, performing normalization calculation on the user label to obtain a user portrait includes:
acquiring fact labels, classifying the correction features according to the fact labels, and obtaining a plurality of user label sets;
calculating the weight of the correction feature, and scoring and calculating the user tag set according to the weight of the correction feature to obtain a user tag score corresponding to the user tag set;
and screening the user labels according to the user label scores to obtain the user portrait.
In order to solve the above problems, the present application further provides a crowd portrayal construction device based on a graph neural network, the device comprising:
the user node generation module is used for acquiring user data, constructing a user association diagram according to the user data and obtaining user nodes;
the feature extraction module is used for extracting the features of the user nodes to obtain node features;
the feature aggregation module is used for carrying out feature aggregation on the node features of the adjacent users to obtain aggregated features;
the correction feature generation module is used for carrying out weighted calculation on the aggregation features and node features corresponding to the aggregation features to obtain correction features;
and the user portrait generation module is used for representing the correction characteristic as the user label of the node, and carrying out normalization calculation on the user label to obtain the user portrait.
In order to solve the above-mentioned problems, the present application also provides an electronic apparatus including:
at least one processor; the method comprises the steps of,
a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the above-described graph neural network-based crowd figure construction method.
In order to solve the above-mentioned problems, the present application further provides a computer readable storage medium, in which at least one computer program is stored, the at least one computer program being executed by a processor in an electronic device to implement the crowd figure constructing method based on a graph neural network.
The embodiment of the application provides a crowd portrayal construction method based on a graph neural network, which carries out weighted calculation on the aggregation characteristics of user nodes and the node characteristics corresponding to the aggregation characteristics to obtain correction characteristics, overcomes the defect of missing user node characteristics and improves the completeness of user node characteristic collection; the correction characteristics consider the adjacent user group attribute of the user node, and the obtained user label is more accurate; by constructing the user association graph according to the user data, the relation and the logic relation between the user data can be clearly and definitely arranged and analyzed, the accuracy of node feature extraction is improved, and therefore the accuracy of the user portrait label is increased. Therefore, the crowd portrayal construction method, the device, the electronic equipment and the computer readable storage medium based on the graph neural network can solve the problem that the feature information collection of the financial users is not comprehensive enough.
Drawings
FIG. 1 is a schematic flow chart of a crowd portrayal construction method based on a neural network according to an embodiment of the application;
fig. 2 is a schematic flow chart of extracting features of the user node to obtain node features according to an embodiment of the present application;
FIG. 3 is a schematic flow chart of feature aggregation of node features of adjacent users according to an embodiment of the present application to obtain aggregated features;
FIG. 4 is a functional block diagram of a crowd portrayal construction device based on a neural network according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device for implementing the crowd portrait construction method based on the graphic neural network according to an embodiment of the present application.
The achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The embodiment of the application provides a crowd portrayal construction method based on a graph neural network. The execution main body of the crowd portrayal construction method based on the graph neural network comprises at least one of a server, a terminal and the like which can be configured to execute the crowd portrayal construction method provided by the embodiment of the application. In other words, the crowd portrayal construction method based on the graph neural network can be executed by software or hardware installed in a terminal device or a server device, and the software can be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (ContentDelivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.
Referring to fig. 1, a flow chart of a crowd portrayal construction method based on a graph neural network according to an embodiment of the application is shown. In this embodiment, the crowd portrayal construction method based on the graph neural network includes:
s1, acquiring user data, and constructing a user association diagram according to the user data to obtain user nodes;
in the embodiment of the application, the user data are the transaction information and the payment behavior information related to the target client in the actual economic application scene, for example, in the application scene of financial data analysis, the user data are transaction account, transaction data and the like.
In the embodiment of the present application, the constructing a user association graph according to the user data to obtain a user node includes:
acquiring a data relationship between the user data, and taking the data relationship as an edge of the user association graph;
and inserting the users corresponding to the data relationship between the edges of the user association graph to obtain the nodes of the user association graph.
In the embodiment of the application, the data relationship is logic between financial user data, for example, the user name and the user age are one-to-one data relationship; the user association diagram is a representation method for analyzing the logic relationship between the user data; for example, for analysis of this application scenario for financial data, the user association graph may be constructed using a synthetic dataset generated by PaySim (mobile currency simulator), with the transaction account of each financial user as a node, each transaction being a continuous edge between two nodes; wherein the attribute of each node comprises mailbox, telephone, etc., and the attribute of the side comprises transaction amount, transaction type, etc.
