Sequence recommendation method for shared account scene

Abstract

The invention discloses a shared account scene-oriented sequence recommendation method, which belongs to the technical field of shared account-oriented sequence recommendation, and comprises the following steps of: constructing a recommendation model based on a capsule graph convolutional network and depth subspace clustering to carry out personalized recommendation, the recommendation model comprises a main capsule graph construction module, a capsule graph convolutional network, an account-level dynamic routing module, a sequence decoder based on depth subspace clustering and a prediction module. The method comprises the following steps: establishing a sequence interaction diagram; further processing the sequence interaction graph into a main capsule graph; and learning the vector representation of the shared account and the vector representation of the mixed sequence, and generating a personalized recommendation result. According to the invention, unique preferences of different potential users in the shared account sequence can be mined, and a more accurate project recommendation result is generated for the shared account.

Description

Technical Field

[0001] The present invention belongs to the technical field of shared account sequence recommendation, and in particular relates to a sequence recommendation method for a shared account scenario. Background Art

[0002] In video platforms and e-commerce networks, users' historical interactive behaviors are usually recorded by the platform in chronological order. As one of the core functions of the platform, the recommendation system captures users' long-term behavior patterns by analyzing their historical behavior sequences, and then provides personalized recommendation results to improve the user experience and platform activity. However, as the research continues to deepen, researchers have found that users tend to share an account among family or close friends. In this scenario, the historical interactive behaviors of multiple potential users are mixed and recorded in the same shared account sequence, which poses a challenge to the current recommendation system. By using methods such as capsule networks, graph convolutional networks, and subspace clustering, the diverse interests of potential users in shared accounts can be effectively modeled, thereby improving the accuracy of the recommendation system.

[0003] In the existing technology, π-net (a shared account sequence recommendation method based on recurrent neural network) and PSJNet (a shared account sequence recommendation method based on hierarchical recurrent network) learn the representation of shared accounts through recurrent neural network and clustering network, but this method cannot analyze the relationship between potential users and account behaviors in the account in a fine-grained manner. DA-GCN (a shared account sequence recommendation method based on graph neural network) and TiDA-GCN (a shared account sequence recommendation method based on graph neural network and self-attention mechanism) respectively capture the interests of potential users in shared accounts through graph neural network and multi-head self-attention mechanism, ignoring the high parameter overhead and training cost of the model. In general, the existing technical methods cannot distinguish the ownership of each interaction in the mixed sequence of shared accounts, and are limited by high computational complexity and parameter overhead. Summary of the invention

[0004] In order to solve the above problems, the present invention proposes a sequence recommendation method for shared account scenarios and constructs a recommendation model based on capsule graph convolutional network and deep subspace clustering. Specifically, a lightweight capsule graph convolutional network is first proposed to identify the ownership of each interaction of potential users. In this part, a master capsule graph is constructed to identify the interaction ownership of different potential users. A graph convolutional network with an attention mechanism is used to propagate messages on the graph, which realizes a fine-grained distinction of the preferences of different potential users. Subsequently, an account-level dynamic routing mechanism is designed to merge the preferences of potential users and generate account-level vector representations. In addition, a sequence decoder based on deep subspace clustering is proposed. The sequence decoder uses a low-rank subspace basis to model the interests of potential users and adopts a contrastive learning strategy to align the preferences of potential users in the sequence representation.

[0005] The technical solution of the present invention is as follows: A sequence recommendation method for shared account scenarios, constructing a recommendation model based on capsule graph convolutional network and deep subspace clustering for personalized recommendation, the recommendation model includes a main capsule graph construction module, a capsule graph convolutional network, an account-level dynamic routing module, a sequence decoder based on deep subspace clustering and a prediction module; the method includes the following steps: Step 1: Obtain the interaction relationship between the shared account and the project and the sequence dependency relationship between the projects in the shared account scenario, and establish a sequence interaction diagram based on these two relationships; Step 2: The sequence interaction graph is further processed into a main capsule graph through the main capsule graph construction module; Step 3: Learn the vector representation of potential users in the shared account and the vector representation of the mixed sequence from the main capsule graph through the capsule graph convolutional network; Step 4: Aggregate the vector representations of potential users in the shared account into the vector representation of the shared account through the account-level dynamic routing module; Step 5: Align user interests with the vector representation of the mixed sequence through a sequence decoder based on deep subspace clustering; Step 6: Generate personalized recommendation results based on the vector representation of the shared account and the vector representation of the mixed sequence.

[0006] Furthermore, the specific process of step 1 is as follows: Get the project collection in the shared account scenario as ,in, Indicates projects, is the total number of projects; the shared account set is ,in, For the Shared accounts, is the total number of shared accounts; the mixed sequence set is ,in, Indicates A mixed sequence of shared accounts; The sequence interaction diagram established is ,in, Represents a collection of nodes, Represents a set of edges, each edge in the set represents a relationship between two different nodes in the sequence interaction graph; the sequence interaction graph contains two types of relationships, namely the interaction relationship between shared accounts and projects and the sequence dependency relationship between projects; the above two relationships are represented by the adjacency matrix express: (1); in, represents the adjacency matrix that stores the sequential dependencies between items; if Projects In the mixed sequence, Projects The predecessor node of Middle Row, No. Elements of a column ,otherwise, ;No. Items corresponding to Row, No. Items corresponding to List; represents the adjacency matrix containing the interaction relationship between shared accounts and projects. If Shared accounts With Projects There is interaction between them, then Middle Row, No. The element corresponding to the column ,otherwise, ;No. The shared account corresponds to Row, No. Items corresponding to List; is the transpose symbol.

[0007] Furthermore, in step 2, the main capsule graph construction module includes a linear attention module and a one-dimensional convolution module; the specific working process of the main capsule graph construction module is: Step 2.1: Convert the feature representation of the items in the sequence interaction graph into the capsule representation of the items through the linear attention module. The specific process is as follows: (2); in, The capsule representation of the item at layer 0 in the main capsule graph obtained after conversion. The number of layers in the main capsule graph corresponds to the number of layers in the sequence interaction graph. , They are the embedding dimensions of feature representation and capsule representation respectively; , , They are The embedded representation of the query, key, and value initialized by the value of It is the feature representation of the items in the 0th layer of the sequence interaction graph obtained by random initialization; is the softmax function; is the dimension transformation matrix of the linear attention module; is the bias term of the linear attention module; Step 2.2: Convert the feature representation of the shared account in the sequence interaction graph into the capsule representation of the shared account through a point-by-point one-dimensional convolution module. The specific calculation process is as follows: (3); in, The capsule representation of the shared account at layer 0 in the main capsule graph obtained after conversion; is the feature representation of the shared account at layer 0 in the randomly initialized sequence interaction graph; Represents a one-dimensional convolution operation; Represents the one-dimensional convolution kernel of the one-dimensional convolution module; Represents the bias of the one-dimensional convolution module; is a hyperparameter that controls the number of potential users in a shared account; Step 2.3: Assume that each shared account contains potential users, and divide the capsule representation of each shared account into The capsule representation of a potential user is as follows: Shared accounts Capsules indicate Divide Capsule representation of potential users , is the first layer of the main capsule image after conversion. of shared accounts Potential users The capsule representation of the main capsule graph is completed while retaining the relationship in the sequence interaction graph.

[0008] Furthermore, in step 3, the capsule graph convolution network includes several graph convolution layers, and each layer must perform a graph convolution operation through the graph convolution network with an attention mechanism. The specific working process is as follows: Step 3.1: Calculate the The information obtained by the capsule representation of potential users during layer graph convolution is used to update the capsule representation of potential users. The specific process is as follows: Step 3.1.1. Calculate the correlation between the capsule representations of potential users in the shared account and the capsule representations of items that have interacted with the shared account on the main capsule graph through the graph convolutional network with attention. The specific calculation process is as follows: (4); in, for and The correlation between For the of shared accounts Potential users The capsule indicates; For the The shared account has interacted Projects The capsule indicates; is an exponential function with base e; Indicates Shared accounts The collection of items that have interacted; Step 3.1.2, calculate the The capsule of the project during the layer graph convolution represents the information conveyed to the potential user, specifically: (5); in, For the When layer graph convolution The capsule indicates the information conveyed; Indicated in The first learnable weight during layer graph convolution; For the When the layer graph is convolved with The shared account has interacted Projects The capsule indicates; For the The second learnable weight during layer graph convolution; For the When layer graph convolution and The correlation between At the same time, calculation The self-connection information is as follows: (6); in, For the When layer graph convolution Self-connection information; For the The third learnable weight during layer graph convolution; For the When layer graph convolution of shared accounts Potential users The capsule indicates; Step 3.1.3: The capsule indicates that an update is being performed. The update process is as follows: (7); in, To conduct the Updated after layer graph convolution The capsule indicates; Indicates all The collection of items that have been interacted with; Step 3.2: Calculate the The capsule representation of the project obtains information during the layer graph convolution and updates the capsule representation of the project; the specific calculation process is as follows: (8); (9); in, Indicates shared account with Projects There is interaction potential users; Indicated in When layer graph convolution The capsule indicates The capsules represent the messages delivered; Indicated in The fourth learnable weight that controls the amount of information during layer graph convolution; For the When layer graph convolution The capsule indicates; Indicates that in a mixed sequence Adjacent projects; Indicated in When layer graph convolution is performed from The capsule indicates The capsules represent the messages delivered; Indicated in The fifth learnable weight that controls the amount of information during layer graph convolution; For the When layer graph convolution The capsule indicates; Step 3.3: The capsule representation update process of the project is as follows: (10); in, To conduct the Updated after layer graph convolution The capsule indicates; Indicates all Potential users who have interacted A collection of; Indicates that in all mixed sequences A collection of adjacent items; Step 3.4: Use a multi-layer aggregation protocol to aggregate the representations obtained through graph convolution on each layer to obtain a complete item vector representation and potential user vector representation; the specific calculation process is as follows: (11); (12); in, For complete The vector representation of ; For complete The vector representation of ; is the total number of layers of graph convolution; Step 3.5: Calculate the vector representation of the final required mixed sequence. The specific calculation process is as follows: (13); in, is the vector representation of the mixed sequence.

[0009] Furthermore, the specific process of step 4 is as follows: Step 4.1: Account-level dynamic routing module iterative calculation sequence The dynamic routing process of this iteration is as follows: (14); in, Indicates At iteration Account-level capsule representation of ; Indicates the function of compressing routing information; For the The first iteration The coupling coefficient between the capsule representation of each potential user and the capsule representation of the account; Step 4.2: After completing a routing process, the coupling coefficient needs to be updated. The specific process is as follows: (15); in, For the The first iteration The coupling coefficient between the capsule representation of each potential user and the capsule representation of the account; It is the learnable weight matrix in the account-level dynamic routing module; It is the element inner product operation; Step 4.3: After the iteration, the complete account capsule representation is obtained, i.e. At iteration Account-level capsule representation ; Step 4.4: Calculate the vector representation of the shared account. The specific calculation process is as follows: (16); in, is the vector representation of the shared account.

[0010] Furthermore, the specific process of step 5 is as follows: Step 5.1: Initialize the subspace basis The affinity between the calculated sequence representation and the subspace basis is represented; different subspaces in the subspace basis represent the user interests of different potential users under the same shared account; Indicates subspaces; is the total number of subspaces; the specific calculation process of affinity is as follows: (17); in, express Middle A vector representation of each item; express and The subspace affinity between them; represents the hyperparameter that controls the smoothness of the calculation; Step 5.2: Use contrastive learning method to align the vector representation of the affinity-enhanced mixed sequence with the vector representation of the original mixed sequence; the specific process is: First, subspace affinities are assigned to the vector representations of items in the mixed sequence: (18); in, Indicates the enhanced A vector representation of items, A vector representation representing the enhanced mixed sequence; Then, the vector representation of the enhanced mixed sequence The vector representation of the original mixed sequence after regularization Compute the loss of contrastive learning between , the specific contrastive learning loss calculation process is as follows: (19); in, , They are Middle Projects, The regularized embedding vector representation of each item; Indicates the enhanced A vector representation of each item; represents the temperature coefficient that controls the influence from negative pairs to positive pairs; Finally, for all mixed sequences, the complete contrastive learning loss is expressed as : (20); Step 5.3: and The residual link method is used for refinement; the vector representation of the refined complete mixed sequence is obtained through the following calculation process: (twenty one); in, The vector representation of the complete mixed sequence after refinement; is the standard L2 regularization function; Represents the learnable weight matrix in the residual link method.

[0011] Furthermore, the specific process of step 6 is as follows: Step 6.1: and Input to the prediction module for prediction. The specific process is as follows: (twenty two); in, represents the predicted probability distribution, For the projects; It is the transformation matrix in the prediction module that maps the predicted value to the dimension of the candidate item set; is the bias term for adjusting the threshold of the activation function in the prediction module; Sort all items in descending order according to the size of the predicted probability distribution. The items are the personalized recommendation results for the shared account, is a preset threshold; Step 6.2: Use cross entropy loss to optimize the learnable parameters of the prediction model. The calculation process is: (twenty three); Step 6.3, add the contrastive learning loss in step 5.2 , the complete loss function of the prediction model is : (twenty four); in, is a hyperparameter that controls the degree of involvement of the self-supervision signal from the contrastive learning loss.

[0012] The beneficial technical effects brought by the present invention are as follows: by constructing a main capsule graph and designing a graph convolutional network with an attention mechanism, the interactive ownership of different potential users in a shared account is identified, and fine-grained differentiation of preferences of different potential users is achieved; by designing a sequence decoder with deep subspace clustering enhanced by contrastive learning, the diverse interests of potential users are simulated, and the sequence representation and the preferences of potential users are aligned. The present invention utilizes a capsule graph convolutional network enhanced by deep subspace clustering to simultaneously model potential users in a shared account in a fine-grained manner, and effectively considers the diverse interests of potential users in the sequence representation learning process, thereby improving the accuracy of sequence modeling, and further recommending more accurate items for shared accounts. The recommended items in the present invention include but are not limited to electronic goods, video programs, etc., so the present invention has a wide range of application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flow chart of the sequence recommendation method for the shared account scenario of the present invention. DETAILED DESCRIPTION

[0014] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments: This paper takes the mixed sequence of shared accounts as the research object, and takes mining the behavioral preferences in the mixed sequence of shared accounts as the core goal. The key technical problems that need to be solved include: First, it is difficult to capture the preferences of different potential users in the shared accounts in a fine-grained manner. Second, the diverse interests of potential users are not fully considered in the learning process of sequence representation. Solving these two problems can more accurately model the mixed sequence of shared accounts.

[0015] Therefore, the specific key issues to be solved by the present invention are: Key technical problem 1: It is difficult to capture the preferences of different potential users in a shared account in a fine-grained manner; This paper proposes a graph capsule convolutional network to identify the ownership of each interaction in a mixed sequence of shared accounts by potential users. In this part, a capsule graph is constructed to represent different potential users and their interaction relationships with items (items include but are not limited to electronic goods, video programs, etc.) in a fine-grained manner. By propagating messages on the capsule graph with a graph convolution method with an attention mechanism, this component achieves fine-grained differentiation of preferences of different potential users.

[0016] Key technical issue 2: The diverse interests of potential users are not fully considered in the sequence representation learning process; This paper designs a sequence representation alignment method based on deep subspace clustering. This part uses low-rank subspace basis to simulate the interests of potential users and uses subspace affinity to refine sequence representation. By applying contrastive learning strategy, this part realizes the alignment of potential users with the diverse interests of potential users in subspace.

[0017] like Figure 1 As shown, the present invention proposes a sequence recommendation method for shared account scenarios, constructs a recommendation model based on capsule graph convolutional network and deep subspace clustering, and performs personalized recommendations based on interactive behavior information and serialized preference information from shared account mixed behavior sequences; the recommendation model includes a main capsule graph construction module, a capsule graph convolutional network, an account-level dynamic routing module, a sequence decoder based on deep subspace clustering, and a prediction module; the method includes the following steps: Step 1: Obtain the interaction relationship between the shared account and the project and the sequence dependency relationship between the projects in the shared account scenario, and establish a sequence interaction diagram based on the above two relationships; the specific process is as follows: Get the project collection in the shared account scenario as ,in, Indicates projects, is the total number of projects; the shared account set is ,in, For the Shared accounts, is the total number of shared accounts; the mixed sequence set is ,in, Indicates A mixed sequence of shared accounts; The sequence interaction diagram established is ,in, Represents a collection of nodes, Represents a set of edges, each edge in the set represents a relationship between two different nodes in the sequence interaction graph. The sequence interaction graph contains two types of relationships, namely the interaction relationship between shared accounts and projects and the sequence dependency relationship between projects. The above two relationships are represented by the adjacency matrix express: (1); in, represents the adjacency matrix that stores the sequential dependencies between items. Projects In the mixed sequence, it is Projects The predecessor node of Middle Row, No. Elements of a column ,otherwise, ;No. Items corresponding to Row, No. Items corresponding to List; represents the adjacency matrix containing the interaction relationship between shared accounts and projects. If Shared accounts With Projects There is interaction between them, then Middle Row, No. The element corresponding to the column ,otherwise, . No. The shared account corresponds to Row, No. Items corresponding to List. is the transpose symbol.

[0018] Step 2: Based on the sequence interaction graph, the sequence interaction graph is further processed into the main capsule graph through the main capsule graph construction module. The main capsule graph construction module includes a linear attention module and a one-dimensional convolution module; the specific working process of the main capsule graph construction module is as follows: Step 2.1: Convert the feature representation of the items in the sequence interaction graph into the capsule representation of the items through the linear attention module. The specific process is as follows: (2); in, The capsule representation of the item at layer 0 in the main capsule graph obtained after conversion. The number of layers in the main capsule graph corresponds to the number of layers in the sequence interaction graph. , They are the embedding dimensions of feature representation and capsule representation respectively; , , They are The embedded representation of the query, key, and value initialized by the value of It is the feature representation of the items in the 0th layer of the sequence interaction graph obtained by random initialization; is the softmax function; is the dimension transformation matrix of the linear attention module; is the bias term of the linear attention module.

[0019] Step 2.2: Convert the feature representation of the shared account in the sequence interaction graph into the capsule representation of the shared account through a point-by-point one-dimensional convolution module. The specific calculation process is as follows: (3); in, The capsule representation of the shared account at layer 0 in the main capsule graph obtained after conversion; is the feature representation of the shared account at layer 0 in the randomly initialized sequence interaction graph; Represents a one-dimensional convolution operation; Represents the one-dimensional convolution kernel of the one-dimensional convolution module; Represents the bias of the one-dimensional convolution module; is a hyperparameter that controls the number of potential users in a shared account.

[0020] Step 2.3: Assume that each shared account contains potential users, for example, ,in Indicates of shared accounts potential users. The capsule representation of each shared account is divided into Capsule representation of potential users, for example, the first capsule in layer 0 of the main capsule graph obtained after conversion Shared accounts Capsules indicate Divide Capsule representation of potential users , is the first layer of the main capsule image after conversion. of shared accounts Potential users The capsule representation of the main capsule graph is completed while retaining the relationship in the sequence interaction graph.

[0021] Step 3: Learn the vector representation of potential users in the shared account and the vector representation of mixed sequences from the main capsule graph through the capsule graph convolution network; the capsule graph convolution network contains several graph convolution layers, each of which must be convolved once through the graph convolution network with an attention mechanism. The specific working process is as follows: Step 3.1: Calculate the The information obtained by the capsule representation of the potential user during the layer graph convolution is used to update the capsule representation of the potential user. For example, the specific process is: Step 3.1.1. Calculate the correlation between the capsule representations of potential users in the shared account and the capsule representations of items that have interacted with the shared account on the main capsule graph through the graph convolutional network with attention. The specific calculation process is as follows: (4); in, for and The correlation between For the of shared accounts Potential users The capsule indicates; For the The shared account has interacted Projects The capsule indicates; is an exponential function with base e; Indicates Shared accounts A collection of items that have interacted.

[0022] Step 3.1.2: After obtaining the correlation information, calculate The capsule of the project during the layer graph convolution represents the information conveyed to potential users, specifically: (5); in, For the When layer graph convolution The capsule indicates the information conveyed; Indicated in The first learnable weights during layer graph convolution, which control the amount of information passed from neighboring items; For the When the layer graph is convolved with The shared account has interacted Projects The capsule indicates; For the The second learnable weight during layer graph convolution controls the influence of the attention mechanism; For the When layer graph convolution and The correlation between them.

[0023] At the same time, calculation The self-connection information is as follows: (6); in, For the When layer graph convolution Self-connection information; For the The third learnable weight during the layer graph convolution controls the The amount of information retained in For the When layer graph convolution of shared accounts Potential users Capsules indicated.

[0024] Step 3.1.3: The capsule indicates that an update is being performed. The update process is as follows: (7); in, To conduct the Updated after layer graph convolution The capsule indicates; Indicates all A collection of items that have been interacted with.

[0025] Step 3.2: Calculate the The capsule representation of the item is obtained during the convolution of the layer graph, and the capsule representation of the item is updated. Projects Take this as an example, the specific calculation process is as follows: (8); (9); in, Indicates shared account with Projects There is interaction potential users; Indicated in When layer graph convolution The capsule indicates The capsules represent the messages delivered; Indicated in The fourth learnable weight that controls the amount of information during layer graph convolution; For the When layer graph convolution The capsule indicates; Indicates that in a mixed sequence Adjacent projects; Indicated in When layer graph convolution is performed from The capsule indicates The capsules represent the messages delivered; Indicated in The fifth learnable weight that controls the amount of information during layer graph convolution; For the When layer graph convolution Capsules indicated.

[0026] Step 3.3: After receiving messages from adjacent potential users and items, the capsule representation of the item is updated as follows: (10); in, To conduct the Updated after layer graph convolution The capsule indicates; Indicates all Potential users who have interacted A collection of; Indicates that in all mixed sequences A collection of adjacent items.

[0027] Step 3.4: Use a multi-layer aggregation protocol to aggregate the representations obtained through graph convolution on each layer to obtain a complete item vector representation and potential user vector representation. The specific calculation process is as follows: (11); (12); in, For complete The vector representation of ; For complete The vector representation of ; is the total number of graph convolution layers.

[0028] Step 3.5: By collecting all the learned item vector representations, we can obtain the vector representation of the final mixed sequence. The specific calculation process is as follows: (13); in, is the vector representation of the mixed sequence.

[0029] Step 4: Aggregate the vector representations of potential users in the shared account into the vector representation of the shared account through the account-level dynamic routing module; the specific process is as follows: Step 4.1: In the process of account-level dynamic routing, the connection strength between different potential user capsules and shared account capsules needs to be calculated, which is represented by a randomly initialized coupling coefficient. Second-rate, is a hyperparameter. The dynamic routing process of this iteration is as follows: (14); in, Indicates At iteration Account-level capsule representation of ; Indicates the function of compressing routing information; For the The first iteration The coupling coefficient between the capsule representation of each potential user and the capsule representation of the account.

[0030] Step 4.2: After completing a routing process, the coupling coefficient needs to be updated. The specific process is as follows: (15); in, For the The first iteration The coupling coefficient between the capsule representation of each potential user and the capsule representation of the account; It is the learnable weight matrix in the account-level dynamic routing module; It is the element-wise inner product operation.

[0031] Step 4.3: After account-level dynamic routing, the model obtains a complete account capsule representation, i.e. At iteration Account-level capsule representation .

[0032] Step 4.4: Calculate the vector representation of the shared account by combining all learned account representations. The specific calculation process is as follows: (16); in, is the vector representation of the shared account.

[0033] Step 5: Align user interests with the vector representation of the mixed sequence through a sequence decoder based on deep subspace clustering; the specific process is as follows: Step 5.1: Initialize the subspace basis The affinity between the calculated sequence representation and the subspace basis is represented; different subspaces in the subspace basis represent the user interests of different potential users under the same shared account; Indicates subspaces; is the total number of subspaces. Mixed sequence of shared accounts For example, The vector representation is , the specific calculation process is as follows: (17); in, express Middle A vector representation of each item; express and The subspace affinity between them; Represents a hyperparameter that controls the smoothness of the computation.

[0034] Step 5.2: Use contrastive learning to align the vector representation of the affinity-enhanced mixed sequence with the vector representation of the original mixed sequence. The specific process is: First, subspace affinities are assigned to the vector representations of items in the mixed sequence: (18); in, Indicates the enhanced A vector representation of items, The vector representation of the enhanced mixed sequence.

[0035] Then, the vector representation of the enhanced mixed sequence The vector representation of the original mixed sequence after regularization Compute the loss of contrastive learning between , the specific contrastive learning loss calculation process is as follows: (19); in, , They are Middle Projects, The regularized embedding vector representation of each item; Indicates the enhanced A vector representation of each item; represents a pair of positive examples in the contrastive learning loss calculation process; represents a pair of negative examples in the contrastive learning loss calculation process; Represents the temperature coefficient that controls the influence from negative pairs to positive pairs.

[0036] Finally, for all mixed sequences, the complete contrastive learning loss is expressed as : (20); Step 5.3: and The residual link method is used for refinement. The vector representation of the refined complete mixed sequence can be obtained through the following calculation process: (twenty one); in, The vector representation of the complete mixed sequence after refinement; is the standard L2 regularization function; Represents the learnable weight matrix in the residual link method.

[0037] Step 6: Generate personalized recommendation results based on the learned vector representation of shared accounts and the vector representation of mixed sequences. The specific process is as follows: Step 6.1: Represent the learned shared account vectors And the refined vector representation of the complete mixed sequence Input to the prediction module for prediction. The specific process is as follows: (twenty two); in, represents the predicted probability distribution, For the projects; It is the transformation matrix in the prediction module that maps the predicted value to the dimension of the candidate item set; It is the bias term for adjusting the threshold of the activation function in the prediction module.

[0038] Sort all items in descending order according to the size of the predicted probability distribution. The items are the personalized recommendation results for the shared account, is a preset threshold.

[0039] Step 6.2: Use cross entropy loss to optimize the learnable parameters of the prediction model. The calculation process is: (twenty three); Step 6.3: Add the contrastive learning loss used in step 5.2 , the complete loss function of the prediction model is : (twenty four); in, is a hyperparameter that controls the degree of involvement of the self-supervision signal from the contrastive learning loss.

[0040] The recommendation model based on capsule graph convolutional network and deep subspace clustering constructed by the present invention is referred to as LightGC²N model. In order to prove the feasibility and superiority of the present invention, a series of comparative experiments were carried out. In terms of experimental design, the experimental data sets used in the present invention include four data sets: Hvideo-E, Hvideo-V, Hamazon-M, and Hamazon-B. These data sets are all from real shared account scenarios and record the mixed behavior sequences of shared accounts. For the four data sets, the ratio of training set to test set is set to 2:8. The specific experiments carried out are as follows: Experiment 1: The recommendation performance of the proposed model LightGC²N on the MRR (mean reciprocal rank) evaluation index was evaluated compared with nine competitive baseline methods. The specific baseline methods used are as follows: NCF (recommendation method based on neural collaborative filtering network), LightGCN (recommendation method based on lightweight graph convolutional network), HRNN (sequence recommendation method based on recurrent neural network), NAIS (sequence recommendation method based on attention), TGSRec (sequence recommendation method based on time-aware collaborative filtering), π-Net (shared account sequence recommendation method based on recurrent neural network), PSJ-Net (shared account sequence recommendation method based on hierarchical recurrent network), DA-GCN (shared account sequence recommendation method based on graph neural network), TiDA-GCN (shared account sequence recommendation method based on graph neural network and self-attention mechanism).

[0041] The experimental results of different methods on four real shared account sequence recommendation datasets are shown in Table 1. Among them, the higher the value of MRR, the higher the model accuracy. It can be observed from Table 1 that the model LightGC²N of the present invention has achieved the best performance on all four datasets. LightGC²N outperforms other state-of-the-art shared account sequence recommendation methods (π-Net, PSJNet, DA-GCN, DA-GCN). This observation shows that it is important to capture the fine-grained differences between potential users in shared accounts. At the same time, LightGC²N outperforms other graph-based shared account sequence recommendation methods (i.e., DA-GCN and TiDA-GCN), proving the advantages of the graph capsule attention network proposed in the present invention in modeling complex relationships in the shared account sequence recommendation scenario. In addition, LightGC²N achieved the best performance on all datasets, proving the superiority of the graph capsule convolutional network and subspace alignment method proposed in the present invention in the shared account sequence recommendation scenario.

[0042] Table 1 Experimental results of different methods on four real shared account sequence recommendation datasets (%) .

[0043] Experiment 2: The present invention also conducts an ablation experiment to verify the performance of the model LightGC²N of the present invention. Specifically, by deleting the test module, the model LightGC²N is subjected to an ablation experimental analysis. The present invention designs two variant models to respectively verify the extent to which the capsule graph convolutional network module and the sequence decoder module based on deep subspace clustering improve the performance of LightGC²N, and the symbol usage is shown in Table 2.

[0044] Table 2 Symbols of ablation experiment variants .

[0045] The same dataset and experimental indicator (i.e., MRR) as in Experiment 1 are used to compare their performance on four datasets. The experimental results are shown in Table 3: Table 3 Experimental results of ablation experiment variants (%) .

[0046] As can be seen from Table 3, the LightGC²N model achieved the best results on four different datasets, verifying the effectiveness of the graph capsule attention network and the deep subspace sequence decoder. Specifically, LightGC²N outperforms , proving that fine-grained differentiation of different potential user preferences does improve the model’s prediction performance on the shared account sequence recommendation task. In addition, LightGC²N outperforms , demonstrating the effectiveness of the deep subspace clustering module in aligning sequence representations with latent user preferences.

[0047] The above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by technicians in this technical field within the essential scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A sequence recommendation method for shared account scenarios, characterized in that: A recommendation model based on capsule graph convolutional network and deep subspace clustering is constructed for personalized recommendation. The recommendation model includes a main capsule graph construction module, a capsule graph convolutional network, an account-level dynamic routing module, a sequence decoder based on deep subspace clustering, and a prediction module. The method includes the following steps: Step 1: Obtain the interaction relationship between the shared account and the project and the sequence dependency relationship between the projects in the shared account scenario, and establish a sequence interaction diagram based on these two relationships; Step 2: The sequence interaction graph is further processed into a main capsule graph through the main capsule graph construction module; Step 3: Learn the vector representation of potential users in the shared account and the vector representation of the mixed sequence from the main capsule graph through the capsule graph convolutional network; Step 4: Aggregate the vector representations of potential users in the shared account into the vector representation of the shared account through the account-level dynamic routing module; Step 5: Align user interests with the vector representation of the mixed sequence through a sequence decoder based on deep subspace clustering; Step 6: Generate personalized recommendation results based on the vector representation of the shared account and the vector representation of the mixed sequence.

2. The sequence recommendation method for shared account scenarios according to claim 1, characterized in that: The specific process of step 1 is as follows: Get the project collection in the shared account scenario as ,in, Indicates projects, is the total number of projects; the shared account set is ,in, For the Shared accounts, is the total number of shared accounts; the mixed sequence set is ,in, Indicates A mixed sequence of shared accounts; The sequence interaction diagram established is ,in, Represents a collection of nodes, Represents a set of edges, each edge in the set represents a relationship between two different nodes in the sequence interaction graph; the sequence interaction graph contains two types of relationships, namely the interaction relationship between shared accounts and projects and the sequence dependency relationship between projects; the above two relationships are represented by the adjacency matrix express: (1)； in, represents the adjacency matrix that stores the sequential dependencies between items; if Projects In the mixed sequence, Projects The predecessor node of Middle Row, No. Elements of a column ,otherwise, ;No. Items corresponding to Row, No. Items corresponding to List; represents the adjacency matrix containing the interaction relationship between shared accounts and projects. If Shared accounts With Projects There is interaction between them, then Middle Row, No. The element corresponding to the column ,otherwise, ;No. The shared account corresponds to Row, No. Items corresponding to List; is the transpose symbol.

3. The sequence recommendation method for shared account scenarios according to claim 2, characterized in that: In step 2, the main capsule graph construction module includes a linear attention module and a one-dimensional convolution module; the specific working process of the main capsule graph construction module is: Step 2.1: Convert the feature representation of the items in the sequence interaction graph into the capsule representation of the items through the linear attention module. The specific process is as follows: (2)； in, The capsule representation of the item at layer 0 in the main capsule graph obtained after conversion. The number of layers in the main capsule graph corresponds to the number of layers in the sequence interaction graph. , They are the embedding dimensions of feature representation and capsule representation respectively; , , They are The embedded representation of the query, key, and value initialized by the value of It is the feature representation of the items in the 0th layer of the sequence interaction graph obtained by random initialization; is the softmax function; is the dimension transformation matrix of the linear attention module; is the bias term of the linear attention module; Step 2.2: Convert the feature representation of the shared account in the sequence interaction graph into the capsule representation of the shared account through a point-by-point one-dimensional convolution module. The specific calculation process is as follows: (3)； in, The capsule representation of the shared account at layer 0 in the main capsule graph obtained after conversion; is the feature representation of the shared account at layer 0 in the randomly initialized sequence interaction graph; Represents a one-dimensional convolution operation; Represents the one-dimensional convolution kernel of the one-dimensional convolution module; Represents the bias of the one-dimensional convolution module; is a hyperparameter that controls the number of potential users in a shared account; Step 2.3: Assume that each shared account contains potential users, and divide the capsule representation of each shared account into The capsule representation of a potential user is as follows: Shared accounts Capsules indicate Divide Capsule representation of potential users , is the first layer of the main capsule image after conversion. of shared accounts Potential users The capsule representation of the main capsule graph is completed while retaining the relationship in the sequence interaction graph.

4. The sequence recommendation method for shared account scenarios according to claim 3, characterized in that: In step 3, the capsule graph convolution network contains several graph convolution layers, each of which needs to be subjected to a graph convolution operation through a graph convolution network with an attention mechanism. The specific working process is as follows: Step 3.1: Calculate the The information obtained by the capsule representation of potential users during layer graph convolution is used to update the capsule representation of potential users. The specific process is as follows: Step 3.1.

1. Calculate the correlation between the capsule representations of potential users in the shared account and the capsule representations of items that have interacted with the shared account on the main capsule graph through the graph convolutional network with attention. The specific calculation process is as follows: (4)； in, for and The correlation between For the of shared accounts Potential users The capsule indicates; For the The shared account has interacted Projects The capsule indicates; is an exponential function with base e; Indicates Shared accounts The collection of items that have interacted; Step 3.1.2, calculate the The capsule of the project during the layer graph convolution represents the information conveyed to the potential user, specifically: (5)； in, For the When layer graph convolution The capsule indicates the information conveyed; Indicated in The first learnable weight during layer graph convolution; For the When the layer graph is convolved with The shared account has interacted Projects The capsule indicates; For the The second learnable weight during layer graph convolution; For the When layer graph convolution and The correlation between At the same time, calculation The self-connection information is as follows: (6)； in, For the When layer graph convolution Self-connection information; For the The third learnable weight during layer graph convolution; For the When layer graph convolution of shared accounts Potential users The capsule indicates; Step 3.1.3: The capsule indicates that an update is being performed. The update process is as follows: (7)； in, To conduct the Updated after layer graph convolution The capsule indicates; Indicates all The collection of items that have been interacted with; Step 3.2: Calculate the The capsule representation of the project obtains information during the layer graph convolution and updates the capsule representation of the project; the specific calculation process is as follows: (8)； (9)； in, Indicates shared account with Projects There is interaction potential users; Indicated in When layer graph convolution The capsule indicates The capsules represent the messages delivered; Indicated in The fourth learnable weight that controls the amount of information during layer graph convolution; For the When layer graph convolution The capsule indicates; Indicates that in a mixed sequence Adjacent projects; Indicated in When layer graph convolution is performed from The capsule indicates The capsules represent the messages delivered; Indicated in The fifth learnable weight that controls the amount of information during layer graph convolution; For the When layer graph convolution The capsule indicates; Step 3.3: The capsule representation update process of the project is as follows: (10)； in, To conduct the Updated after layer graph convolution The capsule indicates; Indicates all Potential users who have interacted A collection of; Indicates that in all mixed sequences A collection of adjacent items; Step 3.4: Use a multi-layer aggregation protocol to aggregate the representations obtained through graph convolution on each layer to obtain a complete item vector representation and potential user vector representation; the specific calculation process is as follows: (11)； (12)； in, For complete The vector representation of ; For complete The vector representation of ; is the total number of layers of graph convolution; Step 3.5: Calculate the vector representation of the final required mixed sequence. The specific calculation process is as follows: (13)； in, is the vector representation of the mixed sequence.

5. The sequence recommendation method for shared account scenarios according to claim 4, characterized in that: The specific process of step 4 is as follows: Step 4.1: Account-level dynamic routing module iterative calculation sequence The dynamic routing process of this iteration is as follows: (14)； in, Indicates At iteration Account-level capsule representation of ; Indicates the function of compressing routing information; For the The first iteration The coupling coefficient between the capsule representation of each potential user and the capsule representation of the account; Step 4.2: After completing a routing process, the coupling coefficient needs to be updated. The specific process is as follows: (15)； in, For the The first iteration The coupling coefficient between the capsule representation of each potential user and the capsule representation of the account; It is the learnable weight matrix in the account-level dynamic routing module; It is the element inner product operation; Step 4.3: After the iteration, the complete account capsule representation is obtained, i.e. At iteration Account-level capsule representation ; Step 4.4: Calculate the vector representation of the shared account. The specific calculation process is as follows: (16)； in, is the vector representation of the shared account.

6. The sequence recommendation method for shared account scenarios according to claim 5, characterized in that: The specific process of step 5 is as follows: Step 5.1: Initialize the subspace basis The affinity between the calculated sequence representation and the subspace basis is represented; different subspaces in the subspace basis represent the user interests of different potential users under the same shared account; Indicates subspaces; is the total number of subspaces; the specific calculation process of affinity is as follows: (17)； in, express Middle A vector representation of each item; express and The subspace affinity between them; represents the hyperparameter that controls the smoothness of the calculation; Step 5.2: Use contrastive learning method to align the vector representation of the affinity-enhanced mixed sequence with the vector representation of the original mixed sequence; the specific process is: First, subspace affinities are assigned to the vector representations of items in the mixed sequence: (18)； in, Indicates the enhanced A vector representation of items, A vector representation representing the enhanced mixed sequence; Then, the vector representation of the enhanced mixed sequence The vector representation of the original mixed sequence after regularization Compute the loss of contrastive learning between , the specific contrastive learning loss calculation process is as follows: (19)； in, , They are Middle Projects, The regularized embedding vector representation of each item; Indicates the enhanced A vector representation of each item; represents the temperature coefficient that controls the influence from negative pairs to positive pairs; Finally, for all mixed sequences, the complete contrastive learning loss is expressed as : (20)； Step 5.3: and The residual link method is used for refinement; the vector representation of the refined complete mixed sequence is obtained through the following calculation process: (21)； in, A vector representation of the complete mixed sequence after refinement; is the standard L2 regularization function; Represents the learnable weight matrix in the residual link method.

7. The sequence recommendation method for shared account scenarios according to claim 6, characterized in that: The specific process of step 6 is as follows: Step 6.1: and Input to the prediction module for prediction. The specific process is as follows: (22)； in, represents the predicted probability distribution, For the projects; It is the transformation matrix in the prediction module that maps the predicted value to the dimension of the candidate item set; is the bias term for adjusting the threshold of the activation function in the prediction module; Sort all items in descending order according to the size of the predicted probability distribution. The items are the personalized recommendation results for the shared account, is a preset threshold; Step 6.2: Use cross entropy loss to optimize the learnable parameters of the prediction model. The calculation process is: (23)； Step 6.3, add the contrastive learning loss in step 5.2 , the complete loss function of the prediction model is : (24)； in, is a hyperparameter that controls the degree of involvement of the self-supervision signal from the contrastive learning loss.

Images