Next Article in Journal
Energy Consumption Analysis of ISO/IEC 29192-2 Standard Lightweight Ciphers
Previous Article in Journal
A Thermodynamic Model for the Solubility of SO2 in Multi-Ion Electrolyte Solutions and Its Applications
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Temporal Enhancement of Top-N Recommendation on Heterogeneous Graphs

1
School of Network and Communication, Nanjing Vocational College of Information Technology, Nanjing 210023, China
2
School of Electronic Science and Engineering, Nanjing University, Nanjing 210023, China
*
Author to whom correspondence should be addressed.
Appl. Sci. 2025, 15(7), 3929; https://doi.org/10.3390/app15073929
Submission received: 9 February 2025 / Revised: 17 March 2025 / Accepted: 18 March 2025 / Published: 3 April 2025

Abstract

:
Heterogeneous information networks (HINs) have seen rapid development and have attracted extensive attention because of their effectiveness in recommender systems. Although many existing models based on HINs for recommender systems have obtained good recommendation performance on account of their superior ability to process heterogeneous data and capture rich semantic information, there are still several problems. Firstly, the temporal relations among different nodes in meta-paths, which include users and items, are rarely considered in HINs. Secondly, the interactions among meta-paths, users, and items are similarly often overlooked. Thirdly, their ability to learn the heterogeneous information of users and items is limited. In view of the above problems, we propose a system for the temporal enhancement of top-N recommendations on HINs called TMRec. Specifically, we first adopted long short-term memory (LSTM) and residual self-attention (RSA) to process users and items and enhance the network’s ability to both learn the heterogeneous information in them and capture the temporal relations among them. Second, we designed a novel method for processing meta-paths, including deep perception self-attention (DPSA), max pooling, and L2-normalization, that can effectively obtain the temporal relations among different nodes in meta-paths. Third, we used collaborative attention to process meta-paths, users, and items to obtain their interactions. Finally, extensive experiments were conducted on four public datasets of recommender systems to verify the superiority of our method compared with state-of-the-art top-N recommendation models.

1. Introduction

Recommender systems, especially top-N recommender systems, play a key role in our daily life [1]. Traditional models of recommender systems recommend interest items to different users based on historical feedback information [2,3] based on methods such as collaborative filtering [2]. Recently, more and more auxiliary data have also been applied in recommender systems for improving their performance [4,5]. However, the complexity and heterogeneity of auxiliary data pose a big challenge for their use in recommender systems.
Today, graph neural network (GNN)-based models [6,7,8,9,10] have achieved significant success in recommender systems. Specifically, in a GNN, first, the interactions of users, items, and auxiliary information are considered to be a graph structure. Then, the information representation of users and items is obtained with an information propagation mechanism. For example, SVD++ [6] enriches the representation of users by incorporating implicit feedback on items, in addition to explicit interactions. ItemRank [7] constructs an item-to-item graph based on the interaction history, where the edges in the graph represent the similarity among items. It then uses a random walk algorithm to obtain the probability of a user visiting each item, which is used to rank the items. However, these networks cannot effectively capture complex heterogeneous information. Consequently, heterogeneous information networks (HINs) [11,12], which include nodes and links, have been proposed for recommender systems due to their effective ability to acquire the semantic and representation information of nodes. The nodes in HINs include users and items, while links are meta-paths representing the relations among nodes. Many HIN-based methods [13,14,15] have been proposed, such as hierarchical attention network (HAN) [13] and meta-path-based context for top-N recommendation (MCRec) [16].
Although many existing models based on HINs for recommender systems have shown better performance due to their superior ability to process heterogeneous data and capture rich semantic information, they still face several problems. Firstly, although some researchers have focused on the explicit information of meta-paths, the proposed methods do not consider the temporal relations among different nodes in meta-paths [17]; this is mainly due to the fact that the static embedding methods used in such studies are subject to limitations and time dependency is often not taken into account. For example, users’ browsing and purchasing behaviors on e-commerce platforms exhibit a certain order over time. The temporal relations among different nodes can help capture the sequence of these behaviors, thereby allowing for a better prediction of what content users may be interested in next. Secondly, although in HINs the relations between nodes are considered, the interactions between nodes and meta-paths are not accounted for [18]. Thirdly, HINs cannot effectively learn the heterogeneous information of nodes [16]. For example, a movie recommender system includes various entity types, such as users, movies, directors, and actors, and the relationships among them exhibit high heterogeneity.
In view of these problems, a system for the temporal enhancement of top-N recommendations on HINs called TMRec is proposed in this study. It is mainly based on two architectures: long short-term memory (LSTM) [19], which can remember important information and forget irrelevant details by controlling the flow of information, and residual self-attention (RSA) [20], which can dynamically focus on the most important parts of information, rather than processing all input information uniformly. Therefore, we adopt these two methods for learning heterogeneous information among users or items, as well as effectively capturing the temporal relations among users or items. Specifically, LSTM focuses on the temporal relations between nodes of the same type, such as users or items, and RSA combines multiple sandwich networks and gated self-attention networks, where each sandwich network includes convolutional neural networks (CNNs) [21] and linear rectification functions. Further, RSA can effectively reduce the problems of information loss and overfitting to enhance the ability to learn relationships among nodes of the same type. In addition, we introduce a deep perception self-attention (DPSA) model with max pooling and L2-normalization to process meta-paths. Compared with the previous methods, it can effectively capture the temporal relations among different nodes in meta-paths. DPSA is a novel self-attention network. Unlike traditional self-attention models, it adopts deep convolutional neural networks (DCNNs) with different convolutional kernels and a residual network (ResNet) to obtain the query, key, and value. In DPSA, the DCNNs with different convolutional kernels can learn from multiple perspectives, thereby improving the performance of the model. ResNet can effectively address the problems of vanishing and exploding gradients, thereby enhancing the model’s learning capability. Their combination can enhance the perceptual ability for different nodes in meta-paths [22]. Finally, we adopt the collaborative attention model to connect the learning results of meta-paths, users, and items to obtain the interactions between nodes and meta-paths. The contributions of our work are summarized as follows:
(1)
We propose a system for the temporal enhancement of top-N recommendations on HINs called TMRec. It can improve the performance of recommender systems by learning the temporal relations of nodes and meta-paths.
(2)
We adopt the architectures of LSTM and RSA for embedding the learning of users and items, which can highlight the temporal relations and interactions among users or items. This can also enhance the ability to learn heterogeneous information from these nodes.
(3)
We use DPSA with max pooling and L2-normalization to learn the interactions among users and items in meta-paths. Additionally, collaborative attention is introduced to obtain the interactions of meta-paths, users, and items.
(4)
Extensive experiments are conducted on four public datasets of recommender systems. The results demonstrate that the performance of the proposed TMRec model is largely improved compared with other SOTA models.
The structure of this paper is organized as follows: In Section 2, we introduce related work on recommender systems, heterogeneous information networks, and network embedding. In Section 3, we introduce the definition of a heterogeneous information network and the main notations used. In Section 4, we explain the framework of the proposed TMRec in detail. In Section 5, we present the experiments conducted and analyze their results, which demonstrate the effectiveness of our model. Section 6 includes the conclusion of our study.

2. Related Work

In this section, related studies associated with our work are presented, with the topics including recommender systems, heterogeneous information networks, and network embedding.

2.1. Recommender Systems

Recommender systems are critical to preventing an information explosion by choosing suitable items for one specific user according to their interaction history and preferences. The two scenarios of recommender systems are rating prediction and top-N recommendation. The latter involves presenting information to users in the form of a recommendation list; for example, the paper recommendations in Academic Toutiao, the music recommendations in Xiami, and the course recommendations in NetEase Cloud Classroom are all top-N recommendations. In early studies, recommendations were always conducted with collaborative filtering (CF) methods, which utilize historical interactions [23]. CF-based models for recommender systems include item-based k-nearest neighbor (ItemKNN) [2], Bayesian personalized ranking (BPR) [3], matrix factorization (MF) [24], and neural collaborative filtering (NCF) [25]. Additional information, including social information [26], location information [27], and heterogeneous information [12], is leveraged in many works to solve the cold-start problems of CF-based methods. Moreover, some universal frameworks, which are intended to combine context information, have been proposed [25]. Recently, owing to the success of deep learning (DL) architectures, researchers have employed deep models such as convolutional neural networks [28], autoencoders [29], and multi-layer perceptrons (MLPs) to obtain latent features from data. Further, meta-paths have been widely employed in recommendation models because of their rich semantic information. Shi et al. [22] proposed a heterogeneous information network for recommendations that constructs homogeneous sequences with a type constraint and filtering in meta-paths. Moreover, Hu et al. [16] proposed MCRec for recommendations, a model which explicitly combines meta-path-based context and designs a three-way neural interaction. Finally, the efficient neighborhood-based interaction model, NIRec, can capture and aggregate rich interaction patterns at the node and path levels through convolution operations for recommendation [30].

2.2. Heterogeneous Information Networks and Network Embedding

Heterogeneous information networks (HINs), also known as heterogeneous graphs, which are made up of different types of nodes and edges, have attracted extensive attention and developed rapidly since they are ubiquitous in the real world. Research studies on learning representations in heterogeneous networks are emerging in classification [31], clustering [32], link prediction [33], and recommendation [34]. Although HIN-based recommender systems have achieved good performance, the embedding representation is always high-dimensional and sparse. Instead, we need a low-dimensional and dense representation for downstream tasks [35]. Network embedding allows one to obtain low-dimensional and dense vector representations for users and items for the understanding of the semantic meaning among nodes. DeepWalk is a classical network embedding method proposed for the learning of representations with the strategy of combining random walk and skip-gram [36]. Likewise, Node2vec, an improvement on DeepWalk, adopts a biased random walk method to replace the previous random walk generation method to more flexibly learn information [37]. However, these models are only applied in homogeneous networks. At present, several works have been conducted to explore representations in heterogeneous networks. For instance, HIN2vec learns the embedding representations of various nodes and meta-paths in a heterogeneous graph with diverse methods of prediction training [38]. Xu et al. [39] first proposed a representative model which encodes the edges of HINs, achieving good results. Heterogeneous graph neural network (HetGNN) encodes the heterogeneous contents of various kinds of nodes and aggregates them by category [40].

3. Preliminaries

In this section, we introduce the definition of a heterogeneous information network and the main notations used.
Recently, more and more attention has been paid to heterogeneous graphs of the type G = ( V , E ) , where V and E represent node sets and edge sets, respectively. The node mapping function α : V A denotes the sets of node types; likewise, the edge mapping function β : E B denotes the sets of edge types, with A + B > 2 . As a promising network, it can model complex objects and links, which correspond to different types of nodes and edges in the graph, respectively.
HINs can expose underlying connections by establishing a logical network, which allows one node on a heterogeneous graph to transfer feedback information to another connection node. We present an example of an HIN for recommendation in Figure 1. Bob interacts with Titanic through the path u s e r m o v i e t y p e m o v i e , which is called a meta-path in the HIN. Here, it is the relationship of specific users, films, genres, and directors, and it is mainly used to combine the features of nodes and the relations among them.
Next, the main notations used in this study and their descriptions are summarized in Table 1.

4. The Proposed Model

4.1. Overall Framework

In this study, we propose a recommender system called TMRec for the determination of the temporal relation of each sequence to enhance the embeddings. The proposed model can be divided into three parts: user embedding representation learning, item embedding representation learning, and meta-path embedding learning. Firstly, LSTM can remember important information and forget irrelevant details by controlling the flow of information. Moreover, RSA, which includes multiple sandwich networks and gated self-attention networks, can dynamically focus on the most important parts of information. Therefore, we adopt LSTM and RSA to process users and items, respectively, as they can effectively capture the heterogeneous and temporal relations among users and items. Secondly, DPSA has better learning ability for different nodes in meta-paths on account of its deep convolutional neural networks with different convolutional kernels and a residual network within the self-attention mechanism; therefore, we use it with max pooling and L2-normalization to process the meta-paths for determining the heterogeneous and temporal relations among nodes of different types in the data to improve the representations. Thirdly, collaborative attention is used for enhancing the ability to obtain the interactions of users, items, and meta-paths. The overall framework of the proposed TMRec is described in Figure 2.
Specifically, we acquire the embeddings of users, items, and meta-paths and combine them as the final embedding to perform a recommendation task. It is worth mentioning that the embeddings of the users and items are generated in the same way, and we take user embeddings as an example to describe the framework. After a lookup layer is employed to obtain low-dimensional and dense vectors, we utilize the LSTM module to process the data and the RSA module to determine the importance of each node and learn more meaningful embeddings. As for the meta-path embeddings, we use priority scores to generate high-quality path instances and aggregate them into a meta-path embedding with the DPSA module to improve the representation. In addition, we use L2-normalization to process the embedding of a single meta-path to avoid overfitting. Then, the final meta-path embedding is obtained with a co-attention mechanism in the case where user and item embeddings are used as input. Finally, we concatenate the above three embeddings into a representation for top-N recommendations.

4.2. User and Item Embeddings

First, we utilize a lookup layer to transform the representations of users and items into low-dimensional and dense vectors. Here, we define a user’s representation and an item’s representation as m u R U × 1 and n i R I × 1 , respectively. In the lookup layer, there are two transformation matrices specific for users and items, namely, M R U × d and N R I × d , respectively, where the numbers of users and items are denoted by U and I , respectively, and d is the dimension of each embedding. The low-dimensional representations of users and items are given as follows:
x u = M T m u
y i = N T n i
The LSTM architecture and the self-attention mechanism are utilized for user and item embedding learning to account for the importance of inner interactions and mutual influence, which are often not considered. LSTM is employed for solving the problem of gradient disappearance and gradient explosion in long-sequence training. There are three gates in LSTM: the input gate, the forget gate, and the output gate. Moreover, there is a “processor”, called a cell, in every gate. When a message enters the LSTM network, it must be judged as useful or not according to the rules. Only the information that meets the criteria is retained, while inconsistent information is forgotten with the forget gate. Taking user embeddings as an example, LSTM can be formulated as follows:
x u = LSTM ( x u , θ )
where all the related parameters are expressed as θ in LSTM. The specific representations of LSTM are shown in the formulas below.
(a) The input state vector of the t-th user in the user sequence is shown as follows:
C ^ t = tanh ( W c [ h t 1 , x t ] + b c )
where tanh is the hyperbolic tangent activation function, ⊙ denotes the Hadamard product, h t 1 denotes the hidden layer output vector of the ( t 1 ) -th user in the user sequence, x t is the low-dimensional embedding representation of the t-th user in the sequence, and W c and b c denote the learnable parameter and the bias vector of the input state, respectively.
(b) The input vector of the t-th user in the user sequence is as follows:
i t = σ ( W i [ h t 1 , x t ] + b i )
where σ is the activation function, and W i and b i denote the learnable parameter and the bias vector of the input gate, respectively.
(c) The forget gate vector of the t-th user in the user sequence is as follows:
f t = σ ( W f [ h t 1 , x t ] + b f )
where W f and b f denote the learnable parameter and the bias vector of the forget gate, respectively.
(d) The overall state vector of the t-th user in the user sequence is as follows:
C t = f t · C t 1 + i t · C ^ t
where C t 1 denotes the overall state vector of the ( t 1 ) -th user in the user sequence and · denotes the pointwise multiplication operator.
(e) The output vector of the t-th user in the user sequence is as follows:
o t = σ ( W 0 [ h t 1 , x t ] + b 0 )
where W 0 and b 0 denote the learnable parameter and the bias vector of the output gate, respectively.
(f) The hidden layer output vector of the t-th user in the user sequence is as follows:
h t = o t · tanh ( C t )
Researchers have introduced self-attention into recommender systems in some HIN-based methods based on the attention mechanism of the natural language neighborhood. However, traditional self-attention has some limitations, such as information loss and overfitting. Therefore, we adopt RSA for processing users and items. Specifically, RSA includes two sandwich networks and gated self-attention (GSA) networks for solving the problem of information loss. At the same time, we use residual connections to solve the problem of overfitting. The process of RSA is shown as follows:
(a) We use the first sandwich network to process x u and obtain the result, which is shown as follows:
x ^ u = C N N ( R e L U ( C N N ( x u ) ) )
where C N N ( · ) is the convolutional neural network and R e L U ( · ) is the rectified linear unit.
(b) We use a residual connection to combine x ^ u and x u and feed them into the second sandwich network. The result can be obtained with the following formula:
x ˇ u = x ^ u + x u + C N N ( R e L U ( C N N ( x ^ u + x u ) ) )
(c) We use a residual connection to combine x u and x ˇ u and feed them into the GSA module. The result can be obtained with the following formula:
x ˜ u = x ˇ u + x u + G S A ( x ˇ u )
where G S A ( · ) is the gated self-attention network, whose formula is shown as follows:
G S A ( x ˇ u ) = S i g m o i d ( F C ( x ˇ u ) ) + t a n h ( F C ( C N N ( x ˇ u ) C N N ( x ˇ u ) T d F C ( x ˇ u ) ) )
where S i g m o i d ( · ) and t a n h ( · ) are nonlinear activation functions and F C ( · ) is a fully connected layer.
Because the process for items is the same as that for users, we can obtain y ˜ i as follows:
y ˜ i = y ˇ u + y u + G S A ( y ˇ u )

4.3. Meta-Path Embedding

Meta-path embedding learning aims at learning the explicit representation of meta-paths to improve mutual interactions. The key point of this representation is to generate path instances according to their importance. In this part, a temporal convolutional network module is introduced to capture the temporal relations and reliance of the nodes in each meta-path instance sequence.

4.3.1. Sampling of High-Quality Path Instances

Different meta-paths mean different semantic and interactive relationships between different types of nodes. There are several meta-paths in Figure 1. For instance, meta-path UMTM is a sequence of a user node, a movie node, a type node, and a movie node, which indicates that people fall in love with movies of the same type. Path-based similarity measures are used in HINs to enhance the representations of nodes and links [41]. Moreover, nodes can be reached with different meanings through meta-paths. Given a meta-path UMTM in a heterogeneous information network, a new term, meta-path instance u 1 m 1 t 1 m 2 , is defined as a sequence consisting of various nodes following the architecture of meta-path UMTM. Specifically, Bob likes watching Roman Holiday, which is a romance, and he is very likely interested in Titanic, which is of the same type; then, B o b R o m a n H o l i d a y R o m a n c e T i t a n i c is a meta-path instance. Multiple meta-path instances of this form can make up a meta-path.
The embedding of meta-path P is obtained from the embedding aggregation of path instances; thus, their high-quality sampling is necessary. However, path instance sampling approaches, such as meta-path-based random walk, are inefficient, as they affect meta-path embedding generation. In a path, a source node should be linked to a node of higher priority in each step. Traditional paths obtained with the random walk method are usually composed of nodes that may not be highly related, so they may not have rich semantic information. Therefore, in order to distinguish the different priorities of different nodes, a priority score based on node similarity is presented. In such a case, nodes are more closely related if their priority score is high. Specifically, we first utilize the singular value decomposition (SVD) feature to learn the latent vector representation of each node [42,43]. Secondly, we calculate the priority score between connected nodes for each path instance. Then, we perform uniform averaging or summation operations. Next, we compare and sort the final priority scores of all path instances. Finally, we choose the path instances with the top-M scores to generate meta-path embeddings.

4.3.2. Single Meta-Path Embedding

After obtaining high-quality meta-path instances, we learn path instance embeddings and aggregate them to generate a meta-path embedding. Considering the sequence characteristics of meta-paths, especially the temporal logic, DPSA, which is an improved self-attention model, is utilized. It is a powerful tool that can be run in parallel on multiple paths and process sequence-type data. Compared with traditional self-attention, DPSA has the following improvements: (1) it uses a deep CNN (DCNN) to obtain the values of query and key, which further includes several CNNs with different kernels, and (2) it uses ResNet to obtain the value of self-attention. These improvements can help to enhance the perceptual ability for meta-paths and obtain the temporal relations among different nodes in a meta-path. The formula of DPSA is shown as follows:
h p = S o f t m a x ( D C N N ( E p ) D C N N ( E p ) T ) R e s N e t ( E p )
where E p is the matrix of meta-path instance p and D C N N ( · ) is the deep CNN, whose formula is shown as follows:
D C N N ( E p ) = C N N 1 ( E p ) + C N N 3 ( E p ) + C N N 5 ( E p ) + C N N 7 ( E p )
where C N N 1 ( · ) , C N N 3 ( · ) , C N N 5 ( · ) and C N N 7 ( · ) are CNNs with different kernel sizes.
Moreover, R e s N e t ( · ) is the residual network, whose formula is shown as follows:
R e s N e t ( E p ) = E p + B N ( R e L U ( F C ( B N ( R e L U ( F C ( E p ) ) ) ) ) )
where B N ( · ) denotes batch normalization.
In addition, because a meta-path is composed of path instances, the max-pooling operation needs to be applied to such instances as follows:
c p = max-pooling ( h p ) p = 1 M
where M is the number of path instances, and max-pooling is the operation performed on M meta-path instances to obtain the most important embedding features among them.
After obtaining the embedding of a single meta-path, we introduce L2-normalization to process it and, importantly, prevent overfitting:
c p = L 2 ( c p )
where L2 denotes the normalization operation, which is a common method for training deep learning models and solving overfitting problems.

4.3.3. Meta-Path Aggregation

Different meta-paths may have different importance because every user may have their own preferences. An individual user is also likely to make different choices and have different interactions with other entities. A good meta-path embedding includes the semantics of each meta-path and accurately distinguishes them. Therefore, the introduction of an attention mechanism is critical. Considering the impact of users and items on the importance of different meta-paths, we consider enhancing the representation of meta-path embeddings by inputting user and item embeddings into the attention mechanism. This structure is called a collaborative attention mechanism. It differs from self-attention precisely in that it considers not only the interaction and influence within each meta-path but also the influence of users and items on the meta-path. Since different meta-paths have different semantics and importance, we introduce a two-layer neural network for the calculation of the attention scores, which are given as follows:
λ u , i , p ( 1 ) = ReLU ( w u ( 1 ) x u + w i ( 1 ) y i + a ( 1 ) )
λ u , i , p ( 2 ) = ReLU ( w ( 2 ) λ u , i , p ( 1 ) + a ( 2 ) )
where w u ( 1 ) and w i ( 1 ) are the weight matrices for users and items, respectively; a ( 1 ) is the bias vector in the first layer; w ( 2 ) and a ( 2 ) are the weight vector and the bias vector, respectively, in the second layer; and ReLU is the activation function.
The final attention weights can be acquired by normalizing the attention scores as follows:
λ u , i , p = λ u , i , p ( 2 ) a M u , i λ u , i , a ( 2 )
Finally, we obtain the meta-path embedding with the co-attention mechanism as follows:
c u , i = p M u , i λ u , i , p · c p

4.4. Embedding Fusion

After having obtained improved user, item, and meta-path embeddings, we combine the three embeddings to obtain the final vector representation of the overall interaction, as shown in the following formula:
r u , i = MLP ( x ˜ u y ˜ i c u , i )
where ⊕ denotes the concatenation operation, and the MLP component is implemented in two hidden layers with ReLU as the activation function and in the output layer of the model with the Sigmoid function. Training an efficient model requires the establishment of a suitable optimization objective function. The objective function of traditional system models for top-N recommendation is the square error loss function. Based on our model, the objective function can be formulated as follows:
σ u , i = log r u , i E n p n e g [ log ( 1 log r u , n ) ]
Here, the first term of expression (24) is the modeled interaction, and the second term denotes the negative feedback generated by noise distribution p n e g , which denotes a uniform distribution of noise in the model.

5. Experiments

The experimental settings, indicators, and results of the proposed TMRec model for top-N recommendations are introduced in detail in this section.

5.1. Datasets and Metrics

We adopted three datasets widely used for recommender systems in different areas: the Movielens dataset for movies, the LastFM dataset for music, and the Yelp dataset for business. The Movielens dataset contains movie rating data, which include movie ratings, movie metadata (genre and era), and demographic data about the user (age, zip code, gender, occupation, etc.); in this dataset, there are 943 users, 1682 movies, and 18 genres. As for the LastFM dataset, it includes a list of users’ most popular artists and the number of times the latter have been played; specifically, there are 1892 users, 17,632 artists, and 11,945 tags. Finally, in the Yelp dataset, there are 16,239 users, 14,267 businesses, and 47 cities. In addition, we also used DUNN to further demonstrate the effectiveness of the proposed TMRec. DUNN [18] is a real-world dataset comprising offline transactions spanning a two-year period from around 2000 households. It has 2500 users, 16,780 items, and 269,974 transactions.
We adopted the widely used metrics of recall at rank-10 ( R e c a l l @ 10 ), normalized discounted cumulative gain at rank-10 ( N D C G @ 10 ), precision at rank-10 ( P r e c @ 10 ), and F-score to quantitatively evaluate the proposed TMRec model. It is worth noting that the higher the metrics, the better the model performance. We averaged the results of 10 experiments to obtain more accurate experimental results.

5.2. Implementation Details

We performed the experiments on the proposed TMRec model with the Python libraries Keras and Tensorflow. The implementation details of our experiments are shown as follows: Before training a neural network, its full value and bias must be initialized. For our experiments, the Gaussian distribution and adaptive moment estimation (Adam) were used. The former was used to initialize model parameters randomly, and the latter, which distributes the same learning rate to each parameter and adapts independently as learning progresses, was utilized to optimize the model. Additionally, we selected appropriate meta-path instances in our datasets: U M U M , U M G M , U U U M and U M M M in Movielens and U A T A , U A U A , U U U A and U U A in LastFM; the number of path instances was 5. In addition, we determined a basis for selecting appropriate hyperparameter values for the TMRec model with extensive experiments. Specifically, when training the network, the batch size, the regularization parameter, and the learning rate were set to 256, 0.001, and 0.001, respectively. In addition, we set the layers of LSTM to 2. We selected a 1 × 1 CNN, a 3 × 3 CNN, a 5 × 5 CNN, and a 7 × 7 CNN as the main components of the DCNN. Lastly, the activation function of the hidden layer was the ReLU function, the number of layers (k) was 5, and the corresponding dimensions of each layer were 512, 256, 128, 64 and 32. For all experiments, we divided the dataset into 80% for training and the remaining 20% for testing with the cross-validation method. In addition, we performed experiments on comparison models from previous studies; their hyperparameter values were set as indicated in the respective publications. Moreover, the split into 80% and 20% of the dataset for training and testing was also applied for these comparison models. All the experimental results, conducted in the same machine under the same conditions, are the averages of 10 experiments.

5.3. Experimental Results Analysis

In this subsection, we compare the proposed TMRec model with several other methods for top-N recommendation. The comparison models can be divided into two categories: CF-based models, including ItemKNN, BPR, MF, and NCF, and HIN-based models, including the meta-paths HAN, MCRec, AMERec, and MEGNN.
(1)
ItemKNN [2]: It is a classical method of collaborative filtering using previous items as the reference. For example, if one wants to predict u 1 ’s rating on movie m 1 , one first needs to keep a list of k movies that u 1 has watched and evaluated and then give a prediction of the rating of m 1 by u 1 according to the historical ratings.
(2)
BPR [3]: It is a commonly used recommendation model in current recommender systems. Unlike other methods based on a user scoring matrix, it mainly adopts the users’ implicit feedback (such as clicks and favorites) to sort items through the maximum posterior probability obtained by the Bayesian analysis of problems and then generates recommendations.
(3)
MF [24]: It is proposed to solve the shortcomings of CF, which has a weak ability to process sparse matrices. It optimizes the MF model with the cross-entropy loss.
(4)
NCF [25]: It is a variant of the traditional collaborative filtering model. Since the previous inner product of collaborative filtering is too simple to perform well, it utilizes an MLP to replace this operation, combining neural networks and collaborative filtering.
(5)
HAN [13]: It adopts a hierarchical structure of the attention mechanism, including node-level and semantic-level attention. The former is used to distinguish the importance of nodes in a meta-path, while the latter is utilized to learn how much each meta-path contributes to the generation of the final embedding.
(6)
MCRec [16]: It is the first method to explicitly express meta-paths, which are combined with the user and item embeddings for recommendation. It also introduces the co-attention mechanism to mutually enhance the representations of meta-paths, users, and items.
(7)
GraFC2T2 [17]: It is a general graph-based framework that combines some side information for top-N recommendation. It encodes content-based features, and temporal and trust information into a complex graph and uses personalized PageRank on this graph to provide recommendations.
(8)
MEGNN [44]: It is capable of discovering and extracting the most expressive meta-paths and avoids manually defining multiple meta-paths. It also uses a heterogeneous convolution module to generate trainable heterogeneous graph structures.
The results of the proposed model and baselines on three datasets are shown in Table 2.
The major findings from Table 2 are reported below.
(1)
The proposed TMRec consistently outperforms all baselines on the three datasets. The results show the effectiveness of TMRec in top-N recommendation tasks; the model considers the temporal relations of nodes of different kinds and the mutual influence of nodes of the same type in user and item embedding learning.
(2)
It can be clearly seen that when comparing the two different types of methods, HIN-based methods perform better than CF-based methods in general. As HIN-based methods can catch richer semantic information, they are particularly useful in addressing cold-start problems in recommendation. Therefore, they have been more widely used in recommendation tasks, especially when there are many different types of nodes and links in a dataset.
(3)
TMRec outperforms HAN, MCRec, and MEGNN among the HIN-based methods. HAN utilizes a hierarchical attention mechanism at the node and semantic levels, while MCRec employs a co-attention mechanism for the user, item, and meta-path embeddings. In addition, MEGNN considers the selection of meta-paths without considering the explicit expression of the meta-paths. In contrast, the proposed TMRec model uses co-attention in meta-path embedding learning and self-attention in user and item embedding learning. From the above results, we can find that the co-attention mechanism is more suitable for paths composed of different types of nodes, while the self-attention mechanism is more useful for paths composed of nodes of the same type. In addition, although GraFC2T2 considers the temporal information among different entities, its ability to capture heterogeneous information is limited. Therefore, the proposed TMRec achieves better performance than GraFC2T2 because of important components such as LSTM, RSA, and DPSA.
(4)
We can see from Table 2 that Prec@10, Recall@10, and NDCG@10 are increased by 0.0406, 0.0395, and 0.0448, respectively, compared with the average values of the comparison models on the LastFM dataset and by 0.0359, 0.0323, and 0.0412, respectively, on the Movielens dataset. These experimental results demonstrate that the proposed TMRec has better performance in these scenarios. In contrast to these two datasets, Prec@10, Recall@10 and NDCG@10 are increased by 0.0359, 0.0323, and 0.0412, respectively, compared with the average values of the comparison models on the Yelp dataset. Although TMRec’s advantage over the other models is relatively lower on the Yelp dataset, it still exhibits better recommendation performance.
In order to further demonstrate the effectiveness of the proposed TMRec, we also carried out some experiments on the DUNN dataset on TOP-10 by using the comparison models. Additionally, we used the F-score to evaluate their performance on TOP-10. The experimental results are shown in Figure 3.
We can see from Figure 3 that the F-score of the proposed TMRec model is 0.8097. It achieves the highest F-score, indicating that it has the best performance among the comparison models. In addition, it is evident that the reduction in the standard deviation from ItemKNN to TMRec reflects an improvement in the stability of the proposed model. This is particularly important in real-world applications where consistent performance is crucial.
In order to show the advantage in time complexity of the proposed TMRec, we investigated the training and inference time of MCRec, GraFC2T2, MEGNN, and TMRec, as shown in Figure 4.
We can see from Figure 4 that the proposed TMRec has a lower time cost than GraFC2T2 and MEGNN, which demonstrates its advantage in time complexity. In addition, although TMRec has a higher time cost than MCRec, it shows better performance. In summary, the experimental results indicate that TMRec has better time complexity and superior performance.

5.4. Ablation Study

Ablation studies were conducted to evaluate the performance and systematically analyze the effectiveness of the TCN, LSTM, and self-attention modules and included (1) TMRec-DPSA: a variant of TMRec without the DPSA module; (2) TMRec-LSTM: a variant of TMRec without the long short-term memory module; and (3) TMRec-RSA: a variant of TMRec without RSA in the user and item embeddings. The results of the ablation studies are presented in Table 3.
We can find from Table 3 that the three improvements clearly have positive effects on recommendation performance. Further, LSTM and RSA in user and item embedding learning play a more important role in TMRec than DPSA. They enhance meta-path embeddings by capturing the temporal relations of internal nodes. In addition, DPSA in meta-path embedding learning is helpful in determining the mutual influence and dependence in a path consisting of nodes of the same type, which also enhances user and item embeddings. In summary, all the modules we introduced in TMRec are important for recommendation performance.

6. Conclusions

In this study, we propose a model for temporally enhanced top-N recommendation on HINs called TMRec for effectively capturing the temporal relations and heterogeneous information of users, items, and meta-paths. Firstly, we adopted long short-term memory (LSTM) and residual self-attention (RSN) to process users and items and enhance the system’s ability to both learn the heterogeneous information in them and capture the temporal relations among them. Secondly, we designed a novel method for processing meta-paths, which includes deep perception self-attention (DPSA), max-pooling, and L2-normalization. It can effectively obtain the temporal relations among different nodes in meta-paths. Thirdly, we use collaborative attention to process meta-paths, users, and items to obtain their interactions. Finally, extensive experiments were conducted on three public datasets of recommender systems to prove the superiority of our method compared with state-of-the-art top-N recommendation models. Although TMRec can improve the performance of recommender systems, most meta-paths are manually selected, which may hinder the application of the model on different tasks and datasets. In the future, we will conduct research on how to automatically generate meta-paths and introduce this strategy into the recommendation task to further improve performance. For example, we could automatically capture path instances between entities that have different semantics. In addition, we have not considered some important auxiliary information for recommendation, so we will develop a method to obtain much more auxiliary information to boost performance.

Author Contributions

F.H. developed the algorithm and designed the experiments; F.H. performed the experiments; F.H. and J.L. wrote the paper. All authors have read and agreed to the published version of the manuscript.

Funding

This research study was funded by the National Natural Science Foundation of China (Nos. 62174084 and 62341408).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Acknowledgments

We would like to thank Nanjing University for providing the necessary technical support.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Wang, Y.; Ma, W.; Zhang, M. A survey on the fairness of recommender systems. ACM Trans. Inf. Syst. 2023, 41, 52. [Google Scholar]
  2. Kabul, M.S.; Setiawan, E.B. Recommender System with User-Based and Item-Based Collaborative Filtering on Twitter using K-Nearest Neighbors Classification. J. Comput. Syst. Inform. 2022, 3, 478–484. [Google Scholar] [CrossRef]
  3. Paul, A.; Wu, Z.; Liu, K.; Gong, S. Robust multi-objective visual bayesian personalized ranking for multimedia recommendation. Appl. Intell. 2022, 52, 3499–3510. [Google Scholar]
  4. He, Z.; Hui, B.; Zhang, S.; Xiao, C.; Zhong, T.; Zhou, F. Exploring indirect entity relations for knowledge graph enhanced recommender system. Expert Syst. Appl. 2023, 213, 11898. [Google Scholar]
  5. Qu, Y.; Fang, B.; Zhang, W.; Tang, R.; Niu, M.; Guo, H.; Yu, Y.; He, X. Product-based neural networks for user response prediction over multi-field categorical data. ACM Trans. Inf. Syst. 2018, 37, 5. [Google Scholar]
  6. Koren, Y. Factorization meets the neighborhood: A multifaceted collaborative filtering model. In Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Las Vegas, NV, USA, 24–27 August 2008; pp. 426–434. [Google Scholar]
  7. Gori, M.; Pucci, A.; Roma, V.; Passerini, A.; Frasconi, P. Itemrank: A random-walk based scoring algorithm for recommender engines. In Proceedings of the International Joint Conferences on Artificial Intelligence, Hyderabad, India, 6–12 January 2007; pp. 2766–2771. [Google Scholar]
  8. Ma, H.; Zhou, D.; Liu, C.; Li, Y.; King, I.; Lyu, M.R. Recommender systems with social regularization. In Proceedings of the Fourth ACM International Conference on Web Search and Data Mining, Hong Kong, China, 9–12 February 2011; pp. 287–296. [Google Scholar]
  9. Lu, Y.; Dong, R.; Smyth, B. Coevolutionary recommendation model: Mutual learning between ratings and reviews. In Proceedings of the 2018 World Wide Web Conference, Lyon, France, 23–27 April 2018; pp. 773–782. [Google Scholar]
  10. Yu, F.; Cui, L.; Guo, W.; Lu, X.; Li, Q.; Lu, H. A category-aware deep model for successive POI recommendation on sparse check-in data. In Proceedings of the Web Conference 2020, Virtual, 20–24 April 2020; pp. 1264–1274. [Google Scholar]
  11. Liu, J.; Song, L.; Wang, G.; Shang, X. Meta-HGT: Metapath-aware HyperGraph Transformer for heterogeneous information network embedding. Neural Netw. 2023, 157, 65–76. [Google Scholar]
  12. Han, Z.; Jin, X.; Xing, H.; Yang, W.; Xiong, H. An effective heterogeneous information network representation learning framework. Future Gener. Comput. Syst. 2023, 148, 66–78. [Google Scholar]
  13. Wang, X.; Bo, D.; Shi, C.; Fan, S. A survey on heterogeneous graph embedding: Methods, techniques, applications and sources. IEEE Trans. Big Data 2022, 9, 415–436. [Google Scholar]
  14. Wang, X.; Ji, H.; Shi, C.; Wang, B.; Cui, P.; Yu, P.; Ye, Y. Heterogeneous graph attention network. In Proceedings of the The World Wide Web Conference, San Francisco, CA, USA, 13–17 May 2019; pp. 2022–2032. [Google Scholar]
  15. Chen, J.; Gong, Z.; Li, Y.; Zhang, H.; Yu, H.; Zhu, J.; Fan, G.; Wu, X.-M.; Wu, K. Meta-path based neighbors for behavioral target generalization in sequential recommendation. IEEE Trans. Netw. Sci. Eng. 2022, 9, 1658–1667. [Google Scholar]
  16. Hu, B.; Shi, C.; Zhao, W.X.; Yu, P.S. Leveraging meta-path based context for top-n recommendation with a neural co-attention model. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, London, UK, 19–23 August 2018; pp. 1531–1540. [Google Scholar]
  17. Nzeko’o, A.; Tchuente, M.; Latapy, M. A general graph-based framework for top-N recommendation using content, temporal and trust information. J. Interdiscip. Methodol. Issues Sci. 2019, 5, 5553. [Google Scholar]
  18. Xu, E.; Zhao, K.; Yu, Z.; Zhang, Y. Limits of predictability in top-N recommendation. Inf. Process. Manag. 2024, 61, 103731. [Google Scholar]
  19. Duan, J.; Zhang, P.F.; Qiu, R.; Huang, Z. Long short-term enhanced memory for sequential recommendation. World Wide Web 2023, 26, 561–583. [Google Scholar] [CrossRef]
  20. Wu, W.; Liu, S.; Xia, Y.; Zhang, Y. Dual residual attention network for image denoising. Pattern Recognit. 2024, 149, 110291. [Google Scholar] [CrossRef]
  21. Chen, M.; Ma, T.; Zhou, X. CoCNN: Co-occurrence CNN for recommendation. Expert Syst. Appl. 2022, 195, 116595. [Google Scholar]
  22. Shi, C.; Hu, B.; Zhao, W.X.; Yu, P.S. Heterogeneous information network embedding for recommendation. IEEE Trans. Knowl. Data Eng. 2018, 31, 357–370. [Google Scholar]
  23. Koren, Y.; Bell, R.; Volinsky, C. Matrix factorization techniques for recommender systems. Computer 2009, 42, 30–37. [Google Scholar] [CrossRef]
  24. Koren, Y.; Rendle, S.; Bell, R. Advances in collaborative filtering. In Recommender Systems Handbook; Springer: New York, NY, USA, 2021; pp. 91–142. [Google Scholar]
  25. He, X.; Chua, T.S. Neural factorization machines for sparse predictive analytics. In Proceedings of the International ACM SIGIR Conference on Research and Development in Information Retrieval, Tokyo, Japan, 7–11 August 2017; pp. 355–364. [Google Scholar]
  26. Wu, L.; Sun, P.; Fu, Y.; Hong, R.; Wang, X.; Wang, M. A neural influence diffusion model for social recommendation. In Proceedings of the International ACM SIGIR Conference on Research and Development in Information Retrieval, Paris, France, 21–25 July 2019; pp. 235–244. [Google Scholar]
  27. Yin, H.; Sun, Y.; Cui, B.; Hu, Z.; Chen, L. LCARS: A location-content-aware recommender system. In Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Chicago, IL, USA, 11–14 August 2013; pp. 221–229. [Google Scholar]
  28. Zheng, L.; Noroozi, V.; Yu, P.S. Joint deep modeling of users and items using reviews for recommendation. In Proceedings of the ACM International Conference on Web Search and Data Mining, Cambridge, UK, 6–10 February 2017; pp. 425–434. [Google Scholar]
  29. He, R.; McAuley, J. VBPR: Visual bayesian personalized ranking from implicit feedback. Proc. AAAI Conf. Artif. Intell. 2016, 30, 144–150. [Google Scholar] [CrossRef]
  30. Jin, J.; Qin, J.; Fang, Y.; Du, K.; Zhang, W.; Yu, Y.; Zhang, Z.; Smola, A.J. An efficient neighborhood-based interaction model for recommendation on heterogeneous graph. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Virtual, 23–27 August 2020; pp. 75–84. [Google Scholar]
  31. Ji, M.; Sun, Y.; Danilevsky, M.; Han, J.; Gao, J. Graph regularized transductive classification on heterogeneous information networks. In Proceedings of the Joint European Conference on Machine Learning and Knowledge Discovery in Databases, Barcelona, Spain, 20–24 September 2010; pp. 570–586. [Google Scholar]
  32. Sun, Y.; Aggarwal, C.C.; Han, J. Relation Strength-Aware Clustering of Heterogeneous Information Networks with Incomplete Attributes. 2012. Available online: https://arxiv.org/abs/1201.6563 (accessed on 9 February 2025).
  33. Chen, H.; Yin, H.; Wang, W.; Wang, H.; Nguyen, Q.V.H.; Li, X. PME: Projected metric embedding on heterogeneous networks for link prediction. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, London, UK, 19–23 August 2018; pp. 1177–1186. [Google Scholar]
  34. Chen, T.; Sun, Y. Task-guided and path-augmented heterogeneous network embedding for author identification. In Proceedings of the ACM International Conference on Web Search and Data Mining, Cambridge, UK, 6–10 February 2017; pp. 295–304. [Google Scholar]
  35. Zhou, S.; Bu, J.; Wang, X.; Chen, J.; Wang, C. HAHE: Hierarchical Attentive Heterogeneous Information Network Embedding. 2019. Available online: https://arxiv.org/abs/1902.01475 (accessed on 9 February 2025).
  36. Li, J.; Wu, L.; Hong, R.; Hou, J. Random walk based distributed representation learning and prediction on social networking services. Inf. Sci. 2021, 549, 328–346. [Google Scholar]
  37. Grover, A.; Leskovec, J. node2vec: Scalable feature learning for networks. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, 13–17 August 2016; pp. 855–864. [Google Scholar]
  38. Fu, T.; Lee, W.C.; Lei, Z. Hin2vec: Explore meta-paths in heterogeneous information networks for representation learning. In Proceedings of the ACM on Conference on Information and Knowledge Management, Singapore, 6–10 November 2017; pp. 1797–1806. [Google Scholar]
  39. Xu, L.; Wei, X.; Cao, J.; Yu, P.S. Embedding of embedding (EOE) joint embedding for coupled heterogeneous networks. In Proceedings of the ACM International Conference on Web Search and Data Mining, Cambridge, UK, 6–10 February 2017; pp. 741–749. [Google Scholar]
  40. Zhao, J.; Wang, X.; Shi, C.; Hu, B.; Song, G.; Ye, Y. Heterogeneous graph structure learning for graph neural networks. In Proceedings of the AAAI Conference on Artificial Intelligence, Virtual, 2–9 February 2021; Volume 35, pp. 4697–4705. [Google Scholar]
  41. Liu, J.; Shi, C.; Hu, B.; Chen, K.; Zheng, Z.; Yu, Y. Personalized ranking recommendation via integrating multiple feedbacks. In Advances in Knowledge Discovery and Data Mining; Springer: Berlin, Germany, 2017; pp. 131–143. [Google Scholar]
  42. Chen, T.; Zhang, W.; Lu, Q.; Chen, K.; Zheng, Z.; Yu, Y. SVDFeature: A toolkit for feature-based collaborative filtering. J. Mach. Learn. Res. 2012, 13, 3619–3622. [Google Scholar]
  43. Yan, S.; Wang, H.; Li, Y.; Zheng, Y.; Han, L. Attention-aware metapath-based network embedding for HIN based recommendation. Expert Syst. Appl. 2021, 174, 11460. [Google Scholar]
  44. Chang, Y.; Chen, C.; Hu, W.; Zheng, Z.; Zhou, X.; Chen, S. Megnn: Meta-path extracted graph neural network for heterogeneous graph representation learning. Knowl.-Based Syst. 2022, 235, 107611. [Google Scholar] [CrossRef]
Figure 1. An instance of an HIN for top-N recommendation.
Figure 1. An instance of an HIN for top-N recommendation.
Applsci 15 03929 g001
Figure 2. The overall architecture of our TMRec.
Figure 2. The overall architecture of our TMRec.
Applsci 15 03929 g002
Figure 3. The performance of different models on TOP-10 on the DUNN dataset: (a) the performance in terms of the F-score on DUNN on TOP-10; (b) the standard deviation of the F-scores of 10 experiments.
Figure 3. The performance of different models on TOP-10 on the DUNN dataset: (a) the performance in terms of the F-score on DUNN on TOP-10; (b) the standard deviation of the F-scores of 10 experiments.
Applsci 15 03929 g003
Figure 4. The time complexity of different models.
Figure 4. The time complexity of different models.
Applsci 15 03929 g004
Table 1. Notations and their descriptions.
Table 1. Notations and their descriptions.
NotationDescription
m u User’s embedding after lookup layer
n i Item’s embedding after lookup layer
x u Low-dimensional representations of users
y i Low-dimensional representations of items
x u Improved user’s embedding after LSTM module
x ˜ u Final user’s embedding after self-attention mechanism
y ˜ i Final item’s embedding after self-attention mechanism
h p Embedding of path instance p
c p Initial embedding of meta-path
c p Embedding of meta-path after L2-normalization
c u , i Final embedding of meta-path after co-attention mechanism module
r u , i Embedding of final fusion
Table 2. Comparison with state-of-the-art models.
Table 2. Comparison with state-of-the-art models.
ModelMovielensLastFMYelp
Prec@10 Recall@10 NDCG@10 Prec@10 Recall@10 NDCG@10 Prec@10 Recall@10 NDCG@10
ItemKNN0.25780.15360.56920.41600.45130.79810.13860.54210.5378
BPR0.30110.19460.64590.41290.44920.80990.14030.56510.5531
MF0.32470.20530.65110.43640.46340.79210.14810.59910.6011
NCF0.32560.21650.66820.4540.46780.81040.14900.60010.6007
HAN0.33560.21550.68450.47200.49780.84210.16020.61230.6134
MCRec0.34150.22130.68760.47700.50280.84910.16740.63120.6300
GraFC2T20.34000.21810.68660.47640.49910.84330.16520.62120.6274
MEGNN0.34100.22140.68610.47730.50180.84900.16700.63050.6305
TMRec0.35680.23810.70110.49330.51860.86910.18510.64970.6485
Table 3. The results of the ablation studies.
Table 3. The results of the ablation studies.
ModelTMRecTMRec-DPSATMRec-LSTMTMRec-RSA
MovielensPrec@100.35680.35270.35610.3564
Recall@100.23810.23380.23540.2355
NDCG@100.70110.69950.70020.7007
LastFMPrec@100.49330.48690.49120.4917
Recall@100.51860.51490.51730.5177
NDCG@100.86810.86420.86300.8639
YelpPrec@100.18510.18050.18230.1827
Recall@100.64970.64360.64710.6475
NDCG@100.64850.64420.64530.6458
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Hu, F.; Lin, J. Temporal Enhancement of Top-N Recommendation on Heterogeneous Graphs. Appl. Sci. 2025, 15, 3929. https://doi.org/10.3390/app15073929

AMA Style

Hu F, Lin J. Temporal Enhancement of Top-N Recommendation on Heterogeneous Graphs. Applied Sciences. 2025; 15(7):3929. https://doi.org/10.3390/app15073929

Chicago/Turabian Style

Hu, Feng, and Jun Lin. 2025. "Temporal Enhancement of Top-N Recommendation on Heterogeneous Graphs" Applied Sciences 15, no. 7: 3929. https://doi.org/10.3390/app15073929

APA Style

Hu, F., & Lin, J. (2025). Temporal Enhancement of Top-N Recommendation on Heterogeneous Graphs. Applied Sciences, 15(7), 3929. https://doi.org/10.3390/app15073929

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop