CN110781407B - User tag generation method, device and computer readable storage medium - Google Patents
User tag generation method, device and computer readable storage medium Download PDFInfo
- Publication number
- CN110781407B CN110781407B CN201911000656.2A CN201911000656A CN110781407B CN 110781407 B CN110781407 B CN 110781407B CN 201911000656 A CN201911000656 A CN 201911000656A CN 110781407 B CN110781407 B CN 110781407B
- Authority
- CN
- China
- Prior art keywords
- user
- feature vector
- text
- texts
- social
- 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 46
- 238000003860 storage Methods 0.000 title claims abstract description 11
- 239000013598 vector Substances 0.000 claims abstract description 240
- 230000006399 behavior Effects 0.000 claims abstract description 65
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 26
- 230000015654 memory Effects 0.000 claims abstract description 22
- 238000005070 sampling Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 7
- 230000001939 inductive effect Effects 0.000 abstract description 9
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 12
- 238000013528 artificial neural network Methods 0.000 description 10
- 230000004927 fusion Effects 0.000 description 10
- 238000012549 training Methods 0.000 description 7
- 230000007246 mechanism Effects 0.000 description 6
- 230000004913 activation Effects 0.000 description 5
- 238000001994 activation Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000003542 behavioural effect Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000011176 pooling Methods 0.000 description 3
- 235000008694 Humulus lupulus Nutrition 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000002457 bidirectional effect Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 230000003997 social interaction Effects 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
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield 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
- G06N3/04—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
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Databases & Information Systems (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- Business, Economics & Management (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the application discloses a user tag generation method, a device and a computer readable storage medium, wherein the method comprises the following steps: acquiring a behavior text sequence of a user, and randomly initializing the behavior text sequence by using a two-way long-short-term memory network BiLSTM to obtain a first feature vector, wherein the behavior text sequence is a sequence formed by texts which are generated according to time sequence and are used for representing the behavior of the user; acquiring social network information of the user, and processing the social network information according to a inductive learning algorithm to obtain a second feature vector; acquiring social statistical information of the user, and acquiring a third feature vector according to the social statistical information; and transversely splicing the first feature vector, the second feature vector and the third feature vector to obtain the user tag of the user. The embodiment of the application is beneficial to improving the precision of generating the user label.
Description
Technical Field
The present application relates to the field of artificial intelligence technology, and in particular, to a method and apparatus for generating a user tag, and a computer readable storage medium.
Background
At present, the multi-mode information fusion technology is a hotspot for research in the field of artificial intelligence, for example, in the production process of user tags, data of different modes such as images, texts, social networks, geographic positions and social statistics information are often required to be organically fused together, so that user behavior characteristics are constructed in a multi-angle and omnibearing manner, accurate user tags are given, accurate predicted values are given, and product operation, personalized recommendation and accurate advertisement delivery are better served.
In addition, social network information plays a role in user labels, and is particularly important for users with sparse features, for example, if friends of the users are mostly in the building industry or the users join multiple discussion groups in the building industry, even if other information of the users is very sparse, we have strong reasons to believe that the users belong to the building industry.
At present, when social network information and information of other modes are fused, a linear splicing method can be adopted, but for a very large-scale social network scene, the social relationship of each user is complicated, the social network vector of each user cannot be generated, in addition, because the contribution degree of different mode information to the user label of the user is different, only simple splicing is carried out, and the precision of the generated user label is low.
Disclosure of Invention
The embodiment of the application provides a user tag generation method, a device and a computer readable storage medium, which are used for generating an accurate user tag by fusing multi-mode data.
In a first aspect, an embodiment of the present application provides a method for generating a user tag, including:
acquiring a behavior text sequence of a user, and randomly initializing the behavior text sequence by using a two-way long-short-term memory network BiLSTM to obtain a first feature vector, wherein the behavior text sequence is a sequence formed by texts which are generated according to time sequence and are used for representing the behavior of the user;
Acquiring social network information of the user, and processing the social network information according to a inductive learning algorithm to obtain a second feature vector;
Acquiring social statistical information of the user, and acquiring a third feature vector according to the social statistical information;
and transversely splicing the first feature vector, the second feature vector and the third feature vector to obtain the user tag of the user.
In a second aspect, an embodiment of the present application provides a user tag generating apparatus, where the user tag generating apparatus includes a processor, a transceiver, and at least one circuit, and the processor and the transceiver are connected through the at least one circuit;
The transceiver is used for acquiring a behavior text sequence, social network information and social statistical information of a user and sending the behavior text sequence, the social network information and the social statistical information to the processor, wherein the behavior text sequence is a sequence formed by texts which are generated according to time sequence and are used for representing the behavior of the user;
the processor is configured to randomly initialize the behavioral text sequence using a bidirectional long-short term memory network BiLSTM to obtain a first feature vector;
the processor is further used for processing the social network information according to a generalized learning algorithm to obtain a second feature vector;
the processor is further used for obtaining a third feature vector according to the social statistical information;
And the processor is further used for transversely splicing the first feature vector, the second feature vector and the third feature vector to obtain the user tag of the user.
In a third aspect, an embodiment of the present application provides an electronic device comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, the programs comprising instructions for performing the steps in the method according to the first aspect.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium storing a computer program that causes a computer to perform the method according to the first aspect.
In a fifth aspect, embodiments of the present application provide a computer program product comprising a non-transitory computer readable storage medium storing a computer program, the computer being operable to cause a computer to perform the method according to the first aspect.
The embodiment of the application has the following beneficial effects:
it can be seen that in the embodiment of the application, the feature vectors corresponding to the social network information in the social mode and the data information in other modes are spliced to obtain the user splice, and the social network information contains rich user features (such as industry information) because the user labels are fused with the social network information in the social mode, so that the accuracy of the user labels is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of multi-modal data fusion according to an embodiment of the present application;
FIG. 2 is a schematic diagram of another multi-modal data fusion provided by an embodiment of the present application;
Fig. 3A is a schematic diagram of a scenario for generating a user tag according to an embodiment of the present application;
fig. 3B is a flowchart of a method for generating a user tag according to an embodiment of the present application;
FIG. 3C is a schematic diagram of a processing behavior sequence based on an attention mechanism according to an embodiment of the present application;
FIG. 3D is a schematic diagram of a social graph network according to an embodiment of the present application;
FIG. 3E is a schematic diagram of a social graph network processing system according to an embodiment of the present application;
Fig. 4 is a flowchart of another method for generating a user tag according to an embodiment of the present application;
FIG. 5 is a schematic diagram of a model for generating user labels according to an embodiment of the present application;
Fig. 6 is a schematic structural diagram of a user tag generating apparatus according to an embodiment of the present application;
fig. 7 is a functional unit composition block diagram of a user tag generating apparatus according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The terms "first," "second," "third," and "fourth" and the like in the description and in the claims and drawings are used for distinguishing between different objects and not necessarily for describing a particular sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus.
Data compliance declaration: the collection and processing of the related data in the application should strictly meet the requirements of relevant national laws and regulations when the example is applied, have legal basis or acquire the informed consent or independent consent of the personal information body, and develop the subsequent data use and processing behaviors within the authorized range of laws and regulations and/or the personal information body.
In the process of generating the user tag, data (such as images, texts, social network information, geographic positions, social statistics information and the like) of different modes are required to be fused, so that the user behavior characteristics are constructed in a multi-angle and all-around mode, and then an accurate user tag is given.
Referring to fig. 1, fig. 1 provides a deep neural network model (Deep Neural Networks, DNN) for merging user browsing video sequences, geographic locations and user social statistics, specifically, after weighting a plurality of browsing vectors and Search behavior sequence vectors of a user, respectively, a video browsing Vector Watch Vector and a Search Vector are generated; and then splicing the Watch Vector and the Search Vector with social statistical information such as address position information, age gender and the like to obtain a target feature Vector, feeding the target feature Vector into a forward neural network, and obtaining a user label of the user through a series of subsequent processes (such as nonlinear activation of a ReLU).
Although the DNN network realizes the fusion of different modal data such as text, images, geographical position information and the like, a fusion scheme for fusing social network information of a user is lacking; in addition, social network information plays a role in generating user labels, particularly important for users with sparse features, a multimodal fusion scheme based on a graph convolution neural network is provided, and as shown in fig. 2, different nodes represent different users, different users are connected in an edge mode to represent social relations of each user, a social graph network is obtained, data (such as a behavior text sequence and social statistics information) of other modes are added to the social graph network as node features, training is performed by using the graph network, and when the users contain a plurality of mode data, simple linear splicing is performed on the node features. When the social network diagram is adopted for training, signal transmission exists among the neighbor nodes of each node, so that social network information of a user is added into the training process, but the splicing method of node characteristics is simple, is not friendly to complex modal data (such as image characteristics), and reduces the expandability of multi-modal data fusion; in addition, when the node characteristics are spliced, the contribution degree of each mode data to the user label is not considered, so that the precision of the spliced node characteristics is low; in addition, when the social network of the user is complex, it is difficult to acquire signals of other nodes to the node, and the feature vector corresponding to the node cannot be obtained.
The scheme of the application is provided for solving the problems in the multi-mode data fusion.
Referring to fig. 3A, fig. 3A is a schematic view of a scenario for generating a user tag according to an embodiment of the present application, including a user terminal 110 and a tag generating device 120;
the user performs network social interaction, online entertainment, and the like through the user terminal 110, industry information (a behavior text sequence) affiliated to the user, social network information of the user, social statistics information (such as age, gender, school, and the like) of the user can be collected from the user terminal 110, the collected behavior text sequence, social network information, and social statistics information are sent to the tag generation device 120, the tag generation device 120 invokes a network model matched with each information to process each information, a first feature vector corresponding to the behavior text sequence is obtained, a second feature vector corresponding to the social network information, and a third feature vector corresponding to the social statistics information are obtained, finally, the first feature vector, the second feature vector, and the third feature vector are spliced to obtain a user tag of the user, and targeted recommendation is performed on the user based on the user tag.
Referring to fig. 3B, fig. 3B is a method for generating a user tag according to an embodiment of the present application, including but not limited to the following steps:
301: and acquiring a behavior text sequence of the user, and randomly initializing the behavior text sequence by using a two-way long-short-term memory network to obtain a first feature vector.
The behavior text sequence is a sequence formed by texts which are generated in time sequence and used for representing the behavior of the user.