S2, extracting characteristics of the user nodes to obtain node characteristics;
referring to fig. 2, in the embodiment of the present application, the feature extraction of the user node to obtain node features includes:
s21, obtaining text information of each user node, and performing word segmentation on the text information to obtain a text identification sequence corresponding to each node;
s22, inputting the text identification sequence into a preset semantic identification model to obtain initial node characteristics;
s23, carrying out convergence calculation on the initial node characteristics to obtain the node characteristics.
In the embodiment of the application, the meaning of the text information in different economic application scenes is different, and the text information can comprise inquiry sentences of banking business, text titles of financial news and the like; the embodiment of the application can perform word segmentation processing by using a dictionary word segmentation algorithm, wherein the dictionary word segmentation algorithm is to match text information to be analyzed with entries in a preset dictionary system according to a certain strategy, and if a certain character string is found in the dictionary system, the matching is successful; the text identification sequence is formed by splicing a plurality of word segmentation sequences, and because each text message is subjected to word segmentation processing, a plurality of word segmentation sequences are generated, for example, the text message A is subjected to word segmentation to obtain word segmentation 1 and word segmentation 2, the word segmentation 1 corresponds to the identification S1, the word segmentation 2 corresponds to the identification S2, and the identification is spliced in order to obtain the text identification sequence S1S2 of the text message A.
In the embodiment of the application, the preset semantic recognition model is a word embedding generator, which can generate an embedding vector for an input text identification sequence and encode and train the generated embedding vector to obtain the initial node characteristic; the convergence calculation feature convergence means that the maximum value or the average value is taken for each one-dimensional feature, and the non-deformation feature expression of the initial node feature can be obtained.
S3, carrying out feature aggregation on the node features of the adjacent users to obtain aggregated features;
referring to fig. 3, in the embodiment of the present application, feature aggregation is performed on the node features of the adjacent users to obtain aggregated features, including:
s31, combining adjacent node characteristics to obtain combined characteristics;
s32, performing polymerization iteration on the combined features to obtain iteration features;
and S33, splicing the iteration features and the node features to obtain an aggregation feature.
In the embodiment of the application, the combination is to combine one node characteristic with other node characteristics adjacent to the node characteristic; the combined feature is based on node sampling of each user node feature, for example, if the required adjacent node feature is N, that is, the sampled data is N, if the number of the user node features is less than N, a sampling method with substitution is adopted until N node features are sampled, and if the number of the user node features is greater than N, sampling without substitution is adopted, so that N node features are obtained.
In the embodiment of the application, the combination feature is subjected to polymerization iteration to obtain the iteration feature, which comprises the following steps:
vector splicing is carried out on the combined features and the node features corresponding to the combined features, so that spliced vectors are obtained;
carrying out average value calculation on the spliced vectors to obtain average value vectors;
and inputting the mean vector into a preset convolution network, and performing aggregation calculation in the convolution layer to obtain aggregation characteristics.
In the embodiment of the application, the preset convolution network is to aggregate the mean value vector from a high-dimensional vector to a low-dimensional vector; the mean value calculation is to average a plurality of spliced vectors according to each dimension, for example, the result obtained by carrying out the mean value calculation on three vectors [1,2,3,4], [2,3,4,5], [3,4,5,6] is [2,3,4,5].
S4, carrying out weighted calculation on the aggregation feature and the node feature corresponding to the aggregation feature to obtain a correction feature;
in the embodiment of the present application, the weighting calculation is performed on the aggregate feature and the node feature corresponding to the aggregate feature to obtain a correction feature, which includes:
performing correlation calculation on the aggregation features and node features corresponding to the aggregation features to obtain correlation coefficients;
performing attention calculation on the node characteristics by using the correlation coefficient to obtain an attention coefficient;
and carrying out weighted summation on node characteristics corresponding to the aggregation characteristics according to the attention coefficient to obtain correction characteristics.
In the embodiment of the application, the correlation calculation can be performed on the aggregation feature and the node feature corresponding to the aggregation feature by using the following formula:
e ij ＝f(wh i ||wh ij )
wherein e ij Attention coefficients between an ith node feature and a jth aggregation feature corresponding to the ith node feature; w is a preset sharing parameter; h is a ij The j aggregation features corresponding to the i node features; h is a i Is the ith node feature; f is a preset mapping function.
And S5, representing the correction characteristic as a user label of the node, and carrying out normalization calculation on the user label to obtain a user portrait.
In the embodiment of the present application, the representing the correction feature as the user label of the node, and performing normalization calculation on the user label to obtain a user portrait includes:
acquiring fact labels, classifying the correction features according to the fact labels, and obtaining a plurality of user label sets;
calculating the weight of the correction feature, and scoring and calculating the user tag set according to the weight of the correction feature to obtain a user tag score corresponding to the user tag set;
and screening the user labels according to the user label scores to obtain the user portrait.
In the embodiment of the application, the weight of the correction feature is obtained by calculating a weighted average value of the correction feature; the fact label is determined by financial user data in an actual application scenario, for example, the application scenario is analyzed for the financial data, and the fact label can be transaction times, transaction types, risk grades and the like of the financial user.
In the embodiment of the application, the scoring calculation can be performed on the user tag set according to the weight corresponding to the correction feature by using the following formula:
score＝x1*y1+x2*y2+…+xn*yn
wherein score represents the user tag score, xn represents the nth of the revised features; yn represents the weight corresponding to the nth correction feature.
In the embodiment of the application, the user labels are screened according to a preset label score threshold, when the user label score exceeds the label score threshold, the user labels are determined to be the component labels of the user portrait, and all the component labels are statistically combined to obtain the user portrait.
The application provides a crowd portrayal construction method based on a graph neural network, which carries out weighted calculation on the aggregation characteristics of user nodes of a financial user and node characteristics corresponding to the aggregation characteristics to obtain correction characteristics, overcomes the defect of missing user node characteristics and improves the completeness of user node characteristic collection; the correction characteristics consider the adjacent user group attribute of the user node, and the obtained user label is more accurate; by constructing the user association diagram according to the user data, the relation and the logic relation between the user data of the financial users can be clearly and definitely arranged and analyzed, the accuracy of node feature extraction is improved, and therefore the accuracy of the user portrait labels is increased. Therefore, the crowd portrayal construction method based on the graph neural network can solve the problem that the feature information of the financial user is not fully collected.
Fig. 4 is a functional block diagram of a crowd image constructing device based on a neural network according to an embodiment of the application.
The crowd figure constructing device 100 based on the graph neural network can be installed in electronic equipment. Depending on the functions implemented, the crowd portrayal construction device 100 based on the graph neural network may include a user node generation module 101, a feature extraction module 102, a feature aggregation module 103, a modified feature generation module 104, and a user portrayal generation module 105. The module of the application, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the present embodiment, the functions concerning the respective modules/units are as follows:
the user node generating module 101 is configured to obtain user data, and construct a user association graph according to the user data to obtain a user node;
the feature extraction module 102 is configured to perform feature extraction on the user node to obtain a node feature;
the feature aggregation module 103 is configured to perform feature aggregation on the node features of the adjacent users to obtain an aggregated feature;
the correction feature generation module 104 is configured to perform weighted calculation on the aggregate feature and a node feature corresponding to the aggregate feature to obtain a correction feature;
the user portrait generation module 105 is configured to represent the correction feature as a user label of the node, and perform normalization calculation on the user label to obtain a user portrait.
In detail, each module in the crowd figure constructing device 100 based on the graph neural network in the embodiment of the application adopts the same technical means as the crowd figure constructing method based on the graph neural network in the drawings when in use, and can generate the same technical effects, and the description thereof is omitted.
Fig. 5 is a schematic structural diagram of an electronic device for implementing a crowd portrait construction method based on a graphic neural network according to an embodiment of the present application.
The electronic device 1 may comprise a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may further comprise a computer program stored in the memory 11 and executable on the processor 10, such as a crowd portrayal construction program based on a neural network.
The processor 10 may be formed by an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be formed by a plurality of integrated circuits packaged with the same function or different functions, including one or more central processing units (Central Processing Unit, CPU), a microprocessor, a digital processing chip, a graphics processor, a combination of various control chips, and so on. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various components of the entire electronic device using various interfaces and lines, executes or executes programs or modules stored in the memory 11 (for example, executes a crowd image construction program based on a neural network, etc.), and invokes data stored in the memory 11 to perform various functions of the electronic device and process data.
The memory 11 includes at least one type of readable storage medium including flash memory, a removable hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device, such as a mobile hard disk of the electronic device. The memory 11 may in other embodiments also be an external storage device of the electronic device, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device. The memory 11 may be used to store not only application software installed in an electronic device and various data, such as codes of a crowd figure construction program based on a neural network, but also data that has been output or is to be output temporarily.
The communication bus 12 may be a peripheral component interconnect standard (Peripheral Component Interconnect, PCI) bus, or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc.
The communication interface 13 is used for communication between the electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), or alternatively a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device and for displaying a visual user interface.
Fig. 5 shows only an electronic device with components, it being understood by a person skilled in the art that the structure shown in fig. 5 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or may combine certain components, or may be arranged in different components.
For example, although not shown, the electronic device may further include a power source (such as a battery) for supplying power to the respective components, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device may further include various sensors, bluetooth modules, wi-Fi modules, etc., which are not described herein.
It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.
The crowd portrayal construction program based on the graph neural network stored in the memory 11 of the electronic device 1 is a combination of instructions, which when run in the processor 10, can implement:
acquiring user data, constructing a user association diagram according to the user data, and obtaining user nodes;
extracting the characteristics of the user nodes to obtain node characteristics;
feature aggregation is carried out on the node features of the adjacent users, so that aggregated features are obtained;
weighting calculation is carried out on the aggregation characteristics and node characteristics corresponding to the aggregation characteristics, so as to obtain correction characteristics;
and representing the correction characteristic as a user label of the node, and carrying out normalization calculation on the user label to obtain a user portrait.
In particular, the specific implementation method of the above instructions by the processor 10 may refer to the description of the relevant steps in the corresponding embodiment of the drawings, which is not repeated herein.
Further, the modules/units integrated in the electronic device 1 may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. The computer readable storage medium may be volatile or nonvolatile. For example, the computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).
The present application also provides a computer readable storage medium storing a computer program which, when executed by a processor of an electronic device, can implement:
acquiring user data, constructing a user association diagram according to the user data, and obtaining user nodes;
extracting the characteristics of the user nodes to obtain node characteristics;
feature aggregation is carried out on the node features of the adjacent users, so that aggregated features are obtained;
weighting calculation is carried out on the aggregation characteristics and node characteristics corresponding to the aggregation characteristics, so as to obtain correction characteristics;
and representing the correction characteristic as a user label of the node, and carrying out normalization calculation on the user label to obtain a user portrait.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. A plurality of units or means recited in the system claims can also be implemented by means of software or hardware by means of one unit or means. The terms first, second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present application and not for limiting the same, and although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present application without departing from the spirit and scope of the technical solution of the present application.
Claims (10)
1. The crowd portrayal construction method based on the graph neural network is characterized by comprising the following steps:
acquiring user data, constructing a user association diagram according to the user data, and obtaining user nodes;
extracting the characteristics of the user nodes to obtain node characteristics;
feature aggregation is carried out on the node features of the adjacent users, so that aggregated features are obtained;
weighting calculation is carried out on the aggregation characteristics and node characteristics corresponding to the aggregation characteristics, so as to obtain correction characteristics;
and representing the correction characteristic as a user label of the node, and carrying out normalization calculation on the user label to obtain a user portrait.
2. The crowd portrayal construction method based on a graph neural network according to claim 1, wherein the feature extraction is performed on the user nodes to obtain node features, including:
acquiring text information of each user node, and performing word segmentation on the text information to obtain a text identification sequence corresponding to each node;
inputting the text identification sequence into a preset semantic identification model to obtain initial node characteristics;
and carrying out convergence calculation on the initial node characteristics to obtain the node characteristics.
3. The crowd portrayal construction method based on a graph neural network according to claim 1, wherein the constructing a user association graph according to the user data to obtain a user node includes:
acquiring a data relationship between the user data, and taking the data relationship as an edge of the user association graph;
and inserting the users corresponding to the data relationship between the edges of the user association graph to obtain the nodes of the user association graph.
4. The crowd portrayal construction method based on a graph neural network according to claim 1, wherein the feature aggregation is performed on the node features of adjacent users to obtain an aggregate feature, and the method comprises:
combining the adjacent node characteristics to obtain combined characteristics;
performing polymerization iteration on the combined features to obtain iterative features;
and splicing the iteration features and the node features to obtain an aggregation feature.
5. The crowd portrayal construction method based on a graph neural network according to claim 4, wherein the performing an aggregation iteration on the combined features to obtain the iterative features includes:
vector splicing is carried out on the combined features and the node features corresponding to the combined features, so that spliced vectors are obtained;
carrying out average value calculation on the spliced vectors to obtain average value vectors;
and inputting the mean vector into a preset convolution network, and performing aggregation calculation in the convolution layer to obtain aggregation characteristics.
6. The crowd portrayal construction method based on a graph neural network according to claim 1, wherein the weighting calculation is performed on the aggregate feature and the node feature corresponding to the aggregate feature to obtain a correction feature, including:
performing correlation calculation on the aggregation features and node features corresponding to the aggregation features to obtain correlation coefficients;
performing attention calculation on the node characteristics by using the correlation coefficient to obtain an attention coefficient;
and carrying out weighted summation on node characteristics corresponding to the aggregation characteristics according to the attention coefficient to obtain correction characteristics.
7. The crowd portrayal construction method based on a graph neural network according to any one of claims 1 to 6, characterized in that the representing the correction feature as a user tag of the node, performing normalization calculation on the user tag to obtain the user portrayal includes:
acquiring fact labels, classifying the correction features according to the fact labels, and obtaining a plurality of user label sets;
calculating the weight of the correction feature, and scoring and calculating the user tag set according to the weight of the correction feature to obtain a user tag score corresponding to the user tag set;
and screening the user labels according to the user label scores to obtain the user portrait.
8. Crowd portrayal construction device based on picture neural network, characterized by that, said device includes:
the user node generation module is used for acquiring user data, constructing a user association diagram according to the user data and obtaining user nodes;
the feature extraction module is used for extracting the features of the user nodes to obtain node features;
the feature aggregation module is used for carrying out feature aggregation on the node features of the adjacent users to obtain aggregated features;
the correction feature generation module is used for carrying out weighted calculation on the aggregation features and node features corresponding to the aggregation features to obtain correction features;
and the user portrait generation module is used for representing the correction characteristic as the user label of the node, and carrying out normalization calculation on the user label to obtain the user portrait.
9. An electronic device, the electronic device comprising:
at least one processor; the method comprises the steps of,
a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the graph neural network-based crowd figure construction method of any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the crowd figure construction method based on a neural network according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310694316.4A CN116663556A (en) | 2023-06-12 | 2023-06-12 | Crowd portrayal construction method, device, equipment and medium based on graph neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310694316.4A CN116663556A (en) | 2023-06-12 | 2023-06-12 | Crowd portrayal construction method, device, equipment and medium based on graph neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116663556A true CN116663556A (en) | 2023-08-29 |
Family
ID=87722289
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310694316.4A Pending CN116663556A (en) | 2023-06-12 | 2023-06-12 | Crowd portrayal construction method, device, equipment and medium based on graph neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116663556A (en) |
-
2023
- 2023-06-12 CN CN202310694316.4A patent/CN116663556A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113822494B (en) | Risk prediction method, device, equipment and storage medium | |
| CN115002200B (en) | Message pushing method, device, equipment and storage medium based on user portrait | |
| CN113157927B (en) | Text classification method, apparatus, electronic device and readable storage medium | |
| CN114398557B (en) | Information recommendation method and device based on double images, electronic equipment and storage medium | |
| CN113592605B (en) | Product recommendation method, device, equipment and storage medium based on similar products | |
| CN113706291A (en) | Fraud risk prediction method, device, equipment and storage medium | |
| CN112328657A (en) | Feature derivation method, feature derivation device, computer equipment and medium | |
| CN116681082A (en) | Discrete text semantic segmentation method, device, equipment and storage medium | |
| CN113360654B (en) | Text classification method, apparatus, electronic device and readable storage medium | |
| CN113157853B (en) | Problem mining method, device, electronic equipment and storage medium | |
| CN113157739B (en) | Cross-modal retrieval method and device, electronic equipment and storage medium | |
| CN115204971B (en) | Product recommendation method, device, electronic equipment and computer readable storage medium | |
| CN116578696A (en) | Text abstract generation method, device, equipment and storage medium | |
| CN113704474B (en) | Bank outlet equipment operation guide generation method, device, equipment and storage medium | |
| CN113806540B (en) | Text labeling method, text labeling device, electronic equipment and storage medium | |
| CN113515591B (en) | Text defect information identification method and device, electronic equipment and storage medium | |
| CN115114408A (en) | Multi-modal emotion classification method, device, equipment and storage medium | |
| CN113706207A (en) | Order transaction rate analysis method, device, equipment and medium based on semantic analysis | |
| CN116663556A (en) | Crowd portrayal construction method, device, equipment and medium based on graph neural network | |
| CN113626605A (en) | Information classification method and device, electronic equipment and readable storage medium | |
| CN112988963B (en) | User intention prediction method, device, equipment and medium based on multi-flow nodes | |
| WO2023178798A1 (en) | Image classification method and apparatus, and device and medium | |
| CN112214556B (en) | Label generation method, label generation device, electronic equipment and computer readable storage medium | |
| CN114723488B (en) | Course recommendation method and device, electronic equipment and storage medium | |
| CN114781833B (en) | Capability assessment method, device and equipment based on business personnel and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |