Recommendation method and device, equipment and storage medium

Abstract

The invention provides a recommendation method and device, equipment and a storage medium, and relates to the technical field of computers.The method comprises the steps that an object feature attribute graph and a spatio-temporal feature attribute graph are constructed according to object information data, object feature values are determined according to the object feature attribute graph, and the object feature values are sent to a server; the server is used for clustering the object characteristic values of the plurality of terminal devices to obtain common characteristics of the plurality of object characteristic values, and determining an individual characteristic vector and a spatial-temporal characteristic vector based on individual characteristics corresponding to the object characteristic attribute graph, spatial-temporal characteristics corresponding to the spatial-temporal characteristic attribute graph and the common characteristics; and determining recommendation data based on the personality feature vector, the spatio-temporal feature vector and a recommendation model. And the accuracy of the recommended data is improved.

Description

Background Art

[0002] As the degree of social informatization continues to deepen, the amount of data that users need to face is gradually increasing. Users need to spend a lot of time and resources to determine the data they need from a large amount of data. How to recommend the data that users need to users is a problem that needs to be solved urgently in this field. Summary of the invention

[0003] The present disclosure provides a recommendation method, apparatus, device and storage medium, which at least to a certain extent improve the accuracy of recommending data to users.

[0004] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0005] According to one aspect of the present disclosure, a recommendation method is provided, which is applied to a terminal device and includes:

[0006] Construct an object feature attribute graph and a spatiotemporal feature attribute graph according to the object information data;

[0007] Determine the object feature value according to the object feature attribute graph;

[0008] The object feature values ​​are sent to a server so that the server clusters the object feature values ​​of multiple terminal devices to obtain common features of the multiple object feature values;

[0009] Determine the individual feature vector and the spatiotemporal feature vector based on the individual features corresponding to the object feature attribute graph, the spatiotemporal features corresponding to the spatiotemporal feature attribute graph, and the common features;

[0010] The recommended data is determined based on the personality feature vector, the spatiotemporal feature vector and the recommendation model.

[0011] In one embodiment of the present disclosure, determining an object feature value according to an object feature attribute graph includes:

[0012] Determine the Laplacian matrix corresponding to the object feature attribute graph;

[0013] Decompose the Laplace matrix to obtain the object eigenvalues.

[0014] In one embodiment of the present disclosure, the method further includes:

[0015] The object feature attribute graph and the spatiotemporal feature attribute graph are respectively input into the graph representation learning network to obtain the individual features corresponding to the object feature attribute graph and the spatiotemporal features corresponding to the spatiotemporal feature attribute graph.

[0016] In one embodiment of the present disclosure, the method further includes:

[0017] Train the intelligent model to obtain recommended model parameters;

[0018] The recommendation model parameters are sent to the server, so that the server aggregates the recommendation model parameters respectively sent by multiple terminal devices to obtain a recommendation model.

[0019] In one embodiment of the present disclosure, training the intelligent model to obtain recommended model parameters includes:

[0020] The intelligent model is trained, and the training is stopped when the loss function value corresponding to the loss function reaches a preset condition, so as to obtain the recommended model parameters;

[0021] The loss function is:

[0022] L = -∑logσ(r ui -r uj )+λ(||U|| 2 +||A|| 2 )

[0023] Among them, L is the loss function value, σ(x) is the sigmoid function, that is, σ(x) = 1 / (1+exp(-x)), r ui and r uj are the output results of inputting i and j into the model respectively, U is the personality feature vector, A is the spatiotemporal feature vector, and λ is the regularization parameter.

[0024] According to another aspect of the present disclosure, a recommendation method is provided, which is applied to a server and includes:

[0025] receiving an object feature value sent by a terminal, where the object feature value is determined by the terminal device based on the object feature map;

[0026] Clustering is performed based on the object feature values ​​sent by multiple terminal devices to obtain a cluster center;

[0027] Determine the cluster center as the common feature, and send the common feature to the terminal device, so that the terminal device determines the individual feature vector and the spatiotemporal feature vector based on the common feature, and determines the recommendation model parameters according to the individual feature vector and the spatiotemporal feature vector;

[0028] The recommendation model is obtained by aggregating the recommendation model parameters respectively sent by multiple terminal devices.

[0029] According to another aspect of the present disclosure, a recommendation device is provided, which is applied to a terminal device, including:

[0030] A construction module, used to construct an object feature attribute graph and a spatiotemporal feature attribute graph according to object information data;

[0031] A first determination module, used to determine an object feature value according to an object feature attribute graph;

[0032] A sending module, used for sending the object feature value to a server, so that the server clusters the object feature values ​​of multiple terminal devices to obtain common features of the multiple object feature values;

[0033] A second determination module is used to determine the individual feature vector and the spatiotemporal feature vector based on the individual features corresponding to the object feature attribute graph, the spatiotemporal features corresponding to the spatiotemporal feature attribute graph, and the common features;

[0034] The third determination module is used to determine the recommended data based on the personality feature vector, the spatiotemporal feature vector and the recommendation model.

[0035] In one embodiment of the present disclosure, the first determining module includes:

[0036] A first determining unit, used to determine the Laplacian matrix corresponding to the object feature attribute graph;

[0037] The decomposition unit is used to decompose the Laplace matrix to obtain the eigenvalues ​​of the object.

[0038] In one embodiment of the present disclosure, the device further includes:

[0039] The input module is used to input the object feature attribute graph and the spatiotemporal feature attribute graph into the graph representation learning network respectively, so as to obtain the individual features corresponding to the object feature attribute graph and the spatiotemporal features corresponding to the spatiotemporal feature attribute graph.

[0040] In one embodiment of the present disclosure, the device further includes:

[0041] The training module is used to train the intelligent model and obtain the recommended model parameters;

[0042] The sending module is used to send the recommendation model parameters to the server, so that the server aggregates the recommendation model parameters sent by multiple terminal devices to obtain the recommendation model.

[0043] In one embodiment of the present disclosure, the training module includes:

[0044] The first training unit is used to train the intelligent model, stop the training when the loss function value corresponding to the loss function reaches a preset condition, and obtain the recommended model parameters;

[0045] The loss function is:

[0046] L = -∑logσ(r ui -r uj )+λ(||U|| 2 +||A|| 2 )

[0047] Among them, L is the loss function value, σ(x) is the sigmoid function, that is, σ(x) = 1 / (1+exp(-x)), r ui and r uj are the output results of inputting i and j into the model respectively, U is the personality feature vector, A is the spatiotemporal feature vector, and λ is the regularization parameter.

[0048] According to another aspect of the present disclosure, a recommendation device is provided, which is applied to a server and includes:

[0049] A receiving module, used for receiving an object feature value sent by a terminal, where the object feature value is determined by the terminal device based on the object feature map;

[0050] A clustering module, used to cluster object feature values ​​sent by multiple terminal devices to obtain cluster centers;

[0051] A fourth determination module, configured to determine the cluster center as a common feature, and send the common feature to the terminal device, so that the terminal device determines an individual feature vector and a spatiotemporal feature vector based on the common feature, and determines a recommendation model parameter according to the individual feature vector and the spatiotemporal feature vector;

[0052] The aggregation module is used to aggregate the recommendation model parameters sent by multiple terminal devices to obtain a recommendation model.

[0053] According to another aspect of the present disclosure, an electronic device is provided, including: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the above-mentioned recommendation method by executing the executable instructions.

[0054] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned recommendation method is implemented.

[0055] The recommendation method, device, equipment and storage medium provided by the embodiments of the present disclosure construct an object feature attribute graph and a spatiotemporal feature attribute graph according to object information data, determine an object feature value according to the object feature attribute graph, and send the object feature value to a server, so that the server clusters the object feature values ​​of multiple terminal devices, obtains common features of multiple object feature values, determines a personality feature vector and a spatiotemporal feature vector based on the personality features corresponding to the object feature attribute graph, the spatiotemporal features corresponding to the spatiotemporal feature attribute graph, and the common features, and determines recommended data based on the personality feature vector, the spatiotemporal feature vector and the recommendation model. The accuracy of the recommended data is improved.

[0056] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0058] Figure 1 A flowchart of a recommended method in an embodiment of the present disclosure is shown;

[0059] Figure 2 Another recommended method flow chart in the embodiment of the present disclosure is shown;

[0060] Figure 3 A flowchart of another recommendation method in an embodiment of the present disclosure is shown;

[0061] Figure 4 A flowchart of another recommendation method in an embodiment of the present disclosure is shown;

[0062] Figure 5 A schematic diagram of the structure of a recommendation system in an embodiment of the present disclosure is shown;

[0063] Figure 6 A schematic diagram of a recommended device in an embodiment of the present disclosure is shown;

[0064] Figure 7 Another schematic diagram of a recommended device in an embodiment of the present disclosure is shown;

[0065] Figure 8 A structural block diagram of an electronic device in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0066] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0067] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0068] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0069] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0070] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0071] With the rapid development of mobile networks and mobile devices, the number of applications is growing. Personalized application recommendation technology can use artificial intelligence technology to recommend applications that match users' interests or habits. The development of this technology can help users discover new and interesting applications and improve user experience. At the same time, it is also of great significance for operators to provide higher quality personalized services. However, the development of recommendation algorithms always has problems with data sparsity and privacy security.

[0072] In order to solve the above technical problems, the present disclosure provides a recommendation method, device, equipment and storage medium. In order to explain the present disclosure in detail, the present disclosure first explains the migration method.

[0073] It should be pointed out that, in the absence of conflict, the embodiments of the present disclosure and the technical features therein may be combined with each other.

[0074] The specific implementation method of the embodiment of the present disclosure is described in detail below with reference to the accompanying drawings.

[0075] Figure 1 A schematic diagram of a recommended method flow in an embodiment of the present disclosure is shown. Figure 1 As shown, the method may include:

[0076] S110, constructing an object feature attribute graph and a spatiotemporal feature attribute graph according to the object information data.

[0077] In some embodiments, the object information data includes the user's personal information and the information of the user's interaction with the APP. Among them, the user's personal information includes the user's gender, age, the location information visited by the user and other information related to the user. The information of the user's interaction with the APP may include the type of APP used and the length of time different types of APPs are used.

[0078] In some embodiments, the object feature attribute graph may include a node graph with user personal information as nodes.

[0079] In some embodiments, the spatiotemporal feature attribute graph may include a node graph with APP as a node.

[0080] S120, determining an object feature value according to the object feature attribute graph.

[0081] In some embodiments, determining the object feature value according to the object feature attribute map includes:

[0082] Determine the Laplacian matrix corresponding to the object feature attribute graph;

[0083] Decompose the Laplace matrix to obtain the object eigenvalues.

[0084] In some embodiments, the object feature attribute graph may be G ua =(V,E,A ua ), where V includes the set of nodes in the object feature attribute graph and the spatiotemporal feature attribute graph, E is the set of edges between nodes in the object feature attribute graph and the spatiotemporal feature attribute graph, and A ua It is the attribute matrix of the object feature attribute graph node, recording the node a used by user u.

[0085] In some embodiments, G ua Spectral clustering is performed to determine the Laplacian matrix corresponding to the object feature attribute graph. The method for determining the Laplacian matrix is ​​not specifically limited in the embodiments of the present disclosure.

[0086] Exemplarily, the Laplacian matrix may be:

[0087] L u =DA=I-DAD -1

[0088] Among them, by performing eigenvalue decomposition on the Laplace matrix, the first k non-zero eigenvectors and eigenvalues ​​of the Laplace matrix are calculated, and the eigenvectors corresponding to the first k smallest eigenvalues ​​are selected as the object eigenvalues.

[0089] S130, sending the object feature value to the server, so that the server clusters the object feature values ​​of multiple terminal devices to obtain common features of the multiple object feature values.

[0090] In some embodiments, after K-means clustering, the server can set the number range of clusters and use the silhouette coefficient for evaluation, and finally obtain the number of clusters k that maximizes the silhouette coefficient, and obtain k cluster center vectors, where the cluster center vectors can be common features.

[0091] S140, determining an individual feature vector and a spatiotemporal feature vector based on the individual features corresponding to the object feature attribute graph, the spatiotemporal features corresponding to the spatiotemporal feature attribute graph, and the common features.

[0092] In some embodiments, individual features, spatiotemporal features, and common features may be input into a feature aggregation network to obtain an individual feature vector and a spatiotemporal feature vector.

[0093] In some embodiments, before inputting the individual features, spatiotemporal features, and common features into the feature aggregation network to obtain the individual feature vector and the spatiotemporal feature vector, the method may further include:

[0094] The intelligent network is trained to obtain a trained feature aggregation network.

[0095] S150, determining recommended data based on the individual feature vector, the spatiotemporal feature vector and the recommendation model.

[0096] In some embodiments, the recommendation model may be a trained model obtained by the server through federated learning.

[0097] The recommendation method provided by the embodiment of the present disclosure constructs an object feature attribute graph and a spatiotemporal feature attribute graph according to object information data, determines an object feature value according to the object feature attribute graph, and sends the object feature value to a server, so that the server clusters the object feature values ​​of multiple terminal devices, obtains common features of multiple object feature values, determines a personality feature vector and a spatiotemporal feature vector based on the personality features corresponding to the object feature attribute graph, the spatiotemporal features corresponding to the spatiotemporal feature attribute graph, and the common features, and determines recommended data based on the personality feature vector, the spatiotemporal feature vector, and the recommendation model. The accuracy of the recommended data is improved.

[0098] Figure 2 FIG. 4 shows another flow chart of a recommended method in an embodiment of the present disclosure. Figure 2 As shown, the method may include:

[0099] S210, constructing an object feature attribute graph and a spatiotemporal feature attribute graph according to the object information data;

[0100] S220, determining an object feature value according to the object feature attribute graph;

[0101] S230, sending the object feature value to the server, so that the server clusters the object feature values ​​of multiple terminal devices to obtain common features of the multiple object feature values;

[0102] S240, inputting the object feature attribute graph and the spatiotemporal feature attribute graph into the graph representation learning network respectively, and obtaining individual features corresponding to the object feature attribute graph and spatiotemporal features corresponding to the spatiotemporal feature attribute graph.

[0103] In some embodiments, before the object feature attribute graph and the spatiotemporal feature attribute graph are respectively input into the graph representation learning network to obtain the individual features corresponding to the object feature attribute graph and the spatiotemporal features corresponding to the spatiotemporal feature attribute graph, the graph representation learning network can be trained. When the training stop condition is met, a trained graph representation learning network can be obtained.

[0104] S250, determining a personality feature vector and a spatiotemporal feature vector based on the personality feature corresponding to the object feature attribute graph, the spatiotemporal feature corresponding to the spatiotemporal feature attribute graph, and the common feature;

[0105] S260, determining recommended data based on the personality feature vector, the spatiotemporal feature vector, and the recommendation model.

[0106] The recommendation method provided by the embodiment of the present disclosure constructs an object feature attribute graph and a spatiotemporal feature attribute graph according to object information data, determines an object feature value according to the object feature attribute graph, and sends the object feature value to a server, so that the server clusters the object feature values ​​of multiple terminal devices, obtains common features of multiple object feature values, determines a personality feature vector and a spatiotemporal feature vector based on the personality features corresponding to the object feature attribute graph, the spatiotemporal features corresponding to the spatiotemporal feature attribute graph, and the common features, and determines recommended data based on the personality feature vector, the spatiotemporal feature vector, and the recommendation model. The accuracy of the recommended data is improved.

[0107] Figure 3 A schematic diagram of another recommended method flow in an embodiment of the present disclosure is shown. Figure 3 As shown, the method may include:

[0108] S310, constructing an object feature attribute graph and a spatiotemporal feature attribute graph according to the object information data;

[0109] S320, determining an object feature value according to the object feature attribute graph;

[0110] S330, sending the object feature value to the server, so that the server clusters the object feature values ​​of multiple terminal devices to obtain common features of the multiple object feature values;

[0111] S340, determining a personality feature vector and a spatiotemporal feature vector based on the personality feature corresponding to the object feature attribute graph, the spatiotemporal feature corresponding to the spatiotemporal feature attribute graph, and the common feature;

[0112] S350, training the intelligent model to obtain recommended model parameters;

[0113] S360: Send the recommendation model parameters to the server, so that the server aggregates the recommendation model parameters respectively sent by multiple terminal devices to obtain a recommendation model.

[0114] S370, determining recommended data based on the individual feature vector, the spatiotemporal feature vector and the recommendation model.

[0115] In some embodiments, the intelligent model is trained to obtain recommended model parameters, including: training the intelligent model, stopping the training when the loss function value corresponding to the loss function reaches a preset condition, and obtaining the recommended model parameters.

[0116] In some embodiments, the intelligent model may be a distributed recommendation model.

[0117] In some embodiments, after the terminal device is trained to obtain the recommendation model parameters, the server may aggregate the recommendation model parameters to obtain a pre-recommendation model, and then determine whether the loss function value meets a preset condition based on the pre-recommendation model.

[0118] In some embodiments, after the terminal device trains and obtains the recommendation model parameters, the server may aggregate the recommendation model parameters to obtain a pre-recommendation model. The server then sends the pre-recommendation model to the terminal device, and the terminal device determines whether the loss function value meets the preset condition based on the pre-recommendation model.

[0119] In some embodiments, the loss function may be:

[0120] L=-Σlogσ(r ui -r uj )+λ(||U|| 2 +||A|| 2 )

[0121] Among them, L is the loss function value, σ(x) is the sigmoid function, that is, σ(x) = 1 / (1+exp(-x)), r ui and r uj are the output results of inputting i and j into the model respectively, U is the personality feature vector, A is the spatiotemporal feature vector, and λ is the regularization parameter.

[0122] In some embodiments, since federated learning is trained at each terminal device, each terminal device will generate a gradient u, but uploading each user gradient to the server will consume a lot of communication resources and reduce the efficiency of model training and updating. Therefore, the server needs to schedule and extract k from N terminal devices. n The terminal devices are asked to upload the gradient parameters to the server after differential privacy encryption, and the server aggregates the parameters.

[0123] In some embodiments, the present disclosure does not specifically limit differential privacy encryption.

[0124] In some embodiments, the preset condition may include obtaining a difference between two consecutive loss function values ​​less than 10 -4 .

[0125] The recommendation method provided by the embodiment of the present disclosure constructs an object feature attribute graph and a spatiotemporal feature attribute graph according to object information data, determines an object feature value according to the object feature attribute graph, and sends the object feature value to a server, so that the server clusters the object feature values ​​of multiple terminal devices, obtains common features of multiple object feature values, determines a personality feature vector and a spatiotemporal feature vector based on the personality features corresponding to the object feature attribute graph, the spatiotemporal features corresponding to the spatiotemporal feature attribute graph, and the common features, and determines recommended data based on the personality feature vector, the spatiotemporal feature vector, and the recommendation model. The accuracy of the recommended data is improved.

[0126] Based on the same inventive concept, the present disclosure also provides a recommended method in the following embodiments. Since the principle of solving the problem in this method embodiment is similar to that in the above method embodiment, the implementation of this method embodiment can refer to the implementation of the above method embodiment, and the repeated parts will not be repeated.

[0127] Figure 4 A flowchart of another recommended method in an embodiment of the present disclosure is shown. It should be noted that the migration method in the present disclosure is applied to a server.

[0128] like Figure 4 As shown, the method may include:

[0129] S410, receiving an object feature value sent by a terminal, where the object feature value is determined by the terminal device based on an object feature map;

[0130] S420, clustering the object feature values ​​sent by multiple terminal devices to obtain a cluster center;

[0131] S430, determining the cluster center as a common feature, and sending the common feature to the terminal device, so that the terminal device determines an individual feature vector and a spatiotemporal feature vector based on the common feature, and determines a recommendation model parameter according to the individual feature vector and the spatiotemporal feature vector;

[0132] S440, aggregating the recommendation model parameters respectively sent by the multiple terminal devices to obtain a recommendation model.

[0133] The recommendation method provided by the embodiment of the present disclosure constructs an object feature attribute graph and a spatiotemporal feature attribute graph according to object information data, determines an object feature value according to the object feature attribute graph, and sends the object feature value to a server, so that the server clusters the object feature values ​​of multiple terminal devices, obtains common features of multiple object feature values, determines a personality feature vector and a spatiotemporal feature vector based on the personality features corresponding to the object feature attribute graph, the spatiotemporal features corresponding to the spatiotemporal feature attribute graph, and the common features, and determines recommended data based on the personality feature vector, the spatiotemporal feature vector, and the recommendation model. The accuracy of the recommended data is improved.

[0134] Based on the same inventive concept, the present disclosure also provides a recommendation system, such as the following embodiment. Since the principle of solving the problem in this method embodiment is similar to that in the above method embodiment, the implementation of this system embodiment can refer to the implementation of the above method embodiment, and the repeated parts will not be repeated.

[0135] Figure 5 A schematic diagram of the structure of a recommendation system in an embodiment of the present disclosure is shown, and the system can be applied to the recommendation method and recommendation device in various embodiments of the present disclosure.

[0136] like Figure 5 As shown, the recommendation system 10 may include a terminal device 101 and a server 102 .

[0137] In some embodiments, the terminal device 101 and the server 102 may be located on different devices, and the terminal device 101 may be a module on an electronic device with data acquisition capability. The server 102 may be a module on an electronic device with data processing capability such as a computer or a server. The terminal device 101 and the server 102 may be located on the same device, such as the terminal device 101 and the server 102 may be an input module and a processing module on a computer or a mobile phone.

[0138] The terminal device 101 and the server 102 are connected to each other via a network, which may be a wired network or a wireless network.

[0139] Optionally, the wireless network or wired network described above uses standard communication technology and / or protocol. The network is usually the Internet, but it can also be any network, including but not limited to a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or any combination of a virtual private network). In some embodiments, technologies and / or formats including Hyper Text Mark-up Language (HTML), Extensible Markup Language (XML), etc. are used to represent data exchanged through the network. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above data communication technologies.

[0140] The following describes a situation where the terminal device 101 and the server 102 are located on two different devices.

[0141] The terminal device 101 may be located on a terminal device, which may be various electronic devices, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, wearable devices, augmented reality devices, virtual reality devices, etc.

[0142] Optionally, the client of the application installed in different terminal devices is the same, or the client of the same type of application based on different operating systems. Based on the different terminal platforms, the specific form of the client of the application can also be different, for example, the application client can be a mobile client, a PC client, etc.

[0143] The server 102 may be located on a server, which may be a server that provides various services, such as a background management server that provides support for devices operated by users using terminal devices. The background management server may analyze and process the received request and other data, and feed back the processing results to the terminal device.

[0144] Optionally, the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), as well as big data and artificial intelligence platforms.

[0145] Based on the same inventive concept, the present disclosure also provides a recommended device in the following embodiments. Since the principle of solving the problem in the device embodiment is similar to that in the above method embodiment, the implementation of the device embodiment can refer to the implementation of the above method embodiment, and the repeated parts will not be repeated.

[0146] Figure 6 A schematic diagram of a recommendation device in an embodiment of the present disclosure is shown. It should be noted that the recommendation device in the present disclosure is applied to a terminal device.

[0147] like Figure 6 As shown, the recommendation device 600 may include:

[0148] A construction module 610 is used to construct an object feature attribute graph and a spatiotemporal feature attribute graph according to the object information data;

[0149] A first determination module 620, configured to determine an object feature value according to an object feature attribute graph;

[0150] A sending module 630 is used to send the object feature value to the server, so that the server clusters the object feature values ​​of multiple terminal devices to obtain common features of the multiple object feature values;

[0151] A second determination module 640 is used to determine a personality feature vector and a spatiotemporal feature vector based on the personality feature corresponding to the object feature attribute graph, the spatiotemporal feature corresponding to the spatiotemporal feature attribute graph, and the common feature;

[0152] The third determination module 650 is used to determine the recommended data based on the personality feature vector, the spatiotemporal feature vector and the recommendation model.

[0153] In one embodiment of the present disclosure, the first determining module includes:

[0154] A first determining unit, used to determine the Laplacian matrix corresponding to the object feature attribute graph;

[0155] The decomposition unit is used to decompose the Laplace matrix to obtain the eigenvalues ​​of the object.

[0156] In one embodiment of the present disclosure, the device further includes:

[0157] The input module is used to input the object feature attribute graph and the spatiotemporal feature attribute graph into the graph representation learning network respectively, so as to obtain the individual features corresponding to the object feature attribute graph and the spatiotemporal features corresponding to the spatiotemporal feature attribute graph.

[0158] In one embodiment of the present disclosure, the device further includes:

[0159] The training module is used to train the intelligent model and obtain the recommended model parameters;

[0160] The sending module is used to send the recommendation model parameters to the server, so that the server aggregates the recommendation model parameters sent by multiple terminal devices to obtain the recommendation model.

[0161] In one embodiment of the present disclosure, the training module includes:

[0162] The first training unit is used to train the intelligent model, stop the training when the loss function value corresponding to the loss function reaches a preset condition, and obtain the recommended model parameters;

[0163] The loss function is:

[0164] L = -∑logσ(r ui -r uj )+λ(||U|| 2 +||A|| 2 )

[0165] Among them, L is the loss function value, σ(x) is the sigmoid function, that is, σ(x) = 1 / (1+exp(-x)), r ui and r uj are the output results of inputting i and j into the model respectively, U is the personality feature vector, A is the spatiotemporal feature vector, and λ is the regularization parameter.

[0166] The recommendation device provided by the embodiment of the present disclosure constructs an object feature attribute graph and a spatiotemporal feature attribute graph according to object information data, determines an object feature value according to the object feature attribute graph, and sends the object feature value to a server, so that the server clusters the object feature values ​​of multiple terminal devices, obtains common features of multiple object feature values, determines a personality feature vector and a spatiotemporal feature vector based on the personality features corresponding to the object feature attribute graph, the spatiotemporal features corresponding to the spatiotemporal feature attribute graph, and the common features, and determines recommended data based on the personality feature vector, the spatiotemporal feature vector, and the recommendation model. The accuracy of the recommended data is improved.

[0167] Figure 7 A schematic diagram of another recommendation device in an embodiment of the present disclosure is shown. It should be noted that the recommendation device in the present disclosure is applied to a server.

[0168] like Figure 7 As shown, the migration device 700 may include:

[0169] A receiving module 710 is configured to receive an object feature value sent by a terminal, where the object feature value is determined by the terminal device based on an object feature map;

[0170] A clustering module 720, configured to perform clustering based on the object feature values ​​sent by multiple terminal devices to obtain a cluster center;

[0171] A fourth determination module 730 is used to determine the cluster center as a common feature, and send the common feature to the terminal device, so that the terminal device determines the individual feature vector and the spatiotemporal feature vector based on the common feature, and determines the recommendation model parameters according to the individual feature vector and the spatiotemporal feature vector;

[0172] The aggregation module 740 is used to aggregate the recommendation model parameters respectively sent by multiple terminal devices to obtain a recommendation model.

[0173] The recommendation device provided in the embodiment of the present disclosure can be used to execute the identification method provided in the above-mentioned method embodiments. The implementation principle and technical effect thereof are similar and will not be described in detail here for the sake of brevity.

[0174] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0175] Refer to the following Figure 8 The electronic device 800 according to this embodiment of the present disclosure is described. Figure 8 The electronic device 800 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0176] like Figure 8 As shown, the electronic device 800 is in the form of a general computing device. The components of the electronic device 800 may include but are not limited to: at least one processing unit 810, at least one storage unit 820, and a bus 830 connecting different system components (including the storage unit 820 and the processing unit 810).

[0177] The storage unit stores program codes, which can be executed by the processing unit 810, so that the processing unit 810 performs the steps of various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification. For example, the processing unit 810 can perform the following steps of the above method embodiment:

[0178] Constructing an object feature attribute graph and a spatiotemporal feature attribute graph according to the object information data;

[0179] Determining an object feature value according to the object feature attribute graph;

[0180] Sending the object feature value to a server so that the server clusters the object feature values ​​of multiple terminal devices to obtain common features of the multiple object feature values;

[0181] Determine a personality feature vector and a spatiotemporal feature vector based on the personality feature corresponding to the object feature attribute graph, the spatiotemporal feature corresponding to the spatiotemporal feature attribute graph, and the common feature;

[0182] Recommendation data is determined based on the personality feature vector, the spatiotemporal feature vector, and a recommendation model.

[0183] Alternatively, receiving an object feature value sent by a terminal, wherein the object feature value is determined by the terminal device based on an object feature map;

[0184] Clustering is performed based on the object feature values ​​sent by the plurality of terminal devices to obtain a cluster center;

[0185] Determine the cluster center as a common feature, and send the common feature to a terminal device, so that the terminal device determines an individual feature vector and a spatiotemporal feature vector based on the common feature, and determines a recommendation model parameter according to the individual feature vector and the spatiotemporal feature vector;

[0186] The recommendation model is obtained by aggregating the recommendation model parameters respectively sent by multiple terminal devices.

[0187] The storage unit 820 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 8201 and / or a cache storage unit 8202 , and may further include a read-only storage unit (ROM) 8203 .

[0188] The storage unit 820 may also include a program / utility 8204 having a set (at least one) of program modules 8205, such program modules 8205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0189] Bus 830 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0190] The electronic device 800 may also process images with one or more external devices 840 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also process images with one or more devices that enable users to interact with the electronic device 800, and / or may process images with any device that enables the electronic device 800 to process images with one or more other computing devices (e.g., routers, modems, etc.). Such image processing may be performed through an input / output (I / O) interface 850. Furthermore, the electronic device 800 may also process images with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 860. As shown, the network adapter 860 processes images with other modules of the electronic device 800 through a bus 830. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0191] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0192] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, which may be a readable signal medium or a readable storage medium. A program product capable of implementing the above method of the present disclosure is stored thereon. In some possible implementations, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above “Exemplary Method” section of this specification.

[0193] More specific examples of computer-readable storage media in the present disclosure may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0194] In the present disclosure, a computer readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, wherein a readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A readable signal medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0195] Alternatively, the program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0196] In a specific implementation, the program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0197] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0198] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0199] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0200] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the appended claims.

Claims

1. A recommendation method, characterized in that: Applied to terminal equipment, including: Construct an object feature attribute graph and a spatiotemporal feature attribute graph according to the object information data; Determining an object feature value according to the object feature attribute graph; Sending the object feature value to a server so that the server clusters the object feature values ​​of multiple terminal devices to obtain common features of the multiple object feature values; Determine a personality feature vector and a spatiotemporal feature vector based on the personality feature corresponding to the object feature attribute graph, the spatiotemporal feature corresponding to the spatiotemporal feature attribute graph, and the common feature; Recommendation data is determined based on the personality feature vector, the spatiotemporal feature vector, and a recommendation model.

2. The recommendation method according to claim 1, characterized in that: Determining the object feature value according to the object feature attribute graph includes: Determine the Laplacian matrix corresponding to the object feature attribute graph; The Laplace matrix is ​​decomposed to obtain object eigenvalues.

3. The recommendation method according to claim 1, characterized in that: The method further comprises: The object feature attribute graph and the spatiotemporal feature attribute graph are respectively input into a graph representation learning network to obtain individual features corresponding to the object feature attribute graph and spatiotemporal features corresponding to the spatiotemporal feature attribute graph.

4. The recommendation method according to claim 1, characterized in that: The method further comprises: Train the intelligent model to obtain recommended model parameters; The recommendation model parameters are sent to the server, so that the server aggregates the recommendation model parameters respectively sent by multiple terminal devices to obtain a recommendation model.

5. The recommendation method according to claim 4, characterized in that: The intelligent model is trained to obtain recommended model parameters, including: The intelligent model is trained, and the training is stopped when the loss function value corresponding to the loss function reaches a preset condition, so as to obtain the recommended model parameters; The loss function is: L=-∑logσ(r ui -r uj )+λ(||U|| 2 +||A|| 2 ) Among them, L is the loss function value, σ(x) is the sigmoid function, that is, σ(x) = 1 / (1+exp(-x)), r ui and r uj are the output results of inputting i and j into the model respectively, U is the personality feature vector, A is the spatiotemporal feature vector, and λ is the regularization parameter.

6. A recommendation method, characterized in that: Applicable to servers, including: receiving an object feature value sent by a terminal, wherein the object feature value is determined by the terminal device based on an object feature map; Clustering is performed based on the object feature values ​​sent by multiple terminal devices to obtain a cluster center; Determine the cluster center as a common feature, and send the common feature to a terminal device, so that the terminal device determines an individual feature vector and a spatiotemporal feature vector based on the common feature, and determines a recommendation model parameter according to the individual feature vector and the spatiotemporal feature vector; The recommendation model is obtained by aggregating the recommendation model parameters respectively sent by multiple terminal devices.

7. A recommendation device, characterized in that: Applied to terminal equipment, including: A construction module, used to construct an object feature attribute graph and a spatiotemporal feature attribute graph according to object information data; A first determining module, configured to determine an object feature value according to the object feature attribute graph; A sending module, used for sending the object feature value to a server, so that the server clusters the object feature values ​​of multiple terminal devices to obtain common features of the multiple object feature values; A second determination module is used to determine a personality feature vector and a spatiotemporal feature vector based on the personality feature corresponding to the object feature attribute graph, the spatiotemporal feature corresponding to the spatiotemporal feature attribute graph, and the common feature; The third determination module is used to determine the recommended data based on the personality feature vector, the spatiotemporal feature vector and the recommendation model.

8. A recommendation device, characterized in that: Applicable to servers, including: A receiving module, configured to receive an object feature value sent by a terminal, wherein the object feature value is determined by the terminal device based on an object feature map; A clustering module, used to cluster object feature values ​​sent by multiple terminal devices to obtain cluster centers; A fourth determination module, configured to determine the cluster center as a common feature, and send the common feature to a terminal device, so that the terminal device determines an individual feature vector and a spatiotemporal feature vector based on the common feature, and determines a recommendation model parameter according to the individual feature vector and the spatiotemporal feature vector; The aggregation module is used to aggregate the recommendation model parameters sent by multiple terminal devices to obtain a recommendation model.

9. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to execute the recommendation method according to any one of claims 1 to 6 by executing the executable instructions.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the recommendation method according to any one of claims 1 to 6 is implemented.

Images