POI recommendation method integrating social relationship network and spatiotemporal context information

Abstract

This invention belongs to the field of big data analysis technology, specifically to a POI recommendation method that integrates social relationship networks and spatiotemporal context information. To address the issues of insufficient consideration of the impact of social relationships on individual visit trajectories and ineffective extraction and integration of spatiotemporal context information features, this method subdivides user and time information features into multiple subfeatures, deeply exploring the relationships between these features. A regional user trajectory graph is constructed to model the POI-POI social relationship network. An encoder is used to comprehensively learn user and spatiotemporal context information features and perform a weighted fusion to capture users' dynamic preferences, thereby improving the accuracy of POI prediction.

Description

Technical Field

[0001] The present invention belongs to the technical field of big data analysis, and in particular relates to a POI recommendation method that integrates social relationship networks and spatiotemporal context information. Background Art

[0002] With the widespread use of mobile devices and the rapid development of internet technology, mobile internet has become an integral part of people's lives and work. This trend has led to an explosive growth in data volumes, bringing inconvenience to daily life. Against this backdrop, point of interest (POI) recommendation has become a hot topic in location-based social network research. It not only helps users quickly filter out locations of interest from a plethora of information, but also helps businesses find potential customers, thus possessing significant practical significance.

[0003] As a branch of recommender systems, POI recommendation initially relied primarily on tedious manual feature extraction and traditional machine learning-based methods, which limited both efficiency and recommendation results. In recent years, the application of deep learning to the POI recommendation task has become a clear trend. Deep learning methods, such as recurrent neural networks (RNNs), transformers, and graph neural networks (GNNs), can effectively model the complex relationships between users and POIs, automatically extract deep features from data, and capture the dynamic changes in user behavior and interests. Li et al. redefined social networks by incorporating features extracted from a user's friends' check-in information into a matrix factorization model, alleviating the cold start problem to some extent. Wang et al. proposed the CPAM recommendation algorithm, which employs an LSTM with a multi-level attention mechanism to capture user preferences, fully considering the different contextual weights of each check-in and the varying impact of each check-in on the next POI visited. Yu et al. proposed a socially aware, self-supervised triple training framework, SEPT, which augments the user data view with user social information and improves recommendations by generating self-supervisory signals from two complementary perspectives of the original data.

[0004] Existing research has achieved many results and progress, but there are still challenges that require further investigation and resolution: Traditional POI recommendation methods fail to consider the spatiotemporal characteristics of user check-in sequences and are unable to effectively model these sequences. POI recommendation methods based on user information features, when faced with very long user visit records, face a trade-off between global and local relationships during feature extraction. POI recommendation methods based on spatiotemporal information, when mining the spatiotemporal information implicit in user check-in data, still need to improve the quality of data processing and the comprehensiveness of feature extraction. Summary of the Invention

[0005] To address the problems of not fully considering the impact of social relationships on individual visit trajectories and not effectively extracting and integrating spatiotemporal context information features, the present invention provides a POI recommendation method that integrates social relationship networks and spatiotemporal context information.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions:

[0007] A POI recommendation method integrating social relationship network and spatiotemporal context information of the present invention comprises the following steps:

[0008] Step 1: Obtain user check-in sequence data, perform data preprocessing on it, and construct a regional user trajectory map;

[0009] Step 2: Input the user check-in sequence data pre-processed in step 1 and the constructed domain user trajectory graph into the pre-trained POI recommendation model to make the next POI recommendation; the POI recommendation model integrates the social relationship network and spatiotemporal context information, including a social relationship learning module, a POI spatiotemporal-category feature fusion module, an encoder and a decoder; the social relationship learning module performs feature aggregation and noise filtering on the regional user trajectory graph through a graph convolutional neural network integrated with a gated recurrent unit, learns the features of each POI, and obtains the feature vector E of each POI. p The POI spatiotemporal-category feature fusion module performs preliminary feature extraction and fusion on the POI spatiotemporal information of different dimensions in the user check-in sequence data through the time-space fusion module and the time-category fusion module to obtain the time-space fusion feature E of the user check-in activity. l,ts and time-category fusion feature E c,t The encoder extracts features from the user trajectory sequence through the user feature learning module and the spatiotemporal feature learning module to obtain the user feature Z u and spatiotemporal characteristics Z t ; The decoder is based on user feature Z u and spatiotemporal characteristics Z t ,The next POI, visit time period and POI category are predicted respectively through three multi-layer perceptrons.

[0010] Furthermore, the regional user trajectory graph in step 1 is defined as: a directed graph consisting of a node set V, an edge set E, a node attribute S, and a weight W. The node set V consists of the POIs that the user has historically visited. The edge set E is a set of directed edges. When a user continuously visits two POIs: p1 and p2 within a period of time, a directed edge e is added between p1 and p2. The weight W of each directed edge is equal to the number of times the sub-trajectory appears in the regional trajectory graph. The node attribute S includes the spatial coordinates, category, and visit frequency of each POI.

[0011] Furthermore, the specific process of the social relationship learning module includes:

[0012] Step 2.1: Calculate the normalized Laplacian matrix of the regional user trajectory graph adjacency matrix

[0013] Step 2.2: Use a single-layer graph convolutional neural network to normalize the Laplacian matrix Perform feature extraction;

[0014] Step 2.3: Update the hidden state of each node using a gated recurrent unit.

[0015] Step 2.4: After repeating steps 2.2 and 2.3 multiple times, the obtained feature vector is compared with the feature vector x extracted by the first single-layer graph convolutional neural network. t Perform splicing and fusion to obtain the feature vector E of each POI p , the calculation process is expressed as follows:

[0016]

[0017] Where H() represents a single-layer graph convolution operation, f T () represents a gated recurrent unit with T time steps, W G,C and b G,C denote weights and biases respectively, · denotes concatenation, is the Laplace matrix.

[0018] Furthermore, the specific process of the POI spatiotemporal-category feature fusion module includes:

[0019] 1) Time-space fusion module: Adopt the encoding strategy of CTLE model to obtain the timestamp feature vector E ts ; Input the coordinates of each POI into the fully connected layer to obtain the spatial feature vector E l ; The timestamp feature vector E ts and the spatial eigenvector E l Splicing is performed and input into the fully connected layer for feature fusion to obtain the time-space fusion feature E of the user's check-in activity l,ts , the calculation process is expressed as follows:

[0020] E l,ts =σ(W l,ts [E ts ·E l ]+b l,ts ) (2)

[0021] Where W l,ts represents the learnable weight vector, bl,ts represents the deviation, σ represents the nonlinear activation function LeakyReLU;

[0022] 2) Time-category fusion module: Adopt the encoding strategy of the time encoding model time2vector to encode the absolute access time and obtain the time feature vector E t ;Embed each POI category into a category feature vector E c Represented; the category feature vector E c and the time feature vector E t The splicing is then input into the fully connected layer for feature fusion to obtain the time-category fusion feature E of the user's check-in activity. c,t , the calculation process is expressed as follows:

[0023] E c,t =σ(W c,t [E c ·E t ]+b c,t ) (3)

[0024] Where W c,t represents the learnable weight vector, b c,t represents the bias, and σ represents the nonlinear activation function LeakyReLU.

[0025] Furthermore, the user trajectory sequence is defined as: the user's continuous check-in records are arranged in chronological order; after obtaining the time-space fusion feature representation E of the user's check-in activity, l,ts And time-category fusion feature representation E c,t Finally, the embedding representation of user check-in q is divided into two parts: and Each user's trajectory sequence Both are embedded by two user check-in sequences and express.

[0026] Furthermore, the specific process of the encoder includes:

[0027] 1) User feature learning module: Construct the user feature vector E of the target user u’s personal information through an embedding layer u , the user check-in data is constructed into two modes: graph-based user trajectory graph and sequence-based user check-in sequence;

[0028] For the user trajectory graph, a graph convolutional neural network with a gated recurrent unit is used for feature extraction, and the obtained POI feature vector is converted into and user feature vector E uSplicing, input into the fully connected layer for nonlinear transformation, thus obtaining the graph-based POI-User fusion feature vector The calculation process is expressed as follows:

[0029]

[0030] in, and represent weight and bias respectively, represents splicing, σ L Represents the nonlinear activation function LeakyReLU;

[0031] For user check-in sequences, a lightweight self-attention mechanism and channel convolution are used to capture global context information and local context information respectively;

[0032] To capture global context information, firstly transform the feature vector E obtained by the social relationship learning module into p and user feature vector E u Splicing, output sequence-based POI-User fusion feature vector Then, a simplified lightweight self-attention mechanism is used to obtain global information, which is input into the fully connected layer FC1 for nonlinear transformation to obtain the sequence-based POI-User global feature vector. The calculation process is expressed as follows:

[0033]

[0034] in, and Represent weights and biases respectively, Q, K, and V represent the input The same query weight matrix, key weight matrix and value weight matrix, d k represents the variance, T represents the transpose of the matrix, Represents the weight of the fully connected layer FC1, AvgPooling represents average pooling, and LayerNorm represents layer normalization;

[0035] To capture local context information, the sequence-based POI-User fusion feature vector The input is sent to the two-dimensional channel convolution layer conv2D to obtain local information, and then input to the fully connected layer FC2 for nonlinear transformation to obtain the sequence-based POI-User local feature vector The calculation process is expressed as follows:

[0036]

[0037] in, Represents the weight of the fully connected layer FC2;

[0038] After obtaining the graph-based POI-User fusion feature vector Sequence-based POI-User global feature vector and local eigenvectors After that, the user feature learning module outputs the user feature Z u , defined by the weighted sum of these three eigenvectors, and introducing two aggregation hyperparameters α1 and α2 to reasonably distribute the fusion ratio. The calculation process is expressed as follows:

[0039]

[0040] Among them, σ t Represents the activation function tanh;

[0041] 2) Spatiotemporal feature learning module: Given a user trajectory sequence Embedding representation of two user check-in sequences and Will and As input tensors, they are respectively passed into the spatiotemporal feature learning module to capture spatiotemporal features of different dimensions. The module has a multi-layer structure. For the pth layer, the input and Transformed by a multi-head self-attention layer, the calculation process is expressed as follows:

[0042]

[0043] Among them, W Q , W K , W V Represent the query weight matrix, key weight matrix and value weight matrix respectively, E ij represents the correlation between the i-th and j-th check-ins, α ij Indicates the influence weight of other check-in activities on the current check-in;

[0044] The output Head obtained by different attention heads i The layers are stacked and the representations from different attention spaces are merged through linear transformation. The encoder uses layer normalization LayerNorm and residual connection technology. The calculation process is expressed as follows:

[0045]

[0046] After the multi-head self-attention layer, the feedforward neural network FNN and activation function ReLU are used to give the model nonlinearity. The calculation process is expressed as follows:

[0047]

[0048] Where W1 and W2 represent weights, and b1 and b2 represent biases, so the output of the encoder layer p is expressed as follows:

[0049]

[0050] After stacking the multi-head self-attention blocks multiple times, the output spatiotemporal feature Z of the spatiotemporal feature learning module is t Output features from time-space and time-category output features The weighted sum definition of , introduces two aggregation hyperparameters β1 and β2 to reasonably distribute the fusion ratio. The calculation process is expressed as follows:

[0051]

[0052] Among them, σ t Represents the activation function tanh, p * Represents the last layer of the spatiotemporal feature learning module.

[0053] Furthermore, the decoder uses three multi-layer perceptrons to predict the next POI, visit time period, and POI category, respectively, as defined in the following equations:

[0054]

[0055] in, Indicates the next POI, Indicates the access time period. represents the POI category, W poi 、W time 、W cat Represent their respective weights, b poi 、b time 、b cat Represent their respective deviations.

[0056] Going further, it also includes the training of the POI recommendation model, and its loss function is defined as follows:

[0057]

[0058] Where, represents the overall loss function of the POI recommendation model, and Represent the loss functions for the next POI and POI category prediction, using cross entropy loss. For performance evaluation of access time prediction, the mean square error (MSE) is used.

[0059] Compared with the prior art, the present invention has the following advantages:

[0060] The method of the present invention is suitable for recommending the next POI that a user is interested in. By constructing a regional user trajectory graph, while filtering out the noise of individual unique preferences, a social relationship network is modeled based on the correlation between different POIs, and the check-in data information of all users is fully utilized. At the same time, the method uses graph-based and sequence-based models to deeply learn user information features, and uses relative visit time difference and absolute visit time to mine POI spatiotemporal-category information features. The method comprehensively and fully processes, utilizes and adaptively integrates the check-in data of target users from multiple dimensions. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a schematic diagram of the regional user trajectory map of the present invention;

[0062] Figure 2 is a schematic diagram of the present invention integrating the social relationship network and spatiotemporal context information model;

[0063] Figure 3 Schematic diagram of the gated graph convolutional neural network of the present invention;

[0064] Figure 4 is a schematic diagram of an encoder module of the present invention;

[0065] Figure 5 Schematic diagram of the present invention in dealing with the new user cold start test in the NYC dataset;

[0066] Figure 6 It is a schematic diagram of the present invention's response to the new user cold start test in the TKY dataset. DETAILED DESCRIPTION

[0067] In order to further illustrate the technical solution of the present invention, the present invention is further described below through examples.

[0068] A POI recommendation method integrating social relationship network and spatiotemporal context information in this embodiment includes the following steps:

[0069] Step 1: Obtain user check-in sequence data, perform data preprocessing on it, and construct a regional user trajectory map;

[0070] The regional user trajectory diagram of this embodiment is as follows Figure 1 As shown, it is defined as: a directed graph consisting of a node set V, an edge set E, a node attribute S and a weight W. The node set V consists of the POIs that the user has visited historically. The edge set E is a set of directed edges. When a user visits two POIs: p1 and p2 continuously within a period of time, a directed edge e is added between p1 and p2. The weight W of each directed edge is equal to the number of times the sub-trajectory appears in the regional trajectory graph. The node attribute S contains the spatial coordinates, category and visit frequency of each POI.

[0071] Specifically, the FourSquare dataset is used as the dataset for the POI recommendation task that integrates social relationship networks and spatiotemporal context information in the present invention. The FourSquare-NYC dataset was collected in New York from April 2012 to February 2013, and the FourSquare-TKY dataset was collected in Tokyo during the same period. Each check-in record contains information such as user id, POI id, POI category, GPS coordinates, and timestamp. The first 80% is taken as the training set, and the remaining 20% ​​is used as the test set. POIs and users with less than 10 check-in records are deleted; the check-in sequence data of each user is divided into tracks with an interval of 24 hours. If a track contains only one check-in record, the track data is deleted.

[0072] Step 2: Input the user check-in sequence data preprocessed in step 1 and the constructed domain user trajectory graph into the pre-trained POI recommendation model (GSTC) to perform the next POI recommendation.

[0073] The POI recommendation model of this embodiment is as follows Figure 2 As shown in Figure 3, this model integrates social relationship networks and spatiotemporal context information, including a social relationship learning module, a POI spatiotemporal-category feature fusion module, an encoder, and a decoder.

[0074] Among them, the social relationship learning module: through the graph convolutional neural network integrated with the gated recurrent unit, the regional user trajectory graph is subjected to feature aggregation and noise filtering, the features of each POI are learned, and the feature vector E of each POI is obtained. p ;

[0075] like Figure 3 As shown in Figure 2, the specific process of the social relationship learning module includes:

[0076] Step 2.1: Calculate the normalized Laplacian matrix of the regional user trajectory graph adjacency matrix

[0077] Step 2.2: Use a single-layer graph convolutional neural network to normalize the Laplacian matrix Perform feature extraction;

[0078] Step 2.3: Update the hidden state of each node using a gated recurrent unit.

[0079] Step 2.4: Repeat steps 2.2 and 2.3 multiple times to obtain the eigenvector h t The feature vector x extracted by the first single-layer graph convolutional neural network t (x1) is spliced ​​and fused to obtain the feature vector E of each POI p , the calculation process is expressed as follows:

[0080]

[0081] Where H() represents a single-layer graph convolution operation, f T () represents a gated recurrent unit with T time steps, W G,C and b G,C denote weights and biases respectively, · denotes concatenation, is the Laplace matrix.

[0082] POI spatiotemporal-category feature fusion module: The time-space fusion module and the time-category fusion module are used to perform preliminary feature extraction and fusion on the POI spatiotemporal information of different dimensions in the user check-in sequence data, and the time-space fusion feature E of the user check-in activity is obtained. l,ts and time-category fusion feature E c,t ;

[0083] 1) Time-space fusion module: Adopt the encoding strategy of CTLE (Context and Time Aware Location Embeddings) model to obtain the timestamp feature vector E ts ; Input the coordinates of each POI into the fully connected layer to obtain the spatial feature vector E l ; The timestamp feature vector E ts and the spatial eigenvector E l Splicing is performed and input into the fully connected layer for feature fusion to obtain the time-space fusion feature E of the user's check-in activity l,ts , the calculation process is expressed as follows:

[0084] E l,ts =σ(W l,ts [E ts ·E l ]+b l,ts ) (2)

[0085] Where W l,ts represents the learnable weight vector, b l,ts represents the deviation, σ represents the nonlinear activation function LeakyReLU;

[0086] 2) Time-category fusion module: Adopt the encoding strategy of the time encoding model time2vector to encode the absolute access time and obtain the time feature vector E t ;Embed each POI category into a category feature vector E c Represented; the category feature vector E c and the time feature vector E tThe splicing is then input into the fully connected layer for feature fusion to obtain the time-category fusion feature E of the user's check-in activity. c,t , the calculation process is expressed as follows:

[0087] E c,t =σ(W c,t [E c ·E t ]+b c,t ) (3)

[0088] Where W c,t represents the learnable weight vector, b c,t represents the bias, and σ represents the nonlinear activation function LeakyReLU.

[0089] Encoder: The user feature learning module and the spatiotemporal feature learning module are used to extract the features of the user trajectory sequence and obtain the user feature Z u and spatiotemporal characteristics Z t ;

[0090] The user trajectory sequence is defined as: the user's continuous check-in records within 48 hours are arranged in chronological order; after obtaining the time-space fusion feature representation E of the user's check-in activity, l,ts And time-category fusion feature representation E c,t Finally, the embedding representation of user check-in q is divided into two parts: and Each user's trajectory sequence Both are embedded by two user check-in sequences and express.

[0091] The structure of the encoder is as follows Figure 4 As shown, the specific process includes:

[0092] 1) User feature learning module: Construct the user feature vector E of the target user u’s personal information through an embedding layer u , the user check-in data is constructed into two modes: graph-based user trajectory graph and sequence-based user check-in sequence;

[0093] For the user trajectory graph, a graph convolutional neural network with a gated recurrent unit is used for feature extraction, and the obtained POI feature vector is converted into and user feature vector E u Splicing, input into the fully connected layer for nonlinear transformation, thus obtaining the graph-based POI-User fusion feature vector The calculation process is expressed as follows:

[0094]

[0095] in, and represent weight and bias respectively, represents splicing, σ L Represents the nonlinear activation function LeakyReLU;

[0096] For user check-in sequences, a lightweight self-attention mechanism and channel convolution are used to capture global context information and local context information respectively;

[0097] To capture global context information, firstly transform the feature vector E obtained by the social relationship learning module into p and user feature vector E u Splicing, output sequence-based POI-User fusion feature vector Then, a simplified lightweight self-attention mechanism is used to obtain global information, which is input into the fully connected layer FC1 for nonlinear transformation to obtain the sequence-based POI-User global feature vector. The calculation process is expressed as follows:

[0098]

[0099] in, and Represent weights and biases respectively, Q, K, and V represent the input The same query weight matrix, key weight matrix and value weight matrix, d k represents the variance, T represents the transpose of the matrix, Represents the weight of the fully connected layer FC1. The encoder uses the softmax function to ensure that the sum of the attention weights is 1. At the same time, to avoid overfitting, average pooling and layer normalization are used. AvgPooling represents average pooling, and LayerNorm represents layer normalization.

[0100] To capture local context information, the sequence-based POI-User fusion feature vector The input is sent to the two-dimensional channel convolution layer conv2D to obtain local information, and then input to the fully connected layer FC2 for nonlinear transformation to obtain the sequence-based POI-User local feature vector The calculation process is expressed as follows:

[0101]

[0102] in, Represents the weight of the fully connected layer FC2;

[0103] After obtaining the graph-based POI-User fusion feature vector Sequence-based POI-User global feature vector and local eigenvectors After that, the user feature learning module outputs the user feature Z u , defined by the weighted sum of these three eigenvectors, in order to improve the adaptive feature fusion capability of the model, two aggregation hyperparameters α1 and α2 are introduced to reasonably distribute the fusion ratio. The calculation process is expressed as follows:

[0104]

[0105] Among them, σ t Represents the activation function tanh;

[0106] 2) Spatiotemporal feature learning module: Given a user trajectory sequence Embedding representation of two user check-in sequences and Will and As input tensors, they are respectively passed into the spatiotemporal feature learning module to capture spatiotemporal features of different dimensions. The module has a multi-layer structure. For the pth layer, the input and Transformed by a multi-head self-attention layer, the calculation process is expressed as follows:

[0107]

[0108] Among them, W Q , W K , W V Represent the query weight matrix, key weight matrix and value weight matrix respectively, E ij represents the correlation between the i-th and j-th check-ins, α ij Indicates the influence weight of other check-in activities on the current check-in. The encoder uses the softmax function to ensure that the sum of attention weights is 1;

[0109] The output Head obtained by different attention heads i The layers are stacked and the representations from different attention spaces are merged through linear transformation. In order to avoid overfitting and gradient disappearance problems, the encoder uses layer normalization LayerNorm and residual connection technology. The calculation process is expressed as follows:

[0110]

[0111] After the multi-head self-attention layer, the feedforward neural network FNN and activation function ReLU are used to give the model nonlinearity. The calculation process is expressed as follows:

[0112]

[0113] Where W1 and W2 represent weights, and b1 and b2 represent biases, so the output of the encoder layer p is expressed as follows:

[0114]

[0115] After stacking the multi-head self-attention blocks multiple times, the spatiotemporal feature learning module outputs the spatiotemporal feature Z t , output features from time-space and time-category output features To improve the adaptive feature fusion capability of the model, two aggregation hyperparameters β1 and β2 are introduced to reasonably distribute the fusion ratio. The calculation process is expressed as follows:

[0116]

[0117] Among them, σ t Represents the activation function tanh, p * Represents the last layer of the spatiotemporal feature learning module.

[0118] Decoder: User feature Z based on user context information u and spatiotemporal characteristics Z t ,The next POI, visit time period and POI category are predicted respectively through three multi-layer perceptrons;

[0119] The decoder uses three multi-layer perceptrons to predict the next POI, visit time period, and POI category, respectively, as defined in the following formula:

[0120]

[0121] in, Indicates the next POI, Indicates the access time period. represents the POI category, W poi 、W time 、W cat Represent their respective weights, b poi 、b time 、b cat Represent their respective deviations.

[0122] In addition, this method also includes the training of the POI recommendation model, and its loss function is defined as follows:

[0123]

[0124] Where, represents the overall loss function of the POI recommendation model, and Represent the loss functions for the next POI and POI category prediction, respectively, using the common cross entropy loss. For performance evaluation of access time prediction, the mean square error (MSE) is used.

[0125] Specifically, the Adam optimizer with an initial learning rate of 0.0001 is used to train the network. The batch size is set to 16, and the network is trained for 150 epochs. The results are obtained through multiple tests. See Tables 1 and 2 for details.

[0126] Table 1 Overall performance comparison of different methods on the NYC dataset

[0127]

[0128] Table 2 Overall performance comparison of different methods on the TKY dataset

[0129]

[0130] Accuracy Acc@k and mean reciprocal ranking MRR are used as measurement indicators on the FourSquare dataset. Tables 1 and 2 compare the measurement indicators of the method of the present invention and other methods for POI recommendation on the NYC and TKY datasets. It can be seen from Tables 1 and 2 that the method of the present invention outperforms other methods in accuracy Acc@k and mean reciprocal ranking MRR. Compared with GETNext, the method of the present invention improves Acc@1 by 2.02%, Acc@5 by 2.99%, Acc@10 by 2.32%, Acc@20 by 2.81%, and MRR by 2.34% in the NYC dataset; in the TKY dataset, Acc@1 improves by 0.85%, Acc@5 improves by 2.02%, Acc@10 improves by 2.68%, Acc@20 improves by 5.22%, and MRR improves by 1.38%.

[0131] The effectiveness of the POI recommendation network GSTC of the present invention is verified by changing the input user trajectory length n. Figure 5 To investigate the effect of user trajectory length n on Acc@10 in the NYC dataset, Figure 6Figure 1 shows the effect of the input user trajectory length n on Acc@10 in the TKY dataset. To test the model's performance in dealing with the cold start of new users, the value of n was set from 2 to 12 to simulate different sparsity levels of user check-in data, while keeping other parameters unchanged. The trend in the figure shows that when n = 2, the GSTC model significantly outperforms the baseline model in accuracy on the NYC and TKY datasets, and the larger the dataset, the more obvious the lead. As the input trajectory length continues to increase, the recommendation effect of the existing model shows an overall upward trend. This shows that the GSTC model can effectively utilize the POI-POI social relationship network modeled by the gated graph convolutional neural network to provide more accurate recommendations for target users, thereby alleviating the cold start problem of new users.

[0132] The foregoing shows and describes the principal features and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the present invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the present invention is defined by the appended claims, not the foregoing description, and all variations that come within the meaning and range of equivalents of the claims are intended to be embraced therein.

[0133] In addition, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A POI recommendation method integrating social relationship network and spatiotemporal context information, characterized in that: The following steps are involved: Step 1: Obtain user check-in sequence data, perform data preprocessing on it, and construct a regional user trajectory map; Step 2: Input the user check-in sequence data pre-processed in step 1 and the constructed domain user trajectory graph into the pre-trained POI recommendation model to make the next POI recommendation; the POI recommendation model integrates the social relationship network and spatiotemporal context information, including a social relationship learning module, a POI spatiotemporal-category feature fusion module, an encoder and a decoder; the social relationship learning module performs feature aggregation and noise filtering on the regional user trajectory graph through a graph convolutional neural network integrated with a gated recurrent unit, learns the features of each POI, and obtains the feature vector of each POI. The POI spatiotemporal-category feature fusion module performs preliminary feature extraction and fusion on the POI spatiotemporal information of different dimensions in the user check-in sequence data through the time-space fusion module and the time-category fusion module to obtain the time-space fusion feature of the user check-in activity. and time-category fusion features The encoder extracts features from the user trajectory sequence through the user feature learning module and the spatiotemporal feature learning module to obtain user features. and spatiotemporal characteristics ; The user feature learning module: constructs the user feature vector of the target user u’s personal information through an embedding layer , the user check-in data is constructed into two modes: the user trajectory graph based on graph and the user check-in sequence based on sequence; for the user trajectory graph, a graph convolutional neural network fused with gated recurrent units is used for feature extraction, and the obtained POI feature vector is converted into and user feature vector Splicing, input into the fully connected layer for nonlinear transformation, thus obtaining the graph-based POI-User fusion feature vector ; For the user check-in sequence, a lightweight self-attention mechanism and channel convolution are used to capture global context information and local context information respectively; for capturing global context information, the feature vector obtained by the social relationship learning module is first and user feature vector Splicing, output sequence-based POI-User fusion feature vector , and then use a simplified lightweight self-attention mechanism to obtain global information, input it into the fully connected layer FC1 for nonlinear transformation, and obtain the sequence-based POI-User global feature vector ; To capture local context information, the sequence-based POI-User fusion feature vector The input is sent to the two-dimensional channel convolution layer conv2D to obtain local information, and then input to the fully connected layer FC2 for nonlinear transformation to obtain the sequence-based POI-User local feature vector ; Based on POI-User fusion feature vector , POI-User global feature vector and local eigenvectors , get user features ; The decoder uses three multi-layer perceptrons to predict the next POI, visit time period, and POI category, respectively, as defined in the following equations: ； ； ； in, Indicates the next POI, Indicates the access time period. Indicates the POI category, 、 、 Represent their respective weights, 、 、 Represent their respective deviations.

2. The POI recommendation method integrating social relationship network and spatiotemporal context information according to claim 1, characterized in that: The regional user trajectory graph in step 1 is defined as: a directed graph consisting of a node set V, an edge set E, node attributes S, and a weight W. The node set V consists of the POIs that the user has visited historically, and the edge set E is a set of directed edges. When a user visits two POIs consecutively within a period of time: and , then in and A directed edge e is added between them, the weight W of each directed edge is equal to the number of times the trajectory appears in the regional trajectory graph, and the node attribute S includes the spatial coordinates, category and visit frequency of each POI.

3. The POI recommendation method integrating social relationship network and spatiotemporal context information according to claim 1, characterized in that: The specific process of the social relationship learning module includes: Step 2.1: Calculate the normalized Laplacian matrix of the regional user trajectory graph adjacency matrix ; Step 2.2: Use a single-layer graph convolutional neural network to normalize the Laplacian matrix Perform feature extraction; Step 2.3: Update the hidden state of each node using a gated recurrent unit. Step 2.4: Repeat steps 2.2 and 2.3 multiple times and compare the obtained feature vector with the feature vector extracted by the first single-layer graph convolutional neural network. Perform splicing and fusion to obtain the feature vector of each POI , the calculation process is expressed as follows: ； Where, represents a single-layer graph convolution operation, represents a gated recurrent unit of T time steps, and denote weights and biases respectively, Indicates splicing, is the Laplace matrix.

4. The POI recommendation method integrating social relationship network and spatiotemporal context information according to claim 3, characterized in that: The specific process of the POI spatiotemporal-category feature fusion module includes: 1) Time-space fusion module: adopts the encoding strategy of the CTLE model to obtain the timestamp feature vector ; Input the coordinates of each POI into the fully connected layer to obtain the spatial feature vector ; The timestamp feature vector and spatial eigenvectors Splicing is performed and input into the fully connected layer for feature fusion to obtain the time-space fusion features of the user's check-in activity , the calculation process is expressed as follows: ； Where, represents the learnable weight vector, Indicates deviation, Represents the nonlinear activation function Leaky ReLU; 2) Time-category fusion module: Adopts the encoding strategy of the time encoding model time2vector to encode the absolute access time and obtain the time feature vector ;Embed each POI category into a category feature vector Representation; the category feature vector and time feature vector The splicing is then input into the fully connected layer for feature fusion to obtain the time-category fusion feature of the user's check-in activity , the calculation process is expressed as follows: ； Where, represents the learnable weight vector, Indicates deviation, Represents the nonlinear activation function Leaky ReLU.

5. The POI recommendation method integrating social relationship network and spatiotemporal context information according to claim 4, characterized in that: The user trajectory sequence is defined as: consisting of the user's continuous check-in records in chronological order; obtaining the time-space fusion feature representation of the user's check-in activity And time-category fusion feature representation Finally, the embedding representation of user check-in q is divided into two parts: and ; Trajectory sequence of each user Both are embedded by two user check-in sequences and express.

6. The POI recommendation method integrating social relationship network and spatiotemporal context information according to claim 5, characterized in that: The specific process of the encoder includes: The POI-User fusion feature vector The calculation process is shown as follows: ； in, and represent weight and bias respectively, Indicates splicing, Represents the nonlinear activation function LeakyReLU; The POI-User global feature vector The calculation process is shown as follows: ； ； ； in, and Represent weights and biases respectively, Q, K, and V represent the input The same query weight matrix, key weight matrix, and value weight matrix, represents the variance, T represents the transpose of the matrix, Represents the weight of the fully connected layer FC1, AvgPooling represents average pooling, and LayerNorm represents layer normalization; The POI-User local feature vector The calculation process is shown as follows: ； ； in, Represents the weight of the fully connected layer FC2; The user characteristics POI-User fusion feature vector , POI-User global feature vector and local eigenvectors The weighted sum of is defined and two aggregation hyperparameters are introduced and The fusion ratio is reasonably distributed, and the calculation process is shown in the following formula: ； in, Represents the activation function tanh; The spatiotemporal feature learning module: given a user trajectory sequence Embedding representation of two user check-in sequences and ,Will and As input tensors, they are respectively passed into the spatiotemporal feature learning module to capture spatiotemporal features of different dimensions. The module has a multi-layer structure. For the pth layer, the input and Transformed by a multi-head self-attention layer, the calculation process is expressed as follows: ； ； ； in, , , Represent the query weight matrix, key weight matrix and value weight matrix respectively, represents the correlation between the i-th and j-th check-ins, Indicates the influence weight of other check-in activities on the current check-in; The outputs obtained by different attention heads The layers are stacked and the representations from different attention spaces are merged through linear transformation. The encoder uses layer normalization LayerNorm and residual connection technology. The calculation process is expressed as follows: ； After the multi-head self-attention layer, the feedforward neural network FNN and activation function ReLU are used to give the model nonlinearity. The calculation process is expressed as follows: ； in, 、 represents the weight, 、 Represents the deviation, so the output of the encoder layer p is expressed as follows: ； After stacking the multi-head self-attention blocks multiple times, the spatiotemporal feature learning module outputs the spatiotemporal features , output features from time-space and time-category output features The weighted sum definition of , introduces two aggregation hyperparameters and The fusion ratio is reasonably distributed, and the calculation process is expressed as follows: ； in, represents the activation function tanh, Represents the last layer of the spatiotemporal feature learning module.

7. A POI recommendation method integrating social relationship network and spatiotemporal context information according to any one of claims 1 to 6, characterized in that: It also includes the training of the POI recommendation model, and its loss function is defined as follows: ； Where, represents the overall loss function of the POI recommendation model, and Represent the loss functions for the next POI and POI category prediction, using cross entropy loss. For performance evaluation of access time prediction, the mean square error (MSE) is used.

Images