Specifically, the behavior text sequence comprises N texts, wherein the N texts are used for representing the behaviors of the user at N moments, each moment corresponds to one text, and the texts used for the behaviors of the user at N different moments are combined according to the sequence of time to obtain the behavior text sequence.
And then, vector encoding each text by using a Bi-directional Short-Term Memory (BiLSTM) network to obtain a feature vector of each text, and weighting the feature vector of each text to obtain the first feature vector.
302: And acquiring social network information of the user, and processing the social network information according to a inductive learning algorithm to obtain a second feature vector.
Wherein the social network information includes: industry information, social object information, and so forth.
Wherein the inductive learning may be GRAPHSAGE algorithm.
Optionally, a social graph network with the user is established according to the social network information, and the node where the user is located in the social graph network is processed according to GRAPHSAGE algorithm to obtain a second feature vector of the user.
303: And acquiring social statistical information of the user, and acquiring a third feature vector according to the social statistical information.
Optionally, the social statistics includes P-item information with the user, including but not limited to: age, gender, academic, home address, job site; and carrying out vector coding on each item of information to obtain sub-vectors corresponding to each item of information, so as to obtain P sub-vectors, and then, carrying out coding on the P sub-vectors, so as to obtain a third feature vector.
304: And transversely splicing the first feature vector, the second feature vector and the third feature vector to obtain the user tag of the user.
And performing dimension splicing, namely transverse splicing, on the first feature vector, the second feature vector and the third feature vector to obtain a target feature vector, performing nonlinear activation on the target feature vector, inputting the target feature vector after nonlinear activation into a softmax classifier, and outputting a user label of the user.
The first feature vector x, the second feature vector y, and the third feature vector z are transversely spliced to obtain a target feature vector w= [ a, b, c, e, f, g, h, i, j ]. It should be noted that, the transverse splicing mentioned later is similar to this, and the transverse splicing will not be described in detail.
It can be seen that in this embodiment, the data information in each mode is processed separately, and in the processor, the behavior text is processed based on the attention mechanism, so that the first feature vector can reflect the industry to which the user belongs, and further the accuracy of the user tag is improved; in addition, social network information is processed based on GRAPHSAGE algorithm, so that when new social network information is added, the second feature vector of the node can be directly obtained without model training, and further calculation cost is reduced.
In some possible embodiments, the implementation process of randomly initializing the behavior text sequence by using the two-way long-short term memory network to obtain the first feature vector may be: extracting keywords of each text in the N texts, randomly initializing the keywords of each text in the N texts by using BiLSTM networks to obtain N feature vectors, and determining the weight of each text according to the N feature vectors; and weighting the N eigenvectors according to the weight of each text to obtain a first eigenvector.
Wherein, the keyword is randomly initialized to the prior art and is not described.
When a key word of each text in the N texts is randomly initialized by using a BiLSTM network, specifically, a key word corresponding to each text is randomly initialized by using a forward LSTM of the BiLSTM network to obtain a forward feature vector corresponding to each text, and simultaneously, a key word corresponding to each text is randomly initialized by using a backward LSTM of BiLSTM to obtain a backward feature vector corresponding to each text, and the forward feature vector and the backward feature vector corresponding to each text in the N texts are transversely spliced to obtain N feature vectors.
For example, if a keyword corresponding to a certain text is "chinese", forward LSTM is used to randomly initialize the keyword "chinese" to obtain forward feature vectors h1= [ s1, s2, s3], backward LSTM is synchronously used to randomly initialize the keyword "chinese" to obtain backward feature vectors h1= [ m1, m2, m3], and then h1 and h2 are transversely spliced to obtain feature vectors p= [ s1, s2, s3, m1, m2, m3] of the keyword, thereby obtaining feature vectors of the text.
The above shows that a text includes a keyword, and for the case that the text includes a plurality of keywords, according to the random initialization mode, a feature vector p corresponding to each keyword is obtained, and then a plurality of feature vectors corresponding to the keywords are transversely spliced to obtain a feature vector corresponding to the text.
It can be seen that in the embodiment, the BiLSTM network is adopted to initialize the behavior text, and the past and future information of the current state is obtained by utilizing the positive and negative time sequence directions in the text, so that the semantic features are enriched, and the feature vector obtained after the initialization is more in line with the behavior of the user; and the weight of each text is obtained first, the weight reflects the contribution degree of the text to the industry of the user, and the first feature vector is further improved to represent the accuracy of the industry of the user by carrying out weighting processing according to the weight.
Optionally, before extracting the keyword from each text in the N texts, denoising the N texts according to a pre-constructed industry dictionary, and filtering text irrelevant to the industry in the N texts to obtain T texts; randomly initializing keywords of each behavior text in T texts by using BiLSTM networks to obtain T feature vectors; determining the weight of each text according to the T feature vectors; and weighting the T feature vectors according to the weight of each text to obtain the first feature vector.
Wherein, the keyword of each behavior text in T texts is randomly initialized by using BiLSTM networks, which is consistent with the random initialization process and is not described,
It can be seen that in this embodiment, the behavioral text sequence is firstly denoised, so as to filter irrelevant text, avoid processing invalid data, and further improve the generation efficiency of the user tag; and the weight of each text is obtained, the weight reflects the contribution degree of the text to the industry of the user, and the accuracy of the first feature vector is further improved by carrying out weighting processing according to the weight.
The process of calculating the weight of each text and weighting a plurality of feature vectors, where the plurality of feature vectors are T feature vectors of T texts after denoising, is described in detail below with reference to fig. 3C.
After the feature vector of each text is obtained, as shown in fig. 3C, weights corresponding to each text are obtained according to the T feature vectors based on the attention mechanism Attention mechanism, and then weighting processing is performed on the T feature vectors based on the weights corresponding to each text, so as to obtain a first feature vector.
Wherein the first eigenvector can be obtained by the formula (1).
et＝tanh(W*ht+bt)
H t is a feature vector of a T text in the T texts, W is a weight matrix, that is, w= (W 1,w2,…,wT),bt is a bias corresponding to the feature vector, α t is a weight corresponding to the T text, and C is a first feature vector.
In some possible embodiments, the generalized learning algorithm may be graphsage algorithm, and the processing the social network information according to the generalized learning algorithm may be performed to obtain the second feature vector, where the implementation process may be: and processing the social network information according to the graphsage algorithm to obtain a second feature vector. Namely, generating a social graph network according to the social network information; sampling neighbor nodes of a target node according to a preset sampling duty ratio to obtain R neighbor nodes, wherein the target node is a node corresponding to the user in the social graph network, and the sampling duty ratio is the duty ratio of the neighbor nodes for sampling to all the neighbor nodes; weighting the feature vectors of the R neighbor nodes to obtain a fourth feature vector, wherein the feature vectors of the R neighbor nodes are obtained according to the feature vectors of the neighbor nodes corresponding to the search depth; and splicing the characteristic vector of the target node and the fourth characteristic vector to obtain the second characteristic vector.
The feature vectors of the R neighbor nodes are weighted, and the R neighbor nodes may be aggregated by a pooling function, where the pooling function may include pooling, mean, and the like.
It can be seen that in this embodiment, the social network information of each user is fused by using GRAPHSAGE algorithm, and when a node is newly added in the social graph network, the feature vector of the node can be directly generated, without retraining the whole model, so that the operation cost during fusion is reduced, and the generalization performance during fusion is improved.
The processing of the GRAPHSAGE algorithm is described in detail below in conjunction with fig. 3D and 3E.
Referring to fig. 3D, fig. 3D is a social graph network of user a, user B, user C, user D, user E, and user F, where node a is a target node, and it is assumed that the sampling duty ratio is 1, that is, all neighboring nodes participate in the calculation of the feature vector of node a, as shown in fig. 3E, the feature vectors of node B, node C, and node D are calculatedAndWeighting to obtain a fourth feature vectorThen, toFeature vector of node aSplicing to obtain a second feature vector of the node AWherein,AndMay be calculated from feature vectors of neighboring nodes to node B, node C and node D.
The preset search depth determines the adjacent depth with the target node. For example, when the search depth is 1, only the feature vector of the neighboring node is needed to calculate the second feature vector, and when the search depth is 2, as shown in fig. 3E, the feature vector of each neighboring node is calculated first by the feature vector of the neighboring node, and then the feature vector of the target node is calculated again using the feature vector of the neighboring node.
In some possible embodiments, after obtaining the user tag of the user, the method further includes: and personalized recommendation is carried out on the user by using the user tag of the user.
Referring to fig. 4, fig. 4 is another method for generating a user tag according to an embodiment of the present application, which is the same as the embodiment shown in fig. 3B, and will not be repeated here, and includes, but is not limited to, the following steps:
401: and acquiring a behavior text sequence of the user, and processing the behavior text sequence by using the random initialization of the two-way long-short-term memory network BiLSTM to obtain a first feature vector.
402: And acquiring social network information of the user, and processing the social network information according to a inductive learning algorithm to obtain a second feature vector.
403: And acquiring social statistical information of the user, and acquiring a third feature vector according to the social statistical information.
404: And transversely splicing the first feature vector, the second feature vector and the third feature vector to obtain the user tag of the user.
405: And carrying out personalized recommendation on the user according to the label of the user.
The personalized recommendation may be an advertisement recommendation, a video recommendation, a news recommendation, a music recommendation, etc.
It can be seen that in this embodiment, the data information in each mode is processed separately, and in the processor, the behavior text is processed based on the attention mechanism, so that the first feature vector can reflect the industry to which the user belongs, and further the accuracy of the user tag is improved; in addition, social network information is processed based on GRAPHSAGE algorithm, so that when new social network information is added, a second feature vector of the node can be directly obtained, model training is not required to be carried out again, and calculation cost is further reduced; because the generated user tag is accurate, the personalized recommendation better meets the actual requirements of the user.
It should be noted that, the specific implementation of each step of the method shown in fig. 4 may be referred to the specific implementation of the method shown in fig. 3B, which is not described herein.
In some possible implementations, the user tag generation method provided by the embodiment of the present application is applied to a tag generation model shown in fig. 5, where the tag generation model includes a first neural network, a second neural network, and a third neural network.
The first neural network is a bidirectional BiLSTM network based on an attention mechanism and is used for processing an input behavior text sequence to obtain a first feature vector (weighted text vector);
The second neural network is a network trained based on GRAPHSAGE algorithm and is used for processing the input social network information to obtain a second feature vector (social network vecto), and specifically, the network is obtained by performing unsupervised optimization by adopting a loss function in a formula (2):
J G(Zu) is the loss corresponding to destination node z u, z v is the neighbor node z u within the specified depth, σ is the sigmoid function, Q is the number of negative samples, P n is the distribution function of negative samples, v n is the negative sample set satisfying the distribution of P n, Is expected for all negative samples v n that obey the P n distribution.
It should be noted that, when training the second neural network, a training data set (social graph network) needs to be constructed, and when constructing the social graph network, only neighboring nodes with 2 hops or 3 hops of the neighboring depth of the target node a need to be sampled, so as to reduce the computing overhead as much as possible under the condition of not reducing the algorithm performance.
The third neural network is a concat network, and is used for connecting a plurality of sub-vectors in the social statistics dimension to obtain a third feature vector (userprofile vector).
Finally, performing dimension splicing on the first feature vector, the second feature vector and the third feature vector based on the label generation model to obtain a target feature vector, performing a series of nonlinear activations, and inputting the target feature vector after the nonlinear activations into a softmax classifier to obtain a label of the user.
Fig. 6 is a schematic structural diagram of a user tag generating apparatus according to an embodiment of the present application, where an apparatus 600 includes a processor, a memory, a communication interface, and one or more programs, where the one or more programs are different from the one or more application programs, and the one or more programs are stored in the memory and configured to be executed by the processor, and the programs include instructions for performing the following steps:
acquiring a behavior text sequence of a user, and randomly initializing the behavior text sequence by using a two-way long-short-term memory network BiLSTM to obtain a first feature vector, wherein the behavior text sequence is a sequence formed by texts which are generated according to time sequence and are used for representing the behavior of the user;
Acquiring social network information of the user, and processing the social network information according to a inductive learning algorithm to obtain a second feature vector;
Acquiring social statistical information of the user, and acquiring a third feature vector according to the social statistical information;
and transversely splicing the first feature vector, the second feature vector and the third feature vector to obtain the user tag of the user.
In some possible embodiments, the behavior text sequence includes N texts, where N texts are used to characterize the behavior of the user at N times, where N is an integer greater than 1, and before the random initialization of the behavior text sequence using the two-way long-short-term memory network BiLSTM, the above program is further configured to execute instructions for:
Denoising the N texts according to a pre-constructed industry dictionary to obtain T texts, wherein T is less than or equal to N, and T is an integer greater than or equal to 1;
The random initialization of the behavioral text sequence using the two-way long-short term memory network BiLSTM to obtain a first feature vector includes:
randomly initializing each text in the T texts by using the BiLSTM to obtain T feature vectors;
determining the weight of each text in the T texts according to the T feature vectors;
and weighting the T feature vectors according to the weight of each text to obtain the first feature vector.
In some possible embodiments, in initializing each text of the T texts randomly using the BiLSTM to obtain T feature vectors, the above program is specifically configured to execute instructions for:
extracting keywords corresponding to each text in the T texts;
Randomly initializing keywords corresponding to each text by using the forward LSTM of BiLSTM to obtain forward feature vectors corresponding to each text, and randomly initializing keywords corresponding to each text by using the backward LSTM of BiLSTM to obtain backward feature vectors corresponding to each text;
and transversely splicing the forward feature vector and the backward feature vector corresponding to each text in the T texts to obtain T feature vectors.
In some possible embodiments, the inductive learning algorithm comprises graphsage algorithm.
In some possible embodiments, in processing the social network information according to a generalized learning algorithm to obtain a second feature vector, the above program is specifically configured to execute instructions for:
Processing the social network information according to the graphsage algorithm to obtain a second feature vector, which specifically includes:
Generating a social graph network according to the social network information;
Sampling neighbor nodes of a target node to obtain R neighbor nodes, wherein the target node is a node corresponding to the user in the social graph network;
weighting the feature vectors of the R neighbor nodes to obtain a fourth feature vector;
And transversely splicing the characteristic vector of the target node and the fourth characteristic vector to obtain the second characteristic vector.
In some possible embodiments, the social statistical information includes P items of information, and the program is specifically configured to execute the following instructions in obtaining a third feature vector according to the social statistical information:
obtaining sub-vectors corresponding to each item of information in the P items of information to obtain P sub-vectors;
And splicing the P sub-vectors to obtain the third feature vector.
Fig. 7 is a block diagram of another subscriber tag generating apparatus according to an embodiment of the present application, where an apparatus 700 includes a processor 710, a transceiver 720, and at least one circuit 730, where the processor and the transceiver are connected by the at least one circuit, and where:
A transceiver 710, configured to obtain a behavior text sequence, social network information, and social statistics information of a user, and send the behavior text sequence, the social network information, and the social statistics information to a processor 720, where the behavior text sequence is a sequence composed of texts generated according to a chronological order and used to characterize a behavior of the user;
a processor 720 for randomly initializing the behavior text sequence using a two-way long-short term memory network BiLSTM to obtain a first feature vector;
The processor 720 is further configured to process the social network information according to a inductive learning algorithm to obtain a second feature vector;
The processor 720 is further configured to obtain a third feature vector according to the social statistical information;
The processor 720 is further configured to laterally splice the first feature vector, the second feature vector, and the third feature vector to obtain a user tag of the user.
In some possible embodiments, the behavior text sequence includes N texts, where N texts are used to characterize the behavior of the user at N times, where N is an integer greater than 1, and before the behavior text sequence is randomly initialized using the two-way long-short-term memory network BiLSTM, the processor 720 is further configured to denoise the N texts according to a pre-constructed industry dictionary to obtain T texts, where T is equal to or less than N, and where T is an integer greater than or equal to 1;
in randomly initializing the behavioral text sequence using the two-way long-short term memory network BiLSTM to obtain a first feature vector, the processor 720 is specifically configured to:
randomly initializing each text in the T texts by using the BiLSTM to obtain T feature vectors;
determining the weight of each text in the T texts according to the T feature vectors;
and weighting the T feature vectors according to the weight of each text to obtain the first feature vector.
In some possible embodiments, the processor 720 is specifically configured to, using the BiLSTM to randomly initialize each of the T texts to obtain T feature vectors:
extracting keywords corresponding to each text in the T texts;
Randomly initializing keywords corresponding to each text by using the forward LSTM of BiLSTM to obtain forward feature vectors corresponding to each text, and randomly initializing keywords corresponding to each text by using the backward LSTM of BiLSTM to obtain backward feature vectors corresponding to each text;
and transversely splicing the forward feature vector and the backward feature vector corresponding to each text in the T texts to obtain T feature vectors.
In some possible embodiments, the inductive learning algorithm comprises graphsage algorithm.
In some possible embodiments, the processor 720 is specifically configured to, in processing the social network information according to a generalized learning algorithm to obtain a second feature vector:
Processing the social network information according to the graphsage algorithm to obtain a second feature vector, which is specifically used for:
Generating a social graph network according to the social network information;
Sampling neighbor nodes of a target node to obtain R neighbor nodes, wherein the target node is a node corresponding to the user in the social graph network;
weighting the feature vectors of the R neighbor nodes to obtain a fourth feature vector;
And transversely splicing the characteristic vector of the target node and the fourth characteristic vector to obtain the second characteristic vector.
In some possible embodiments, the social statistical information includes P items of information, and the processor 720 is specifically configured to:
obtaining sub-vectors corresponding to each item of information in the P items of information to obtain P sub-vectors;
And splicing the P sub-vectors to obtain the third feature vector.
Embodiments of the present application also provide a computer storage medium storing a computer program that is executed by a processor to implement some or all of the steps of any one of the user tag generation methods described in the method embodiments above.
Embodiments of the present application also provide a computer program product comprising a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform part or all of the steps of any one of the user tag generation methods described in the method embodiments above.
It should be noted that, for simplicity of description, the foregoing method embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present application is not limited by the order of acts described, as some steps may be performed in other orders or concurrently in accordance with the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are alternative embodiments, and that the acts and modules referred to are not necessarily required for the present application.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and for parts of one embodiment that are not described in detail, reference may be made to related descriptions of other embodiments.
In the several embodiments provided by the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, such as the division of the units, merely a logical function division, and there may be additional manners of dividing the actual implementation, such as multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, or may be in electrical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit 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 described above may be implemented either in hardware or in software program modules.
The integrated units, if implemented in the form of software program modules, may be stored in a computer-readable memory for sale or use as a stand-alone product. Based on this understanding, the technical solution of the present application may be embodied essentially or partly in the form of a software product, or all or part of the technical solution, which is stored in a memory, and includes several instructions for causing a computer device (which may be a personal computer, a server, a network device, or the like) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned memory includes: a usb disk, a Read-Only TeTory, a random access memory (RAT, randoT Access TeTory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Those of ordinary skill in the art will appreciate that all or a portion of the steps in the various methods of the above embodiments may be implemented by a program that instructs associated hardware, and the program may be stored in a computer readable memory, which may include: flash disk, read-Only TeTory (English: read-Only TeTory, for short: ROT), random access device (English: randoT Access TeTory, for short: RAT), magnetic disk or optical disk, etc.
The foregoing has outlined rather broadly the more detailed description of embodiments of the application, wherein the principles and embodiments of the application are explained in detail using specific examples, the above examples being provided solely to facilitate the understanding of the method and core concepts of the application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.
Claims (6)
1. A method for generating a user tag, comprising:
Acquiring a behavior text sequence of a user, wherein the behavior text sequence comprises N texts, the N texts are used for representing the behaviors of the user at N moments, and N is an integer greater than 1; denoising the N texts according to a pre-constructed industry dictionary to obtain T texts, wherein T is less than or equal to N, and T is an integer greater than or equal to 1; randomly initializing each text in the T texts by using a two-way long-short-term memory network BiLSTM to obtain T feature vectors; determining the weight of each text in the T texts according to the T feature vectors; weighting the T feature vectors according to the weight of each text to obtain a first feature vector; the behavior text sequence is a sequence formed by texts which are generated according to time sequence and used for representing the behavior of the user;
Acquiring social network information of the user, processing the social network information according to graphsage algorithm to obtain a second feature vector, including: generating a social graph network according to the social network information; sampling neighbor nodes of a target node according to a preset sampling duty ratio to obtain R neighbor nodes, wherein the target node is a node corresponding to the user in the social graph network, and the sampling duty ratio is the duty ratio of the neighbor nodes for sampling to all the neighbor nodes; weighting the feature vectors of the R neighbor nodes to obtain a fourth feature vector; transversely splicing the characteristic vector of the target node and the fourth characteristic vector to obtain the second characteristic vector;
Acquiring social statistical information of the user, and acquiring a third feature vector according to the social statistical information;
and transversely splicing the first feature vector, the second feature vector and the third feature vector to obtain the user tag of the user.
2. The method of claim 1, wherein said randomly initializing each of said T texts using said BiLSTM to obtain T feature vectors comprises:
extracting keywords corresponding to each text in the T texts;
Randomly initializing keywords corresponding to each text by using the forward LSTM of BiLSTM to obtain forward feature vectors corresponding to each text, and randomly initializing keywords corresponding to each text by using the backward LSTM of BiLSTM to obtain backward feature vectors corresponding to each text;
and transversely splicing the forward feature vector and the backward feature vector corresponding to each text in the T texts to obtain T feature vectors.
3. The method according to claim 1 or 2, wherein the social statistical information includes P items of information, and the obtaining a third feature vector according to the social statistical information includes:
obtaining sub-vectors corresponding to each item of information in the P items of information to obtain P sub-vectors;
And splicing the P sub-vectors to obtain the third feature vector.
4. A user tag generating apparatus, comprising a processor, a transceiver and at least one circuit, the processor and the transceiver being connected by the at least one circuit;
The transceiver is used for acquiring a behavior text sequence, social network information and social statistical information of a user and sending the behavior text sequence, the social network information and the social statistical information to the processor, wherein the behavior text sequence is a sequence formed by texts which are generated according to time sequence and are used for representing the behavior of the user; the behavior text sequence comprises N texts, wherein the N texts are used for representing the behaviors of the user at N moments, and N is an integer greater than 1;
The processor is used for denoising the N texts according to a pre-constructed industry dictionary to obtain T texts, T is less than or equal to N, and T is an integer greater than or equal to 1; randomly initializing each text in the T texts by using a two-way long-short-term memory network BiLSTM to obtain T feature vectors; determining the weight of each text in the T texts according to the T feature vectors; weighting the T feature vectors according to the weight of each text to obtain a first feature vector;
The processor is further configured to process the social network information according to graphsage algorithm to obtain a second feature vector, including: generating a social graph network according to the social network information; sampling neighbor nodes of a target node according to a preset sampling duty ratio to obtain R neighbor nodes, wherein the target node is a node corresponding to the user in the social graph network, and the sampling duty ratio is the duty ratio of the neighbor nodes for sampling to all the neighbor nodes; weighting the feature vectors of the R neighbor nodes to obtain a fourth feature vector; transversely splicing the characteristic vector of the target node and the fourth characteristic vector to obtain the second characteristic vector;
the processor is further used for obtaining a third feature vector according to the social statistical information;
And the processor is further used for transversely splicing the first feature vector, the second feature vector and the third feature vector to obtain the user tag of the user.
5. The apparatus of claim 4, wherein the processor is configured to, using the BiLSTM, randomly initialize each of the T texts to obtain T feature vectors, specifically:
extracting keywords corresponding to each text in the T texts;
Randomly initializing keywords corresponding to each text by using the forward LSTM of BiLSTM to obtain forward feature vectors corresponding to each text, and randomly initializing keywords corresponding to each text by using the backward LSTM of BiLSTM to obtain backward feature vectors corresponding to each text;
and transversely splicing the forward feature vector and the backward feature vector corresponding to each text in the T texts to obtain T feature vectors.
6. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program, which is executed by a processor to implement the method of any of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911000656.2A CN110781407B (en) | 2019-10-21 | 2019-10-21 | User tag generation method, device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911000656.2A CN110781407B (en) | 2019-10-21 | 2019-10-21 | User tag generation method, device and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110781407A CN110781407A (en) | 2020-02-11 |
| CN110781407B true CN110781407B (en) | 2024-07-23 |
Family
ID=69386111
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911000656.2A Active CN110781407B (en) | 2019-10-21 | 2019-10-21 | User tag generation method, device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110781407B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111488807B (en) * | 2020-03-29 | 2023-10-10 | 复旦大学 | Video description generation system based on graph rolling network |
| CN113591886B (en) * | 2020-04-30 | 2023-11-07 | 北京百度网讯科技有限公司 | Method, apparatus, device and computer readable storage medium for information classification |
| CN112765481B (en) * | 2020-06-29 | 2023-09-19 | 腾讯科技（深圳）有限公司 | Data processing method, device, computer and readable storage medium |
| CN112347316A (en) * | 2020-10-21 | 2021-02-09 | 上海淇玥信息技术有限公司 | GraphSAGE-based bad preference behavior detection method and device and electronic equipment |
| CN112559904B (en) * | 2020-12-16 | 2021-10-22 | 中国计量大学 | Conversational social recommendation method based on door mechanism and multi-modal graph network |
| CN114742569A (en) * | 2021-01-08 | 2022-07-12 | 广州视源电子科技股份有限公司 | User life stage prediction method and device, computer equipment and storage medium |
| CN113486235A (en) * | 2021-05-31 | 2021-10-08 | 杭州摸象大数据科技有限公司 | Method and system for identifying user interest |
| CN113344067A (en) * | 2021-05-31 | 2021-09-03 | 中国工商银行股份有限公司 | Method, device and equipment for generating customer portrait |
| CN113407845A (en) * | 2021-07-14 | 2021-09-17 | 上海明略人工智能(集团)有限公司 | Method and device for information recommendation, electronic equipment and storage medium |
| CN113553433B (en) * | 2021-09-17 | 2022-01-07 | 平安科技（深圳）有限公司 | Product classification method, device, medium and terminal equipment based on artificial intelligence |
| CN114493535B (en) * | 2022-01-28 | 2022-09-06 | 北京乐开科技有限责任公司 | Asset model-based data center system for data driving |
| CN114792406A (en) * | 2022-05-11 | 2022-07-26 | 上海任意门科技有限公司 | Method, apparatus, and medium for generating a tag for user-generated content |
| CN115114934B (en) * | 2022-07-15 | 2024-07-12 | 广东工业大学 | Combined extraction method for label fusion |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110197389A (en) * | 2019-03-04 | 2019-09-03 | 腾讯科技（深圳）有限公司 | A kind of user identification method and device |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8386574B2 (en) * | 2009-10-29 | 2013-02-26 | Xerox Corporation | Multi-modality classification for one-class classification in social networks |
| CN104899253B (en) * | 2015-05-13 | 2018-06-26 | 复旦大学 | Towards the society image across modality images-label degree of correlation learning method |
| US10922609B2 (en) * | 2017-05-17 | 2021-02-16 | Facebook, Inc. | Semi-supervised learning via deep label propagation |
| KR102003487B1 (en) * | 2017-08-28 | 2019-10-01 | 서울시립대학교 산학협력단 | Apparatus and method for providing body tag recognizing model, and apparatus for applying body tag recognizing model |
| CN108763319B (en) * | 2018-04-28 | 2022-02-08 | 中国科学院自动化研究所 | Social robot detection method and system fusing user behaviors and text information |
| CN109522406A (en) * | 2018-10-11 | 2019-03-26 | 平安科技（深圳）有限公司 | Text semantic matching process, device, computer equipment and storage medium |
| CN109697285B (en) * | 2018-12-13 | 2022-06-21 | 中南大学 | Hierarchical BilSt Chinese electronic medical record disease coding and labeling method for enhancing semantic representation |
| CN110097125B (en) * | 2019-05-07 | 2022-10-14 | 郑州轻工业学院 | Cross-network account association method based on embedded representation |
| CN110209823B (en) * | 2019-06-12 | 2021-04-13 | 齐鲁工业大学 | Multi-label text classification method and system |
| CN110321560B (en) * | 2019-06-25 | 2021-10-01 | 北京邮电大学 | Method and device for determining position information from text information and electronic equipment |
-
2019
- 2019-10-21 CN CN201911000656.2A patent/CN110781407B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110197389A (en) * | 2019-03-04 | 2019-09-03 | 腾讯科技（深圳）有限公司 | A kind of user identification method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110781407A (en) | 2020-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110781407B (en) | User tag generation method, device and computer readable storage medium | |
| CN110163220B (en) | Picture feature extraction model training method and device and computer equipment | |
| CN111931062A (en) | Training method and related device of information recommendation model | |
| CN109284406B (en) | Intention identification method based on difference cyclic neural network | |
| Xie et al. | A Survey on Machine Learning‐Based Mobile Big Data Analysis: Challenges and Applications | |
| CN116664719B (en) | Image redrawing model training method, image redrawing method and device | |
| TW201915790A (en) | Generating document for a point of interest | |
| CN111581966A (en) | Context feature fusion aspect level emotion classification method and device | |
| CN111382555A (en) | Data processing method, medium, device and computing equipment | |
| CN107918778B (en) | Information matching method and related device | |
| CN111353303A (en) | Word vector construction method and device, electronic equipment and storage medium | |
| CN111125386B (en) | Media resource processing method and device, storage medium and electronic device | |
| CN112149604A (en) | Training method of video feature extraction model, video recommendation method and device | |
| CN114398973B (en) | Media content tag identification method, device, equipment and storage medium | |
| CN114330966A (en) | Risk prediction method, device, equipment and readable storage medium | |
| CN113392179A (en) | Text labeling method and device, electronic equipment and storage medium | |
| CN116977701A (en) | Video classification model training method, video classification method and device | |
| CN113742572B (en) | Data recommendation method and device, electronic equipment and storage medium | |
| CN116628345A (en) | Content recommendation method and device, electronic equipment and storage medium | |
| CN114528491A (en) | Information processing method, information processing device, computer equipment and storage medium | |
| CN113128225B (en) | Named entity identification method and device, electronic equipment and computer storage medium | |
| CN112417260B (en) | Localized recommendation method, device and storage medium | |
| CN114547308A (en) | Text processing method and device, electronic equipment and storage medium | |
| Oh et al. | Dynamic programming approach to visual place recognition in changing environments | |
| CN111581455A (en) | Text generation model generation method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40022449 Country of ref document: HK |
|
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TG01 | Patent term adjustment | ||
| TG01 | Patent term adjustment |