With the wide application of location-based social networks(LBSNs),personalized point of interest(POI)recommendation becomes popular,especially in the commercial field.Unfortunately,it is challenging to accurately rec...With the wide application of location-based social networks(LBSNs),personalized point of interest(POI)recommendation becomes popular,especially in the commercial field.Unfortunately,it is challenging to accurately recommend POIs to users because the user-POI matrix is extremely sparse.In addition,a user's check-in activities are affected by many influential factors.However,most of existing studies capture only few influential factors.It is hard for them to be extended to incorporate other heterogeneous information in a unified way.To address these problems,we propose a meta-path-based deep representation learning(MPDRL)model for personalized POI recommendation.In this model,we design eight types of meta-paths to fully utilize the rich heterogeneous information in LBSNs for the representations of users and POIs,and deeply mine the correlations between users and POIs.To further improve the recommendation performance,we design an attention-based long short-term memory(LSTM)network to learn the importance of different influential factors on a user's specific check-in activity.To verify the effectiveness of our proposed method,we conduct extensive experiments on a real-world dataset,Foursquare.Experimental results show that the MPDRL model improves at least 16.97%and 23.55%over all comparison methods in terms of the metric Precision@N(Pre@N)and Recall@N(Rec@N)respectively.展开更多
Heterogeneous information networks(HINs)have been extensively applied to real-world tasks,such as recommendation systems,social networks,and citation networks.While existing HIN representation learning methods can eff...Heterogeneous information networks(HINs)have been extensively applied to real-world tasks,such as recommendation systems,social networks,and citation networks.While existing HIN representation learning methods can effectively learn the semantic and structural features in the network,little awareness was given to the distribution discrepancy of subgraphs within a single HIN.However,we find that ignoring such distribution discrepancy among subgraphs from multiple sources would hinder the effectiveness of graph embedding learning algorithms.This motivates us to propose SUMSHINE(Scalable Unsupervised Multi-Source Heterogeneous Information Network Embedding)-a scalable unsupervised framework to align the embedding distributions among multiple sources of an HiN.Experimental results on real-world datasets in a variety of downstream tasks validate the performance of our method over the state-of-the-art heterogeneous information network embedding algorithms.展开更多
由于异构信息网络HIN(heterogeneous information network)具有丰富的语义信息而在推荐任务中得到广泛应用.传统的面向异构信息网络的推荐方法忽略了网络中关联关系的异质性,以及不同关联类型之间的相互影响.提出了一种基于多视角嵌入...由于异构信息网络HIN(heterogeneous information network)具有丰富的语义信息而在推荐任务中得到广泛应用.传统的面向异构信息网络的推荐方法忽略了网络中关联关系的异质性,以及不同关联类型之间的相互影响.提出了一种基于多视角嵌入融合的推荐模型,分别从同质关联视角和异质关联视角来挖掘异构信息网络的深层潜在特征并加以融合,有效地保证了推荐结果的准确性.针对同质关联视角,提出了一种基于图卷积神经网络的嵌入融合方法,通过对同质关联作用下节点邻域信息的轻量式卷积,实现节点嵌入的局部融合.针对异质关联视角,提出了一种基于注意力的嵌入融合方法,利用注意力机制来区分不同关联类型对节点嵌入的影响,实现节点嵌入的全局融合.通过实验验证了所提出的关键技术的可行性和有效性.展开更多
异构信息网络(Heterogeneous Information Network, HIN)凭借其丰富的语义信息和结构信息被广泛应用于推荐系统中,虽然取得了很好的推荐效果,但较少考虑局部特征放大、信息交互和多嵌入聚合等问题。针对这些问题,提出了一种新的用于top-...异构信息网络(Heterogeneous Information Network, HIN)凭借其丰富的语义信息和结构信息被广泛应用于推荐系统中,虽然取得了很好的推荐效果,但较少考虑局部特征放大、信息交互和多嵌入聚合等问题。针对这些问题,提出了一种新的用于top-N推荐的多嵌入融合推荐(Multi-embedding Fusion Recommendation, MFRec)模型。首先,该模型在用户和项目学习分支中都采用对象上下文表示网络,充分利用上下文信息以放大局部特征,增强相邻节点的交互性;其次,将空洞卷积和空间金字塔池化引入元路径学习分支,以便获取多尺度信息并增强元路径的节点表示;然后,采用多嵌入融合模块以便更好地进行用户、项目以及元路径的嵌入融合,细粒度地进行多嵌入之间的交互学习,并强调了各特征的不同重要性程度;最后,在两个公共推荐系统数据集上进行了实验,结果表明所提模型MFRec优于现有的其他top-N推荐系统模型。展开更多
针对现有模型对异质信息网络(heterogeneous information network, HIN)信息提取大部分依赖于元路径,缺乏元路径信息补充以及很少学习异质图中复杂的结构信息等问题,提出一种异质网中基于邻居节点和元路径的推荐算法(NMRec)。提取用户...针对现有模型对异质信息网络(heterogeneous information network, HIN)信息提取大部分依赖于元路径,缺乏元路径信息补充以及很少学习异质图中复杂的结构信息等问题,提出一种异质网中基于邻居节点和元路径的推荐算法(NMRec)。提取用户和物品邻居节点补充元路径缺失的信息,以卷积的方式捕获节点之间丰富的交互,通过注意力机制得到节点和元路径的嵌入表示,拼接用户、物品、邻居节点及元路径进行TOP-N推荐。在两个公开数据集上的实验结果表明,NMRec推荐性能良好,对推荐结果有良好的可解释性,与7种推荐基准算法相比,NMRec在评价指标Pre@10、Recall@10、NDGG@10上至少提升了0.21%、29%、1.46%。展开更多
Due to the heterogeneity of nodes and edges,heterogeneous network embedding is a very challenging task to embed highly coupled networks into a set of low-dimensional vectors.Existing models either only learn embedding...Due to the heterogeneity of nodes and edges,heterogeneous network embedding is a very challenging task to embed highly coupled networks into a set of low-dimensional vectors.Existing models either only learn embedding vectors for nodes or only for edges.These two methods of embedding learning are rarely performed in the same model,and they both overlook the internal correlation between nodes and edges.To solve these problems,a node and edge joint embedding model is proposed for Heterogeneous Information Networks(HINs),called NEJE.The NEJE model can better capture the latent structural and semantic information from an HIN through two joint learning strategies:type-level joint learning and element-level joint learning.Firstly,node-type-aware structure learning and edge-type-aware semantic learning are sequentially performed on the original network and its line graph to get the initial embedding of nodes and the embedding of edges.Then,to optimize performance,type-level joint learning is performed through the alternating training of node embedding on the original network and edge embedding on the line graph.Finally,a new homogeneous network is constructed from the original heterogeneous network,and the graph attention model is further used on the new network to perform element-level joint learning.Experiments on three tasks and five public datasets show that our NEJE model performance improves by about 2.83%over other models,and even improves by 6.42%on average for the node clustering task on Digital Bibliography&Library Project(DBLP)dataset.展开更多
As a powerful tool for elucidating the embedding representation of graph-structured data,Graph Neural Networks(GNNs),which are a series of powerful tools built on homogeneous networks,have been widely used in various ...As a powerful tool for elucidating the embedding representation of graph-structured data,Graph Neural Networks(GNNs),which are a series of powerful tools built on homogeneous networks,have been widely used in various data mining tasks.It is a huge challenge to apply a GNN to an embedding Heterogeneous Information Network(HIN).The main reason for this challenge is that HINs contain many different types of nodes and different types of relationships between nodes.HIN contains rich semantic and structural information,which requires a specially designed graph neural network.However,the existing HIN-based graph neural network models rarely consider the interactive information hidden between the meta-paths of HIN in the poor embedding of nodes in the HIN.In this paper,we propose an Attention-aware Heterogeneous graph Neural Network(AHNN)model to effectively extract useful information from HIN and use it to learn the embedding representation of nodes.Specifically,we first use node-level attention to aggregate and update the embedding representation of nodes,and then concatenate the embedding representation of the nodes on different meta-paths.Finally,the semantic-level neural network is proposed to extract the feature interaction relationships on different meta-paths and learn the final embedding of nodes.Experimental results on three widely used datasets showed that the AHNN model could significantly outperform the state-of-the-art models.展开更多
基金National Natural Science Foundation of China(No.61972080)Shanghai Rising-Star Program,China(No.19QA1400300)。
文摘With the wide application of location-based social networks(LBSNs),personalized point of interest(POI)recommendation becomes popular,especially in the commercial field.Unfortunately,it is challenging to accurately recommend POIs to users because the user-POI matrix is extremely sparse.In addition,a user's check-in activities are affected by many influential factors.However,most of existing studies capture only few influential factors.It is hard for them to be extended to incorporate other heterogeneous information in a unified way.To address these problems,we propose a meta-path-based deep representation learning(MPDRL)model for personalized POI recommendation.In this model,we design eight types of meta-paths to fully utilize the rich heterogeneous information in LBSNs for the representations of users and POIs,and deeply mine the correlations between users and POIs.To further improve the recommendation performance,we design an attention-based long short-term memory(LSTM)network to learn the importance of different influential factors on a user's specific check-in activity.To verify the effectiveness of our proposed method,we conduct extensive experiments on a real-world dataset,Foursquare.Experimental results show that the MPDRL model improves at least 16.97%and 23.55%over all comparison methods in terms of the metric Precision@N(Pre@N)and Recall@N(Rec@N)respectively.
基金supported by the Research Grants Council of Hong Kong(17308321)the HKUTCL Joint Research Center for Artificial Intelligence sponsored by TCL Corporate Research(Hong Kong).
文摘Heterogeneous information networks(HINs)have been extensively applied to real-world tasks,such as recommendation systems,social networks,and citation networks.While existing HIN representation learning methods can effectively learn the semantic and structural features in the network,little awareness was given to the distribution discrepancy of subgraphs within a single HIN.However,we find that ignoring such distribution discrepancy among subgraphs from multiple sources would hinder the effectiveness of graph embedding learning algorithms.This motivates us to propose SUMSHINE(Scalable Unsupervised Multi-Source Heterogeneous Information Network Embedding)-a scalable unsupervised framework to align the embedding distributions among multiple sources of an HiN.Experimental results on real-world datasets in a variety of downstream tasks validate the performance of our method over the state-of-the-art heterogeneous information network embedding algorithms.
文摘由于异构信息网络HIN(heterogeneous information network)具有丰富的语义信息而在推荐任务中得到广泛应用.传统的面向异构信息网络的推荐方法忽略了网络中关联关系的异质性,以及不同关联类型之间的相互影响.提出了一种基于多视角嵌入融合的推荐模型,分别从同质关联视角和异质关联视角来挖掘异构信息网络的深层潜在特征并加以融合,有效地保证了推荐结果的准确性.针对同质关联视角,提出了一种基于图卷积神经网络的嵌入融合方法,通过对同质关联作用下节点邻域信息的轻量式卷积,实现节点嵌入的局部融合.针对异质关联视角,提出了一种基于注意力的嵌入融合方法,利用注意力机制来区分不同关联类型对节点嵌入的影响,实现节点嵌入的全局融合.通过实验验证了所提出的关键技术的可行性和有效性.
文摘针对现有模型对异质信息网络(heterogeneous information network, HIN)信息提取大部分依赖于元路径,缺乏元路径信息补充以及很少学习异质图中复杂的结构信息等问题,提出一种异质网中基于邻居节点和元路径的推荐算法(NMRec)。提取用户和物品邻居节点补充元路径缺失的信息,以卷积的方式捕获节点之间丰富的交互,通过注意力机制得到节点和元路径的嵌入表示,拼接用户、物品、邻居节点及元路径进行TOP-N推荐。在两个公开数据集上的实验结果表明,NMRec推荐性能良好,对推荐结果有良好的可解释性,与7种推荐基准算法相比,NMRec在评价指标Pre@10、Recall@10、NDGG@10上至少提升了0.21%、29%、1.46%。
基金supported by the National Natural Science Foundation of China(No.62103143)the Hunan Province Key Research and Development Program(No.2022WK2006)+2 种基金the Special Project for the Construction of Innovative Provinces in Hunan(Nos.2020TP2018 and 2019GK4030)the Young Backbone Teacher of Hunan Province(No.2022101)the Scientific Research Fund of Hunan Provincial Education Department(No.22B0471).
文摘Due to the heterogeneity of nodes and edges,heterogeneous network embedding is a very challenging task to embed highly coupled networks into a set of low-dimensional vectors.Existing models either only learn embedding vectors for nodes or only for edges.These two methods of embedding learning are rarely performed in the same model,and they both overlook the internal correlation between nodes and edges.To solve these problems,a node and edge joint embedding model is proposed for Heterogeneous Information Networks(HINs),called NEJE.The NEJE model can better capture the latent structural and semantic information from an HIN through two joint learning strategies:type-level joint learning and element-level joint learning.Firstly,node-type-aware structure learning and edge-type-aware semantic learning are sequentially performed on the original network and its line graph to get the initial embedding of nodes and the embedding of edges.Then,to optimize performance,type-level joint learning is performed through the alternating training of node embedding on the original network and edge embedding on the line graph.Finally,a new homogeneous network is constructed from the original heterogeneous network,and the graph attention model is further used on the new network to perform element-level joint learning.Experiments on three tasks and five public datasets show that our NEJE model performance improves by about 2.83%over other models,and even improves by 6.42%on average for the node clustering task on Digital Bibliography&Library Project(DBLP)dataset.
基金supported by the Key Scientific Guiding Project for the Central Universities Research Funds(No.N2008005)the Major Science and Technology Project of Liaoning Province of China(No.2020JH1/10100008)the National Key Research and Development Program of China(No.2018YFB1701104)。
文摘As a powerful tool for elucidating the embedding representation of graph-structured data,Graph Neural Networks(GNNs),which are a series of powerful tools built on homogeneous networks,have been widely used in various data mining tasks.It is a huge challenge to apply a GNN to an embedding Heterogeneous Information Network(HIN).The main reason for this challenge is that HINs contain many different types of nodes and different types of relationships between nodes.HIN contains rich semantic and structural information,which requires a specially designed graph neural network.However,the existing HIN-based graph neural network models rarely consider the interactive information hidden between the meta-paths of HIN in the poor embedding of nodes in the HIN.In this paper,we propose an Attention-aware Heterogeneous graph Neural Network(AHNN)model to effectively extract useful information from HIN and use it to learn the embedding representation of nodes.Specifically,we first use node-level attention to aggregate and update the embedding representation of nodes,and then concatenate the embedding representation of the nodes on different meta-paths.Finally,the semantic-level neural network is proposed to extract the feature interaction relationships on different meta-paths and learn the final embedding of nodes.Experimental results on three widely used datasets showed that the AHNN model could significantly outperform the state-of-the-art models.