CN111476673A - Method, device and medium for aligning users among social networks based on neural network - Google Patents
Method, device and medium for aligning users among social networks based on neural network Download PDFInfo
- Publication number
- CN111476673A CN111476673A CN202010255397.4A CN202010255397A CN111476673A CN 111476673 A CN111476673 A CN 111476673A CN 202010255397 A CN202010255397 A CN 202010255397A CN 111476673 A CN111476673 A CN 111476673A
- Authority
- CN
- China
- Prior art keywords
- user
- alignment
- model
- social network
- nodes
- 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 48
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 27
- 239000013598 vector Substances 0.000 claims abstract description 75
- 238000012549 training Methods 0.000 claims abstract description 28
- 238000005295 random walk Methods 0.000 claims abstract description 15
- 238000013135 deep learning Methods 0.000 claims abstract description 4
- 238000005070 sampling Methods 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 16
- 238000013507 mapping Methods 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 8
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 230000007246 mechanism Effects 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 3
- 238000012545 processing Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000000547 structure data Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012916 structural analysis Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06N3/045—Combinations of networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Business, Economics & Management (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Machine Translation (AREA)
Abstract
The invention provides a method, a device and a medium for aligning users among social networks based on a neural network, which can effectively improve the alignment precision of the social networks, and the method comprises the following steps: step 1: sampling each user node of the social network to generate a random walk sequence of the user nodes; step 2: representing the network nodes in the random walk sequence as user vectors by using a Word2vec tool; and step 3: adopting a deep learning neural network to construct an alignment model, training by taking the aligned user vector as input, and optimizing model parameters to obtain the alignment model; and 4, step 4: and inputting user vectors corresponding to all user nodes in the first social network into the alignment model, outputting the representation of the user vectors of the user nodes in the second social network, searching the user vectors which are most similar to the output of the alignment model in the second social network respectively, and judging whether the users corresponding to the user nodes are the same user or not according to the similarity.
Description
Technical Field
The invention relates to the field of neural networks in machine learning, in particular to a method, a device and a medium for aligning users among social networks based on the neural networks.
Background
With the gradual popularization of online social networks, online social network services have penetrated the aspects of life and gradually become a part of life, and meanwhile, different online social platforms have different starting points and are not limited to online communication. Online social network data has become an important source of internet big data, with an average of 8.5 online social network accounts per netizen, and each platform is intended for different purposes. When the user uses the online social platform, massive data including attention, concerned and personal information of the user can be generated, and the public information is used for associating the same user entity in different social platforms, so that commodity popularization, friend recommendation and the like can be performed more conveniently and effectively. When a popularization person carries out product popularization, the same user of different social networks is identified, then commodity popularization can be carried out for different users, and resource waste due to repeated pushing can be effectively avoided; similarly, according to the structural analysis of the user in different online social networks, possible friends of the user are recommended, so that the user can establish more friend relationships, and the user stickiness is improved.
Most of the existing network alignment methods are based on a network structure, firstly, a network is modeled, various social relationships among users in the network are simulated according to network characteristics and self-defined node characteristics such as node shortest paths, node degree distribution, common neighbor numbers and the like, the nodes in the social network to be mapped are calculated and ranked according to the probability, and two nodes with the highest matching degree are considered as the same user.
Due to the heterogeneity of the social network and the real social relationship network, people who have intersection in the real society are not always connected on the social platform, that is, the online and offline friend relationship networks are not equal, and the incomplete network structure can cause data processing difficulty; secondly, the method based on the network structure only considers local or integral network structure information, does not deeply mine potential network structure relationship, and inevitably omits user identity information during calculation.
The method has the advantages that the network feature extraction is needed when the user is linked by the network structure, the manual feature extraction efficiency of the existing method is low, meanwhile, the method is not easy to expand among other social networks, and the calculation complexity is increased when a new network is faced. The difference of different online social network service contents causes great difference of activities of users on different platforms, and the manual feature extraction cannot take local and overall structural information into consideration at the same time, so that the matching result precision is low.
Disclosure of Invention
In view of the above problems, the present invention provides a method, an apparatus, and a medium for aligning users between social networks based on a neural network, which can effectively improve the accuracy of social network alignment.
The technical scheme is as follows: a method for user alignment between social networks based on neural networks, comprising the steps of:
step 1: sampling each user node of the social network to generate a random walk sequence of the user nodes;
step 2: representing the network nodes in the random walk sequence as user vectors by using a Word2vec tool;
and step 3: adopting a deep learning neural network to construct an alignment model, training by taking the aligned user vector as input, and optimizing model parameters to obtain the alignment model;
and 4, step 4: and inputting user vectors corresponding to all user nodes in the first social network into the alignment model, outputting the representation of the user vectors of the user nodes in the second social network, searching the user vectors which are most similar to the output of the alignment model in the second social network respectively, and judging whether the users corresponding to the user nodes are the same user or not according to the similarity.
Further, in step 1, for each user in the social network, taking the user node corresponding to the user as a starting point, randomly accessing the neighbor node of the user node as a next hop, then randomly accessing the neighbor node again with the next hop as the starting point, according to the user average degree w in the network, until a fixed length sequence length is 2 x w, and then repeating the above process N times, where N is a natural number, to obtain a random walk sequence of the user node.
Further, in step 2, training a random walk sequence of user nodes based on a Skip-Gram model, obtaining a weight matrix learned by model training after the model training is completed, encoding all the user nodes into a one-hot form, inputting and mapping the one-hot form into the weight matrix, calculating the conditional probability of a given central node generating adjacent nodes by using gradient descent and maximizing the probability to obtain a mapping table of the nodes and embedded vectors, and respectively representing all the nodes in different social networks as user vectors in a low-dimensional space through table lookup after the training is completed.
Further, the step 3 specifically includes the following steps: dividing user vectors corresponding to aligned user nodes in two social networks into a test set and a training set, constructing an alignment model based on a Seq2Seq model added with an Attention mechanism, wherein the alignment model is used for mapping the user vectors in one social network to the other social network, taking the user vectors in the training set as an input training alignment model, performing supervised learning, learning the mapping relationship in the two social networks, outputting vectors with the same dimensionality as the input user vectors by the alignment model, indicating that the model tends to be stable when the overall loss function value of the alignment model tends to be stable and does not change, stopping training, and testing the effect of the alignment model through the data of the test set to judge whether the model is effective.
Further, in step 3, Dropout method is used to discard part of neurons randomly for preventing the alignment model from overfitting.
Further, in step 3, the loss function is calculated by the following formula:
where V represents a user vector, cos (V (u)i))，v(uj) Cosine values between two vectors, the smaller the loss function is, the better the alignment model is represented, and according to the size of the loss function, the corresponding weight is adjusted, so that the same user can be similar to the alignment model as much as possible.
Further, in step 4, the user vector in the first social network is input into the alignment model and then output, whether the output user vector is the same user is judged through the cosine similarity between the output user vector and the user vector in the second social network, the cosine similarity is calculated through the user vector in the second social network, and if the value is greater than the judgment value under the condition of taking the maximum value of the cosine value, the user corresponding to the user vector in the second social network under the condition of the maximum value of the cosine value is considered to be the same user of the input user corresponding to the user vector in the first social network.
Compared with the prior art, the method, the device and the medium for aligning the users among the social networks based on the neural network have the advantages that the characteristics do not need to be manually extracted in the whole aligning process, the local and overall network structure information can be effectively reserved, the link precision is generally improved in different indexes compared with other user aligning methods, the average link precision is improved by 7%, the aligned users can be locked in a smaller range, the network structure characteristics are automatically extracted under the condition that only the social network structure data exist, the hidden information of the social network structure is mined, and the local information and the overall information of the network structure are combined to achieve higher-precision social network user alignment.
Drawings
FIG. 1 is a flow chart of a method for user alignment between social networks based on neural networks of the present invention.
Detailed Description
Referring to fig. 1, a method for user alignment between social networks based on a neural network according to the present invention includes the following steps:
step 1: sampling each user node of the social network to generate a random walk sequence of the user nodes;
step 2: representing the network nodes in the random walk sequence as user vectors by using a Word2vec tool;
and step 3: adopting a deep learning neural network to construct an alignment model, training by taking the aligned user vector as input, and optimizing model parameters to obtain the alignment model;
and 4, step 4: and inputting user vectors corresponding to all user nodes in the first social network into the alignment model, outputting the representation of the user vectors of the user nodes in the second social network, searching the user vectors which are most similar to the output of the alignment model in the second social network respectively, and judging whether the users corresponding to the user nodes are the same user or not according to the similarity.
Specifically, in step 1, for each user in the social network, taking the user node corresponding to the user as a starting point, randomly accessing the neighbor node of the user node as a next hop, then randomly accessing the neighbor node again with the next hop as the starting point, until a fixed length sequence length is 2 × w according to the user average degree w in the network, and then, in this embodiment, repeating the above process 10 times to obtain a random walk sequence of the user node.
Specifically, in step 2, a random walk sequence of user nodes is trained based on a Skip-Gram model, after the training of the model is completed, a weight matrix learned by the training of the model is obtained, all the user nodes are encoded into a one-hot form and input and mapped into the weight matrix, the co-occurrence probability between the nodes in a window in the sequence is maximized, namely the conditional probability of generating adjacent nodes by a given central node is calculated by using gradient descent and the probability is maximized, a mapping table of the nodes and embedded vectors is obtained, and after the training is completed, all the nodes in different social networks are respectively represented as user vectors in low-dimensional spaces through table lookup, so that the user nodes which are closely connected in the social networks are closer in distance in the embedded space, and the nodes which are weaker in distance.
Specifically, in step 3, two pieces of social network data are available, wherein part of nodes are known to be the same user, that is, part of aligned users are known, the part of known users is segmented, part of the known users is used as training data, part of the known users is used as test data, the aligned user vectors in the two social networks are divided into a test set and a training set, an alignment model is constructed based on a Seq2Seq model added with an Attention mechanism, dimension reduction is performed through the Attention mechanism, key information influencing output, that is, dimension data information of the model is learned, and each dimension corresponds to a weight value needing weighting and is used for further improving alignment accuracy; the alignment model is used for mapping the user vector in one social network to another social network, training the alignment model by taking the user vector in the training set as input, and learning the mapping relation between the two social networks by supervised learning; meanwhile, a Dropout method is adopted, part of neurons are abandoned randomly, overfitting of the alignment model is prevented, the generalization capability of the alignment model is enabled to be stronger, the alignment model outputs vectors with the same dimensionality as the input user vectors, when the overall loss function value of the alignment model tends to be stable and does not change, the model tends to be stable, training is stopped, and whether the model is effective is judged by testing the effect of the alignment model through data of a test set;
specifically, the loss function is calculated by the following formula:
where V represents a user vector, cos (V (u)i))，v(uj) Cosine values between two vectors, the smaller the loss function is, the better the alignment model is represented, and according to the size of the loss function, the corresponding weight is adjusted, so that the same user can be similar to the alignment model as much as possible.
Specifically, in step 4, the user vector in the first social network is input into the alignment model and then output, judging whether the output user vector is the same user or not through the cosine similarity of the output user vector and the user vector in the second social network, calculating the cosine similarity through the user vector in the second social network, if the value is larger than the judgment value under the condition of taking the maximum value of the cosine value, the user corresponding to the user vector of the second social network under the condition of the maximum cosine value is considered as the same user of the input users corresponding to the user vector of the first social network, in step 4, the user vector in the first social network obtains an output vector through the alignment model, and then finding a user vector in the second social network that is most similar to the output vector, so that the user in the second social network and the user in the first social network are considered to be the same user.
The invention relates to a method for aligning users among social networks based on a neural network, which maintains the original structure of the social networks to the maximum extent by constructing a random walk sequence of user nodes, then constructs an alignment model by adding a Seq2Seq model of an Attention mechanism, learns the mapping relation in the two social networks by the alignment model, then inputs user vectors of all the user nodes in the social networks into the alignment model, searches the most similar node in the corresponding social networks by utilizing cosine value similarity according to the output vector of the alignment model, judges whether the two nodes are the same user or not according to the similarity, compared with the prior art, the technical scheme provided by the invention does not need to manually extract features in the whole alignment process, can effectively retain local and whole network structure information, improves the link precision in different index universes compared with other user alignment methods, the average link precision is improved by 7%, meanwhile, the aligned users can be locked in a smaller range, network structure features are automatically extracted under the condition that only social network structure data exist, hidden information of the social network structure is mined, and meanwhile, local information and global information of the network structure are combined, so that the social network users can be aligned with higher precision.
In an embodiment of the present invention, there is also provided an apparatus for user alignment between social networks based on a neural network, including: comprising a processor, a memory, and a program;
a program is stored in the memory and the processor invokes the program stored in the memory to perform the method for user alignment between social networks based on neural networks described above.
In the implementation of the above-described device for user alignment between social networks based on neural networks, the memory and the processor are electrically connected directly or indirectly to enable transmission or interaction of data. For example, the components may be electrically connected to each other via one or more communication buses or signal lines, such as a bus. The memory stores computer-executable instructions for implementing the data access control method, and includes at least one software functional module which can be stored in the memory in the form of software or firmware, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory.
The Memory may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The memory is used for storing programs, and the processor executes the programs after receiving the execution instructions.
The processor may be an integrated circuit chip having signal processing capabilities. The processor may be a general-purpose processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
In an embodiment of the present invention, there is also provided a computer-readable storage medium, wherein the computer-readable storage medium is configured to store a program configured to execute the above-mentioned method for user alignment between social networks based on a neural network.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart.
These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart.
The method for aligning users among social networks based on a neural network, the device for aligning users among social networks based on a neural network, and the application of a computer-readable storage medium provided by the present invention are described in detail above, specific examples are applied herein to illustrate the principles and embodiments of the present invention, and the description of the above embodiments is only used to help understand the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (9)
1. A method for user alignment between social networks based on neural networks, comprising the steps of:
step 1: sampling each user node of the social network to generate a random walk sequence of the user nodes;
step 2: representing the network nodes in the random walk sequence as user vectors by using a Word2vec tool;
and step 3: adopting a deep learning neural network to construct an alignment model, training by taking the aligned user vector as input, and optimizing model parameters to obtain the alignment model;
and 4, step 4: and inputting user vectors corresponding to all user nodes in the first social network into the alignment model, outputting the representation of the user vectors of the user nodes in the second social network, searching the user vectors which are most similar to the output of the alignment model in the second social network respectively, and judging whether the users corresponding to the user nodes are the same user or not according to the similarity.
2. The method of claim 1, wherein the method comprises: in step 1, for each user in the social network, taking a user node corresponding to the user as a starting point, randomly accessing a neighbor node of the user node as a next hop, then randomly accessing the neighbor node again with the next hop as the starting point, according to the user average degree w in the network, until a fixed-length sequence length is 2 x w, and then repeating the above process for N times, wherein N is a natural number, to obtain a random walk sequence of the user node.
3. The method of claim 1, wherein the method comprises: in step 2, training a random walk sequence of user nodes based on a Skip-Gram model, obtaining a weight matrix learned by model training after the model training is finished, coding all the user nodes into a one-hot form, inputting and mapping the one-hot form into the weight matrix, calculating the conditional probability of a given central node for generating adjacent nodes by using gradient descent and maximizing the probability to obtain a mapping table of the nodes and embedded vectors, and respectively representing all the nodes in different social networks as user vectors of a low-dimensional space through table lookup after the training is finished.
4. The method of claim 1, wherein the method comprises: the step 3 specifically comprises the following steps: dividing user vectors corresponding to aligned user nodes in two social networks into a test set and a training set, constructing an alignment model based on a Seq2Seq model added with an Attention mechanism, wherein the alignment model is used for mapping the user vectors in one social network to the other social network, taking the user vectors in the training set as an input training alignment model, performing supervised learning, learning the mapping relationship in the two social networks, outputting vectors with the same dimensionality as the input user vectors by the alignment model, indicating that the model tends to be stable when the overall loss function value of the alignment model tends to be stable and does not change, stopping training, and testing the effect of the alignment model through the data of the test set to judge whether the model is effective.
5. The method of claim 4, wherein the social network user alignment based on the neural network is performed by: in step 3, Dropout method is used to discard part of neurons randomly for preventing the alignment model from overfitting.
6. The method of claim 4, wherein the social network user alignment based on the neural network is performed by: in step 3, the loss function is calculated by the following formula:
l(v(ui)，v(uj))＝min(1-cos(Φ(v(ui))，v(uj)))
where V represents a user vector, cos (V (u)i))，v(uj) Cosine values between two vectors, the smaller the loss function is, the better the alignment model is represented, and according to the size of the loss function, the corresponding weight is adjusted, so that the same user can be similar to the alignment model as much as possible.
7. The method of claim 1, wherein the method comprises: in step 4, the user vector in the first social network is input into the alignment model and then output, whether the output user vector is the same user is judged through cosine similarity between the output user vector and the user vector in the second social network, the cosine similarity is calculated through the user vector in the second social network, and if the value of the output user vector is greater than the judgment value under the condition of taking the maximum value of the cosine value, the user corresponding to the user vector in the second social network under the condition of the maximum value of the cosine value is considered to be the same user of the input user corresponding to the user vector in the first social network.
8. An apparatus for user alignment between social networks based on neural networks, comprising: comprising a processor, a memory, and a program;
the program is stored in the memory and the processor invokes the memory-stored program to perform the method for user alignment between neural network-based social networks of claim 1.
9. A computer-readable storage medium characterized by: the computer readable storage medium is configured to store a program configured to perform the method of user alignment between neural network-based social networks of claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010255397.4A CN111476673A (en) | 2020-04-02 | 2020-04-02 | Method, device and medium for aligning users among social networks based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010255397.4A CN111476673A (en) | 2020-04-02 | 2020-04-02 | Method, device and medium for aligning users among social networks based on neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111476673A true CN111476673A (en) | 2020-07-31 |
Family
ID=71749597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010255397.4A Pending CN111476673A (en) | 2020-04-02 | 2020-04-02 | Method, device and medium for aligning users among social networks based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111476673A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112036445A (en) * | 2020-08-06 | 2020-12-04 | 中国人民解放军战略支援部队信息工程大学 | Cross-social-network user identity recognition method based on neural tensor network |
| CN112084373A (en) * | 2020-08-05 | 2020-12-15 | 国家计算机网络与信息安全管理中心 | Multi-source heterogeneous network user alignment method based on graph embedding |
| CN112507247A (en) * | 2020-12-15 | 2021-03-16 | 重庆邮电大学 | Cross-social network user alignment method fusing user state information |
| CN112508724A (en) * | 2020-10-21 | 2021-03-16 | 上海交通大学 | User identity alignment method based on cycle consistency counterstudy |
| CN112686766A (en) * | 2020-12-26 | 2021-04-20 | 中山大学 | Embedded representation method, device, equipment and storage medium of social network |
| CN117056239A (en) * | 2023-10-11 | 2023-11-14 | 腾讯科技（深圳）有限公司 | Method, device, equipment and storage medium for determining test function using characteristics |
| CN117495481A (en) * | 2023-10-25 | 2024-02-02 | 宁夏大学 | Article recommendation method based on heterogeneous timing diagram attention network |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107392782A (en) * | 2017-06-29 | 2017-11-24 | 上海斐讯数据通信技术有限公司 | Corporations' construction method, device and computer-processing equipment based on word2Vec |
| CN108959418A (en) * | 2018-06-06 | 2018-12-07 | 中国人民解放军国防科技大学 | Character relation extraction method and device, computer device and computer readable storage medium |
| CN109635989A (en) * | 2018-08-30 | 2019-04-16 | 电子科技大学 | A kind of social networks link prediction method based on multi-source heterogeneous data fusion |
| CN109636658A (en) * | 2019-01-17 | 2019-04-16 | 电子科技大学 | A kind of social networks alignment schemes based on picture scroll product |
| CN109697288A (en) * | 2018-12-25 | 2019-04-30 | 北京理工大学 | A kind of example alignment schemes based on deep learning |
| CN110489567A (en) * | 2019-08-26 | 2019-11-22 | 重庆邮电大学 | A kind of node information acquisition method and its device based on across a network Feature Mapping |
| CN110532436A (en) * | 2019-07-17 | 2019-12-03 | 中国人民解放军战略支援部队信息工程大学 | Across social network user personal identification method based on community structure |
-
2020
- 2020-04-02 CN CN202010255397.4A patent/CN111476673A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107392782A (en) * | 2017-06-29 | 2017-11-24 | 上海斐讯数据通信技术有限公司 | Corporations' construction method, device and computer-processing equipment based on word2Vec |
| CN108959418A (en) * | 2018-06-06 | 2018-12-07 | 中国人民解放军国防科技大学 | Character relation extraction method and device, computer device and computer readable storage medium |
| CN109635989A (en) * | 2018-08-30 | 2019-04-16 | 电子科技大学 | A kind of social networks link prediction method based on multi-source heterogeneous data fusion |
| CN109697288A (en) * | 2018-12-25 | 2019-04-30 | 北京理工大学 | A kind of example alignment schemes based on deep learning |
| CN109636658A (en) * | 2019-01-17 | 2019-04-16 | 电子科技大学 | A kind of social networks alignment schemes based on picture scroll product |
| CN110532436A (en) * | 2019-07-17 | 2019-12-03 | 中国人民解放军战略支援部队信息工程大学 | Across social network user personal identification method based on community structure |
| CN110489567A (en) * | 2019-08-26 | 2019-11-22 | 重庆邮电大学 | A kind of node information acquisition method and its device based on across a network Feature Mapping |
Non-Patent Citations (4)
| Title |
|---|
| 杨奕卓;于洪涛;黄瑞阳;刘正铭;: "基于融合表示学习的跨社交网络用户身份匹配", no. 09 * |
| 裴杨;瞿学鑫;郭晓博;段丁阳;: "基于node2vec的社交网络用户属性补全攻击", no. 12 * |
| 解书颖: "基于节点表示的跨网络节点关联研究", no. 2018, pages 139 - 79 * |
| 高扬: "人工智能与机器人先进技术丛书 智能摘要与深度学习", 北京理工大学出版社, pages: 120 - 122 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112084373B (en) * | 2020-08-05 | 2022-06-03 | 国家计算机网络与信息安全管理中心 | Graph embedding-based multi-source heterogeneous network user alignment method |
| CN112084373A (en) * | 2020-08-05 | 2020-12-15 | 国家计算机网络与信息安全管理中心 | Multi-source heterogeneous network user alignment method based on graph embedding |
| CN112036445B (en) * | 2020-08-06 | 2023-05-26 | 中国人民解放军战略支援部队信息工程大学 | User identity recognition method across social networks based on neural tensor network |
| CN112036445A (en) * | 2020-08-06 | 2020-12-04 | 中国人民解放军战略支援部队信息工程大学 | Cross-social-network user identity recognition method based on neural tensor network |
| CN112508724A (en) * | 2020-10-21 | 2021-03-16 | 上海交通大学 | User identity alignment method based on cycle consistency counterstudy |
| CN112508724B (en) * | 2020-10-21 | 2022-03-11 | 上海交通大学 | User identity alignment method based on cycle consistency counterstudy |
| CN112507247B (en) * | 2020-12-15 | 2022-09-23 | 重庆邮电大学 | Cross-social network user alignment method fusing user state information |
| CN112507247A (en) * | 2020-12-15 | 2021-03-16 | 重庆邮电大学 | Cross-social network user alignment method fusing user state information |
| CN112686766A (en) * | 2020-12-26 | 2021-04-20 | 中山大学 | Embedded representation method, device, equipment and storage medium of social network |
| CN112686766B (en) * | 2020-12-26 | 2024-05-31 | 中山大学 | Method, device, equipment and storage medium for embedded representation of social network |
| CN117056239A (en) * | 2023-10-11 | 2023-11-14 | 腾讯科技（深圳）有限公司 | Method, device, equipment and storage medium for determining test function using characteristics |
| CN117056239B (en) * | 2023-10-11 | 2024-01-30 | 腾讯科技（深圳）有限公司 | Method, device, equipment and storage medium for determining test function using characteristics |
| CN117495481A (en) * | 2023-10-25 | 2024-02-02 | 宁夏大学 | Article recommendation method based on heterogeneous timing diagram attention network |
| CN117495481B (en) * | 2023-10-25 | 2024-06-11 | 宁夏大学 | Article recommendation method based on heterogeneous timing diagram attention network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111476673A (en) | Method, device and medium for aligning users among social networks based on neural network | |
| CN110837602B (en) | User recommendation method based on representation learning and multi-mode convolutional neural network | |
| CN104598611B (en) | The method and system being ranked up to search entry | |
| CN114330312B (en) | Title text processing method, title text processing device, title text processing program, and recording medium | |
| CN109753602B (en) | Cross-social network user identity recognition method and system based on machine learning | |
| CN112633973B (en) | Commodity recommendation method and related equipment thereof | |
| CN111859986A (en) | Semantic matching method, device, equipment and medium based on multitask twin network | |
| CN110009486B (en) | Method, system, equipment and computer readable storage medium for fraud detection | |
| CN108171059A (en) | Malicious Code Detection and recognition methods and device based on software gene | |
| CN112085091B (en) | Short text matching method, device, equipment and storage medium based on artificial intelligence | |
| CN116091152A (en) | Recommendation method and system based on multi-level comparison learning and multi-mode knowledge graph | |
| CN111241298A (en) | Information processing method, apparatus and computer readable storage medium | |
| CN113821587A (en) | Text relevance determination method, model training method, device and storage medium | |
| CN111460783A (en) | Data processing method and device, computer equipment and storage medium | |
| CN114726823B (en) | Domain name generation method, device and equipment based on generation countermeasure network | |
| CN113420212A (en) | Deep feature learning-based recommendation method, device, equipment and storage medium | |
| Zhou et al. | Learning to correlate accounts across online social networks: An embedding-based approach | |
| CN112966095B (en) | Software code recommendation method based on JEAN | |
| CN114330704A (en) | Statement generation model updating method and device, computer equipment and storage medium | |
| CN117390480A (en) | Information extraction method, device, equipment and storage medium | |
| CN113591881A (en) | Intention recognition method and device based on model fusion, electronic equipment and medium | |
| CN111325578B (en) | Sample determination method and device of prediction model, medium and equipment | |
| CN113204643A (en) | Entity alignment method, device, equipment and medium | |
| CN116957128A (en) | Service index prediction method, device, equipment and storage medium | |
| CN113918696A (en) | Question-answer matching method, device, equipment and medium based on K-means clustering algorithm |
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 |