CN113297500B - Social network isolated node link prediction method - Google Patents
Social network isolated node link prediction method Download PDFInfo
- Publication number
- CN113297500B CN113297500B CN202110697470.8A CN202110697470A CN113297500B CN 113297500 B CN113297500 B CN 113297500B CN 202110697470 A CN202110697470 A CN 202110697470A CN 113297500 B CN113297500 B CN 113297500B
- Authority
- CN
- China
- Prior art keywords
- vector
- node
- supervised
- network
- semi
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 239000013598 vector Substances 0.000 claims abstract description 83
- 238000013507 mapping Methods 0.000 claims abstract description 15
- 230000006870 function Effects 0.000 claims description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000010801 machine learning Methods 0.000 claims description 6
- 210000002569 neuron Anatomy 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 9
- 230000003042 antagnostic effect Effects 0.000 abstract description 3
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 238000007476 Maximum Likelihood Methods 0.000 description 2
- 230000008485 antagonism Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- 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
-
- 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
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/061—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using biological neurons, e.g. biological neurons connected to an integrated circuit
-
- 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
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Business, Economics & Management (AREA)
- Neurology (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Primary Health Care (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Marketing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of social network link prediction, and particularly relates to a social network isolated node link prediction method. Aiming at an isolated node prediction task in social network link prediction, the invention provides a social network isolated node link prediction method for semi-supervised link prediction by adopting auxiliary information. According to the invention, a mapping model is learned through attribute vectors and topology vectors of nodes in a known network; mapping the attribute vector of the node to be predicted into a topology vector by using the model; semi-supervised generation is used for antagonizing a network, a link prediction task is completed by using the topology vector of a node to be predicted and the topology vector of a current network node, and feasibility and advantages of a model are verified on a social data set. The method and the device can be used for solving the problem of predicting the isolated node in the social network link prediction process, and get rid of the dependence on the sample containing the label to a certain extent in the prediction process.
Description
Technical Field
The invention belongs to the technical field of social network link prediction, and particularly relates to a social network isolated node link prediction method.
Background
With the progress of the computer software and hardware level and the continuous development of network and communication means, many excellent social network applications are created and gradually penetrate into the aspects of people's life. As the number of users has increased geometrically throughout the year, these platforms accumulate vast amounts of data that can be analyzed to generate information useful for production and life. Outside the country, people commonly chat with friends using Twitter and Facebook, sharing their own state of life; and meanwhile, linkedIn is used for performing activities such as office vein expansion, technical experience exchange and company information acquisition. In China, people usually use microblogs to browse various hot news, conduct activities such as praise, forwarding and comment posting, and can pay attention to accounts of other people to acquire the latest dynamics of the accounts concerned. Meanwhile, QQ and WeChat provide functions such as real-time information exchange and text content browsing among friends for people. Social networks provide many application services including friend recommendations, product recommendations, knowledge network construction, etc., and the core process of implementing such services is to accurately mine relationships between various entities in the social network. This process may be referred to as social network entity link prediction, hereinafter simply link prediction.
One class of methods of link prediction is based on maximum likelihood estimation. Newman, clauset and Moore consider links to be a reflection of the underlying hierarchy and propose an algorithm for maximum likelihood estimation based on this assumption to make link predictions. This approach is applicable to networks with obvious hierarchies, such as family relation networks or food chain networks in local environments. Another class of methods is predictive methods that utilize node attributes. Heaukulani et al classify by directly extracting feature information associated with nodes as input to a random forest. O' Madadhain et al consider that existing similarity metrics are not fully applicable to all networks, and therefore propose a method of employing metric learning to generate a metric between node attributes, thereby predicting the likelihood of links between nodes.
In addition, some scholars propose a link prediction method based on node similarity. The link prediction method based on node similarity is generally based on the following assumption: the higher the similarity of two nodes, the greater the likelihood that a link exists between the two nodes. The core idea of the method is to judge whether the point pairs are links or not based on the threshold value of the index of the similarity of the specific description point pairs.
Therefore, many scholars propose different similarity indexes to measure the similarity of the nodes, and common neighbors, cosine similarity, jaccard coefficients and the like are commonly used. The earliest used similarity evaluation index is a common neighbor, which means that if two nodes have more common neighbors, then the two nodes can be considered to have stronger similarity. Many link prediction related studies are improved based on common neighbor similarity indicators. Jaccard has proposed that Jaccard coefficients be applied in link prediction. In recent years, research on link prediction similarity indexes has been carried out in various versions, and the research has been successfully applied to link prediction of various networks.
With the development of machine learning, some scholars have proposed a link prediction algorithm based on machine learning. Al Hasan M et Al applied a machine-learned Support Vector Machine (SVM) model to the link prediction and performed experimental verification on DBLP and BIOBASE and both datasets, proving the effectiveness of the support vector machine model relative to other machine learning algorithms. The Hasan et al uses a plurality of different classification algorithms to carry out link prediction experiments on a real data set, most of the Hasan et al obtain better experimental effects, and the feasibility of the link prediction algorithm adopting supervised learning is also proved.
However, in the development of the above-mentioned link prediction algorithm, researchers have focused on the problem of link prediction over a complete network. But in real world demands can sometimes abstract a task of predicting the relationship that an orphan node may have with the current network. The method has important significance for improving the service quality of the social network and has a strong practical application background. The greatest difficulty with such tasks is how to obtain topology information and other ancillary information of the nodes to be predicted. The topology information of the node refers to the link relation between the node and other nodes in the network; and the auxiliary information includes, for example, user profile (User profile), text, etc. of the node. Because most social networks require users to register first when they are first used, the social network platform can obtain the above auxiliary information, such as user portraits, before these newly added users are not in contact with other users. Thus, based on this behavior of the new user, we can predict the link relationships that the user may have in the future by using these initial assistance information.
Except for the problem that the link prediction task may encounter an isolated node prediction task in practical application, the label quality of the training data set is depended on based on the supervised learning social network link prediction model, which brings high cost to the algorithm. The semi-supervised model can efficiently complete model training in a data set using a small number of labels, so that the model adopting semi-supervised learning can be better suitable for a link prediction task.
Disclosure of Invention
The invention aims to provide a social network isolated node link prediction method.
The aim of the invention is realized by the following technical scheme: the method comprises the following steps:
step 1: inputting a current network topology structure of a social network, and generating an embedded vector for nodes in the current network by adopting a network embedding algorithm;
step 2: collecting attribute vectors of current network nodes, and learning a mapping model formed by a multi-layer perceptron MLP by using the attribute vectors of each node and the embedded vectors thereof;
if the attribute vector dimension of the user is D and the dimension of the embedded vector of the user is D, obviously the number of neurons of the input layer of the multi-layer perceptron is D and the number of neurons of the output layer is D; let the input user attribute vector be x= [ X ] 1 ,x 2 ,x 3 ,...,x D ]The multi-layer perceptron can be regarded as a function f MLP (x) The output of the input vector through the multi-layer perceptron is:
X'＝[x' 1 ,x' 2 ,...,x' d ]＝f MLP (X)
in order to map the attribute vector to the embedded vector well by the model, the mean square error is adopted as a loss function, and the embedded vector of the corresponding user is set as Y= [ Y ] 1 ,y 2 ,...,y d ]The loss function of the model is:
step 3: collecting attribute vectors of the isolated nodes, and learning to obtain potential embedded vectors of the isolated nodes by using the learned multi-layer perceptron mapping model;
step 4: splicing the embedded vector of the isolated node with the embedded vector of the current node in the network, and partially endowing the label with a sample for machine learning;
step 5: constructing a semi-supervised generation countermeasure network;
step 6: learning from a sample by adopting a semi-supervised generation countermeasure network to obtain a semi-supervised link prediction model, and carrying out link prediction by using the semi-supervised link prediction model;
after obtaining the embedded vector of the isolated node, the vector of the two users is spliced to form the input of the semi-supervision generation countermeasure network; let user u 1 Is [ x ] 1 ,x 2 ,...,x d ]The method comprises the steps of carrying out a first treatment on the surface of the User u 2 Is [ y ] 1 ,y 2 ,...,y d ]The input of semi-supervised generation of the challenge model is [ x ] 1 ,x 2 ,...,x d ,y 1 ,y 2 ,...,y d ；L]Wherein L represents a label for indicating whether a link relationship exists between two user pairs; the user can be the existing node or the isolated node of the current network, and the input node pair vector does not have labels;
since there are 3 types of predictors: the real data, the generated data and the tagged data are changed, so that the output shape of the discriminator is changed, and the output of a single scalar is changed into the output of a three-dimensional vector: [ Link, unlink, fake ]; wherein Link indicates that links exist between users; the Unlink indicates that no links are present; fake represents the generation of samples;
the loss function of the arbiter uses the sum of the supervised and semi-supervised losses:
L D ＝L supervised +L unsupervised
the loss function of the generator is:
L G ＝L adversarial +L feature
L adversarial ＝-E x～g(x) log[1-D(y|x,y＝K+1)]
in the semi-supervised link prediction model, a generator and a discriminator all adopt fully connected networks, and in order to prevent the phenomenon of mode collapse and the occurrence of over fitting, batch regularization and dropout technologies are adopted, and the generator and the discriminator adopt symmetrical structures.
The invention has the beneficial effects that:
aiming at an isolated node prediction task in social network link prediction, the invention provides a social network isolated node link prediction method for semi-supervised link prediction by adopting auxiliary information. According to the invention, a mapping model is learned through attribute vectors and topology vectors of nodes in a known network; mapping the attribute vector of the node to be predicted into a topology vector by using the model; semi-supervised generation is used for antagonizing a network, a link prediction task is completed by using the topology vector of a node to be predicted and the topology vector of a current network node, and feasibility and advantages of a model are verified on a social data set. The method and the device can be used for solving the problem of predicting the isolated node in the social network link prediction process, and get rid of the dependence on the sample containing the label to a certain extent in the prediction process.
Drawings
FIG. 1 is a map model diagram of a multi-layer perceptron.
FIG. 2 is a model diagram of a semi-supervised generation countermeasure network.
FIG. 3 is an overall block diagram of a semi-supervised orphan predictive model.
Fig. 4 is a table of settings for generating an impedance network.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
The invention relates to the field of social network link prediction, in particular to a prediction method of isolated nodes in a social network. The invention aims to solve the problem of predicting isolated nodes in the social network link prediction process, and get rid of the dependence on a sample containing labels to a certain extent in the prediction process.
The purpose of the invention is realized in the following way:
step 1, inputting a current network topology structure of a social network, and generating an embedded vector for a node in the current network by adopting a certain network embedding algorithm;
step 2, collecting attribute vectors of the current network node, and learning a mapping model formed by a multi-layer perceptron (MLP) by using the attribute vectors of each node and the embedded vectors thereof;
step 3, collecting attribute vectors of the isolated nodes, and learning to obtain potential embedded vectors of the isolated nodes by using the learned multi-layer perceptron mapping model;
step 4, splicing the embedded vector of the isolated node with the embedded vector of the current node in the network, and partially endowing the label with a sample for forming machine learning;
step 5, constructing a semi-supervised generation countermeasure network;
and 6, learning from the sample by adopting the generated countermeasure network to obtain a semi-supervised link prediction model, and carrying out link prediction by using the semi-supervised link prediction model.
The invention provides a semi-supervised link prediction model by adopting auxiliary information aiming at an isolated node prediction task in social network link prediction. Firstly, a mapping model is learned through the attribute vector and the topology vector of the nodes in the known network, and the attribute vector of the node to be predicted is mapped into the topology vector by using the mapping model. Semi-supervised generation is used for antagonizing a network, a link prediction task is completed by using the topology vector of a node to be predicted and the topology vector of a current network node, and feasibility and advantages of a model are verified on a social data set.
Example 1:
1. mapping of node attribute vectors to embedded vectors
Designating an existing node in the network as u 1 ～u N1 It has both attribute vector and network structure information, so in the present invention, the network embedding algorithm is first adopted as u 1 ～u N1 Each node generates an embedded vector; and then a mapping is learned through the embedded vector and the attribute vector of the node, and the mapping can well convert the attribute vector of the node into a corresponding embedded vector.
If the attribute vector dimension of the user is D and the dimension of the embedded vector of the user is D, obviously the number of neurons of the input layer of the multi-layer perceptron is D and the number of neurons of the output layer is D; let the input user attribute vector be x= [ X ] 1 ,x 2 ,x 3 ,...,x D ]The multi-layer perceptron can be regarded as a function f MLP (x) The output of the input vector through the multi-layer perceptron is:
X'＝[x' 1 ,x' 2 ,...,x' d ]＝f MLP (X) (1)
in order to well map the attribute vector to the embedded vector by the model, the invention adopts the mean square error as a loss function, and sets the embedded vector of the corresponding user as Y= [ Y ] 1 ,y 2 ,...,y d ]The loss function of the model is:
2. semi-supervised link prediction method based on generation of countermeasure network
After obtaining the embedded vector of the isolated node, the invention constructs the input of the semi-supervised generation countermeasure network by splicing the vectors of the two users. Let user u 1 Is [ x ] 1 ,x 2 ,...,x d ]The method comprises the steps of carrying out a first treatment on the surface of the User u 2 Is [ y ] 1 ,y 2 ,...,y d ]The input of the semi-supervised generation countermeasure model in the invention is [ x ] 1 ,x 2 ,...,x d ,y 1 ,y 2 ,...,y d ；L]Where L represents a label to indicate whether a link relationship exists between two user pairs. The user may be a node existing in the current network or an isolated node, and the input node pair vector does not have a label.
After the generation of the antagonism network model, a learner tries to complete the semi-supervised learning task by generating the antagonism network through different technical angles. The invention refers to the related thought, and the dimension of the output result of the discriminator is increased to enable the generation of the countermeasure network to process the data with the classification labels:
in a conventional generation countermeasure network, the arbiter D outputs a single scalar representing the magnitude of the probability that the input belongs to a true sample. In the present invention, 3 kinds of prediction results exist: the invention changes the output shape of the discriminator, and changes the output of a single scalar into the output of a three-dimensional vector: [ Link, unlink, fake ], where Link indicates that there is a Link between users, unlink indicates that there is no Link, and like indicates that the loss function of the arbiter generates samples such that the sum of supervised and semi-supervised losses can be used:
for the loss of the generator, the model of the invention uses the sum of classical generated sample loss and feature matching loss. The feature matching loss refers to the difference between the distribution of the data generated by the generator after the feature extraction of the discriminator and the distribution of the real data, and is expressed as follows:
thus, the loss function of the generator can be expressed as follows:
the semi-supervised generation of a link prediction model for the countermeasure network is shown in fig. 2.
The generator and the arbiter all adopt fully connected networks in the model, and batch regularization (batch_normalization) and dropout technologies are adopted in order to prevent the phenomenon of pattern collapse and the occurrence of over fitting problems. The generator and the arbiter are of symmetrical construction, the detailed setup of which is shown in fig. 4.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (1)
1. The social network isolated node link prediction method is characterized by comprising the following steps of:
step 1: inputting a current network topology structure of a social network, and generating an embedded vector for nodes in the current network by adopting a network embedding algorithm;
step 2: collecting attribute vectors of current network nodes, and learning a mapping model formed by a multi-layer perceptron MLP by using the attribute vectors of each node and the embedded vectors thereof;
if the attribute vector dimension of the user is D and the dimension of the embedded vector of the user is D, obviously the number of neurons of the input layer of the multi-layer perceptron is D and the number of neurons of the output layer is D; let the input user attribute vector be x= [ X ] 1 ,x 2 ,x 3 ,...,x D ]The multi-layer perceptron can be regarded as a function f MLP (x) The output of the input vector through the multi-layer perceptron is:
X′＝[x′ 1 ,x′ 2 ,...,x′ d ]＝f MLP (X)
in order to map the attribute vector to the embedded vector by the model, the mean square error is used as a loss function, and the embedded vector of the corresponding user is set as Y= [ Y ] 1 ,y 2 ,...,y d ]The loss function of the model is:
step 3: collecting attribute vectors of the isolated nodes, and learning to obtain potential embedded vectors of the isolated nodes by using the learned multi-layer perceptron mapping model;
step 4: splicing the embedded vector of the isolated node with the embedded vector of the current node in the network, and partially endowing the label with a sample for machine learning;
step 5: constructing a semi-supervised generation countermeasure network;
step 6: learning from a sample by adopting a semi-supervised generation countermeasure network to obtain a semi-supervised link prediction model, and carrying out link prediction by using the semi-supervised link prediction model;
after obtaining the embedded vector of the isolated node, the vector of the two users is spliced to form the input of the semi-supervision generation countermeasure network; let user u 1 Is [ x ] 1 ,x 2 ,...,x d ]The method comprises the steps of carrying out a first treatment on the surface of the User u 2 Is [ y ] 1 ,y 2 ,...,y d ]The input of semi-supervised generation of the challenge model is [ x ] 1 ,x 2 ,...,x d ,y 1 ,y 2 ,...,y d ；L]Wherein L represents a label for indicating whether a link relationship exists between two user pairs; the user can be the existing node or the isolated node of the current network, and the input node pair vector does not have labels;
since there are 3 types of predictors: the real data, the generated data and the tagged data are changed, so that the output shape of the discriminator is changed, and the output of a single scalar is changed into the output of a three-dimensional vector: [ Link, unlink, fake ]; wherein Link indicates that links exist between users; the Unlink indicates that no links are present; fake represents the generation of samples;
the loss function of the arbiter uses the sum of the supervised and semi-supervised losses:
L D ＝L supervised +L unsupervised
the loss function of the generator is:
L G ＝L adversarial +L feature
L adversarial ＝-E x～g(x) log[1-D(y|x，y＝K+1)]
in the semi-supervised link prediction model, a generator and a discriminator all adopt fully connected networks, and in order to prevent the phenomenon of mode collapse and the occurrence of over fitting, batch regularization and dropout technologies are adopted, and the generator and the discriminator adopt symmetrical structures.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110697470.8A CN113297500B (en) | 2021-06-23 | 2021-06-23 | Social network isolated node link prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110697470.8A CN113297500B (en) | 2021-06-23 | 2021-06-23 | Social network isolated node link prediction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113297500A CN113297500A (en) | 2021-08-24 |
| CN113297500B true CN113297500B (en) | 2023-07-25 |
Family
ID=77329298
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110697470.8A Active CN113297500B (en) | 2021-06-23 | 2021-06-23 | Social network isolated node link prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113297500B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117349950B (en) * | 2023-12-05 | 2024-02-02 | 华设检测科技有限公司 | Assembly type roadbed optimization method based on generated countermeasure type network and random forest unit |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104134159A (en) * | 2014-08-04 | 2014-11-05 | 中国科学院软件研究所 | Method for predicting maximum information spreading range on basis of random model |
| CN113011282A (en) * | 2021-02-26 | 2021-06-22 | 腾讯科技（深圳）有限公司 | Graph data processing method and device, electronic equipment and computer storage medium |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107784124B (en) * | 2017-11-23 | 2021-08-24 | 重庆邮电大学 | LBSN (location based service) hyper-network link prediction method based on space-time relationship |
| CN108449209A (en) * | 2018-03-17 | 2018-08-24 | 北京工业大学 | The social networks friend recommendation method merged based on routing information and nodal information |
| CN109543114A (en) * | 2018-11-14 | 2019-03-29 | 南京邮电大学 | Heterogeneous Information network linking prediction technique, readable storage medium storing program for executing and terminal |
| CN110677284B (en) * | 2019-09-24 | 2022-06-17 | 北京工商大学 | Heterogeneous network link prediction method based on meta path |
| KR102234850B1 (en) * | 2019-11-15 | 2021-04-02 | 숭실대학교산학협력단 | Method and apparatus for complementing knowledge based on relation network |
| CN111611785A (en) * | 2020-04-30 | 2020-09-01 | 北京理工大学 | Generation type confrontation network embedded representation learning method |
| CN111475739B (en) * | 2020-05-22 | 2022-07-29 | 哈尔滨工程大学 | Heterogeneous social network user anchor link identification method based on meta-path |
| CN112767186B (en) * | 2021-01-26 | 2022-11-18 | 东南大学 | Social network link prediction method based on 7-subgraph topological structure |
-
2021
- 2021-06-23 CN CN202110697470.8A patent/CN113297500B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104134159A (en) * | 2014-08-04 | 2014-11-05 | 中国科学院软件研究所 | Method for predicting maximum information spreading range on basis of random model |
| CN113011282A (en) * | 2021-02-26 | 2021-06-22 | 腾讯科技（深圳）有限公司 | Graph data processing method and device, electronic equipment and computer storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113297500A (en) | 2021-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yan et al. | Learning URL embedding for malicious website detection | |
| CN113961759B (en) | Abnormality detection method based on attribute map representation learning | |
| CN111881350A (en) | Recommendation method and system based on mixed graph structured modeling | |
| CN112215604B (en) | Method and device for identifying transaction mutual-party relationship information | |
| CN104995870A (en) | Multi-objective server placement determination | |
| Qiu et al. | An adaptive social spammer detection model with semi-supervised broad learning | |
| Xu et al. | A scientific research topic trend prediction model based on multi‐LSTM and graph convolutional network | |
| Liu et al. | Cloud model-based PROMETHEE method under 2D uncertain linguistic environment | |
| Zheng et al. | Jora: Weakly supervised user identity linkage via jointly learning to represent and align | |
| Farhadi et al. | Friendship selection and management in social internet of things: A systematic review | |
| Feng et al. | H-Diffu: hyperbolic representations for information diffusion prediction | |
| CN113297500B (en) | Social network isolated node link prediction method | |
| Geng et al. | Applying the blockchain-based deep reinforcement consensus algorithm to the intelligent manufacturing model under internet of things | |
| Wu et al. | Instant Representation Learning for Recommendation over Large Dynamic Graphs | |
| Xiao et al. | WSGCN4SLP: Weighted signed graph convolutional network for service link prediction | |
| Shirgahi et al. | Trust estimation of the semantic web using semantic web clustering | |
| Gong et al. | Detecting malicious accounts in online developer communities using deep learning | |
| Le et al. | Enhancing Anchor Link Prediction in Information Networks through Integrated Embedding Techniques | |
| Chen et al. | Information cascades prediction with graph attention | |
| Chen et al. | A supervised link prediction method for dynamic networks | |
| CN116306834A (en) | Link prediction method based on global path perception graph neural network model | |
| Gao et al. | A Novel Blockchain-Based Responsible Recommendation System for Service Process Creation and Recommendation | |
| Zhao et al. | Detecting fake reviews via dynamic multimode network | |
| Tao et al. | Structural identity representation learning of blockchain transaction network for metaverse | |
| CN112396237A (en) | Link prediction method in social network |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |