Pub Date : 2024-08-02DOI: 10.1109/TKDE.2024.3436909
Yao Cheng;Minjie Chen;Caihua Shan;Xiang Li
Graph neural networks (GNNs) have recently received significant attention. Learning node-wise message propagation in GNNs aims to set personalized propagation steps for different nodes in the graph. Despite the success, existing methods ignore node priority that can be reflected by node influence and heterophily. In this paper, we propose a versatile framework PriPro, which can be integrated with most existing GNN models and aim to learn prioritized node-wise message propagation in GNNs. Specifically, the framework consists of three components: a backbone GNN model, a propagation controller to determine the optimal propagation steps for nodes, and a weight controller to compute the priority scores for nodes. We design a mutually enhanced mechanism to compute node priority, optimal propagation step and label prediction. We also propose an alternative optimization strategy to learn the parameters in the backbone GNN model and two parametric controllers. We conduct extensive experiments to compare our framework with other 12 state-of-the-art competitors on 10 benchmark datasets. Experimental results show that our framework can lead to superior performance in terms of propagation strategies and node representations.
{"title":"Learning Prioritized Node-Wise Message Propagation in Graph Neural Networks","authors":"Yao Cheng;Minjie Chen;Caihua Shan;Xiang Li","doi":"10.1109/TKDE.2024.3436909","DOIUrl":"10.1109/TKDE.2024.3436909","url":null,"abstract":"Graph neural networks (GNNs) have recently received significant attention. Learning node-wise message propagation in GNNs aims to set personalized propagation steps for different nodes in the graph. Despite the success, existing methods ignore node priority that can be reflected by node influence and heterophily. In this paper, we propose a versatile framework PriPro, which can be integrated with most existing GNN models and aim to learn prioritized node-wise message propagation in GNNs. Specifically, the framework consists of three components: a backbone GNN model, a propagation controller to determine the optimal propagation steps for nodes, and a weight controller to compute the priority scores for nodes. We design a mutually enhanced mechanism to compute node priority, optimal propagation step and label prediction. We also propose an alternative optimization strategy to learn the parameters in the backbone GNN model and two parametric controllers. We conduct extensive experiments to compare our framework with other 12 state-of-the-art competitors on 10 benchmark datasets. Experimental results show that our framework can lead to superior performance in terms of propagation strategies and node representations.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"36 12","pages":"8670-8681"},"PeriodicalIF":8.9,"publicationDate":"2024-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141883007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-02DOI: 10.1109/TKDE.2024.3437781
Jie Bai;Kang Zhao;Linjing Li;Daniel Zeng;Qiudan Li;Fan Yang;Quannan Zu
Graph representation learning is an emerging area for graph analysis and inference. However, existing approaches for large-scale graphs either sample nodes in sequential walks or manipulate the adjacency matrices of graphs. The former approach can cause sampling bias against less-connected nodes, whereas the latter may suffer from sparsity that exists in many real-world graphs. To learn from structural information in a graph more efficiently and comprehensively, this paper proposes a new graph representation learning approach inspired by the cognitive model of spreading-activation mechanisms in human memory. This approach learns node embeddings by adopting a graph activation model that allows nodes to “activate” their neighbors and spread their own structural information to other nodes through the paths simultaneously. Comprehensive experiments demonstrate that the proposed model performs better than existing methods on several empirical datasets for multiple graph inference tasks. Meanwhile, the spreading-activation-based model is computationally more efficient than existing approaches–the training process converges after only a small number of iterations, and the training time is linear in the number of edges in a graph. The proposed method works for both homogeneous and heterogeneous graphs.
{"title":"Graph Representation Learning Based on Cognitive Spreading Activations","authors":"Jie Bai;Kang Zhao;Linjing Li;Daniel Zeng;Qiudan Li;Fan Yang;Quannan Zu","doi":"10.1109/TKDE.2024.3437781","DOIUrl":"10.1109/TKDE.2024.3437781","url":null,"abstract":"Graph representation learning is an emerging area for graph analysis and inference. However, existing approaches for large-scale graphs either sample nodes in sequential walks or manipulate the adjacency matrices of graphs. The former approach can cause sampling bias against less-connected nodes, whereas the latter may suffer from sparsity that exists in many real-world graphs. To learn from structural information in a graph more efficiently and comprehensively, this paper proposes a new graph representation learning approach inspired by the cognitive model of spreading-activation mechanisms in human memory. This approach learns node embeddings by adopting a graph activation model that allows nodes to “activate” their neighbors and spread their own structural information to other nodes through the paths simultaneously. Comprehensive experiments demonstrate that the proposed model performs better than existing methods on several empirical datasets for multiple graph inference tasks. Meanwhile, the spreading-activation-based model is computationally more efficient than existing approaches–the training process converges after only a small number of iterations, and the training time is linear in the number of edges in a graph. The proposed method works for both homogeneous and heterogeneous graphs.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"36 12","pages":"8408-8420"},"PeriodicalIF":8.9,"publicationDate":"2024-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141883008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-02DOI: 10.1109/TKDE.2024.3437364
Aihong Yuan;Mengbo You;Yuhan Wang;Xun Li;Xuelong Li
Unsupervised feature selection (UFS) is an important technology for dimensionality reduction and has gained great interest in a wide range of fields. Recently, most popular methods are spectral-based which frequently use adaptive graph constraints to promote performance. However, no literature has considered the grouping characteristic of the data features, which is the most basic and important characteristic for arbitrary data. In this paper, based on the spectral analysis method, we first simulate the data feature grouping characteristic. Then, the similarity between data is adaptively reconstructed through the similarity between groups, which can explore the more fine-grained relationship between data than the previous adaptive graph methods. In order to achieve the aforementioned goal, the local similarity matrix and the global similarity matrix are defined, and the weighted KL entropy is used to constrain the relationship between the global similarity matrix and the local similarity matrices. Furthermore, the symmetrical self-representation structure is used to improve the performance of the reconstruction error term in the conventional spectral-based methods. After the model is constructed, a simple but efficient algorithm is proposed to solve the full model. Extensive experiments on 8 benchmark dataset with different types to show the effectiveness of the proposed method.
{"title":"Symmetrical Self-Representation and Data-Grouping Strategy for Unsupervised Feature Selection","authors":"Aihong Yuan;Mengbo You;Yuhan Wang;Xun Li;Xuelong Li","doi":"10.1109/TKDE.2024.3437364","DOIUrl":"10.1109/TKDE.2024.3437364","url":null,"abstract":"<italic>Unsupervised feature selection (UFS)</i>\u0000 is an important technology for dimensionality reduction and has gained great interest in a wide range of fields. Recently, most popular methods are spectral-based which frequently use adaptive graph constraints to promote performance. However, no literature has considered the grouping characteristic of the data features, which is the most basic and important characteristic for arbitrary data. In this paper, based on the spectral analysis method, we first simulate the data feature grouping characteristic. Then, the similarity between data is adaptively reconstructed through the similarity between groups, which can explore the more fine-grained relationship between data than the previous adaptive graph methods. In order to achieve the aforementioned goal, the local similarity matrix and the global similarity matrix are defined, and the weighted KL entropy is used to constrain the relationship between the global similarity matrix and the local similarity matrices. Furthermore, the symmetrical self-representation structure is used to improve the performance of the reconstruction error term in the conventional spectral-based methods. After the model is constructed, a simple but efficient algorithm is proposed to solve the full model. Extensive experiments on 8 benchmark dataset with different types to show the effectiveness of the proposed method.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"36 12","pages":"9348-9360"},"PeriodicalIF":8.9,"publicationDate":"2024-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141883078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-31DOI: 10.1109/TKDE.2024.3436076
Ziyang Liu;Chaokun Wang;Hao Feng;Ziyang Chen
Graph embedding aims to extract low-dimensional representation vectors, commonly referred to as embeddings, from graph data. The generated embeddings simplify subsequent data analysis and machine learning tasks. Recently, researchers have proposed the use of contrastive learning on graphs to extract node embeddings in an unsupervised manner. Although existing graph contrastive learning methods have significantly advanced this field, there is still potential for further exploration, particularly in optimizing training efficiency and enhancing embedding quality . In this paper, we propose an efficient unsupervised graph embedding method named GEARED. First, the method involves an attributed graph reduction module that converts the raw graph into a reduced graph, greatly improving model training efficiency. Second, GEARED employs a dual-level loss with adaptive scaling factors to ensure the acquisition of high-quality embeddings. Finally, we conduct a partial derivative analysis to elucidate the specific mechanisms through which GEARED is capable of generating high-quality embeddings. Extensive experimental evaluations on 14 benchmark datasets show that GEARED consistently outperforms state-of-the-art methods in terms of training efficiency and classification accuracy. For instance, GEARED achieves a training speedup of over 40 times on both the CS and Physics datasets while maintaining superior classification accuracy.
{"title":"Efficient Unsupervised Graph Embedding With Attributed Graph Reduction and Dual-Level Loss","authors":"Ziyang Liu;Chaokun Wang;Hao Feng;Ziyang Chen","doi":"10.1109/TKDE.2024.3436076","DOIUrl":"10.1109/TKDE.2024.3436076","url":null,"abstract":"Graph embedding aims to extract low-dimensional representation vectors, commonly referred to as embeddings, from graph data. The generated embeddings simplify subsequent data analysis and machine learning tasks. Recently, researchers have proposed the use of contrastive learning on graphs to extract node embeddings in an unsupervised manner. Although existing graph contrastive learning methods have significantly advanced this field, there is still potential for further exploration, particularly in optimizing \u0000<italic>training efficiency</i>\u0000 and enhancing \u0000<italic>embedding quality</i>\u0000. In this paper, we propose an efficient unsupervised graph embedding method named GEARED. First, the method involves an attributed graph reduction module that converts the raw graph into a reduced graph, greatly improving model training efficiency. Second, GEARED employs a dual-level loss with adaptive scaling factors to ensure the acquisition of high-quality embeddings. Finally, we conduct a partial derivative analysis to elucidate the specific mechanisms through which GEARED is capable of generating high-quality embeddings. Extensive experimental evaluations on 14 benchmark datasets show that GEARED consistently outperforms state-of-the-art methods in terms of training efficiency and classification accuracy. For instance, GEARED achieves a training speedup of over 40 times on both the CS and Physics datasets while maintaining superior classification accuracy.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"36 12","pages":"8120-8134"},"PeriodicalIF":8.9,"publicationDate":"2024-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141871191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Accurately forecasting long-term future wind power is critical to achieve safe power grid integration. This problem is quite challenging due to wind power's high volatility and randomness. In this paper, we propose a novel time series forecasting method, namely Deep Conditional Generative Spatio-Temporal model (DCGST), and its high accuracy is achieved by tackling two critical issues simultaneously: a proper handling of the non-stationarity of multiple wind power time series, and a fine-grained modeling of their complicated yet dynamic spatio-temporal dependencies. Specifically, we first formally define the Spatio-Temporal Concept Drift (STCD) problem of wind power, and then we propose a novel deep conditional generative model to learn probabilistic distributions of future wind power values under STCD. Three different tailored neural networks are designed for distributions parameterization, including a graph-based prior network, an attention-based recognition network, and a stochastic seq2seq-based generation network. They are able to encode the dynamic spatio-temporal dependencies of multiple wind power time series and infer one-to-many mappings for future wind power generation. Compared to existing methods, DCGST can learn better spatio-temporal representations of wind power data and learn better uncertainties of data distribution to generate future values. Comprehensive experiments on real-world datasets including the largest public turbine-level wind power dataset verify the effectiveness, efficiency, generality and scalability of our method.
{"title":"Towards Effective Long-Term Wind Power Forecasting: A Deep Conditional Generative Spatio-Temporal Approach","authors":"Peiyu Yi;Zhifeng Bao;Feihu Huang;Jince Wang;Jian Peng;Linghao Zhang","doi":"10.1109/TKDE.2024.3435859","DOIUrl":"10.1109/TKDE.2024.3435859","url":null,"abstract":"Accurately forecasting long-term future wind power is critical to achieve safe power grid integration. This problem is quite challenging due to wind power's high volatility and randomness. In this paper, we propose a novel time series forecasting method, namely Deep Conditional Generative Spatio-Temporal model (DCGST), and its high accuracy is achieved by tackling two critical issues simultaneously: a proper handling of the non-stationarity of multiple wind power time series, and a fine-grained modeling of their complicated yet dynamic spatio-temporal dependencies. Specifically, we first formally define the \u0000<italic>Spatio-Temporal Concept Drift</i>\u0000 (STCD) problem of wind power, and then we propose a novel deep conditional generative model to learn probabilistic distributions of future wind power values under STCD. Three different tailored neural networks are designed for distributions parameterization, including a graph-based prior network, an attention-based recognition network, and a stochastic seq2seq-based generation network. They are able to encode the dynamic spatio-temporal dependencies of multiple wind power time series and infer one-to-many mappings for future wind power generation. Compared to existing methods, DCGST can learn better spatio-temporal representations of wind power data and learn better uncertainties of data distribution to generate future values. Comprehensive experiments on real-world datasets including the largest public turbine-level wind power dataset verify the effectiveness, efficiency, generality and scalability of our method.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"36 12","pages":"9403-9417"},"PeriodicalIF":8.9,"publicationDate":"2024-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141871123","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Graph clustering, which involves the partitioning of nodes within a graph into disjoint clusters, holds significant importance for numerous subsequent applications. Recently, contrastive learning, known for utilizing supervisory information, has demonstrated encouraging results in deep graph clustering. This methodology facilitates the learning of favorable node representations for clustering by attracting positively correlated node pairs and distancing negatively correlated pairs within the representation space. Nevertheless, a significant limitation of existing methods is their inadequacy in thoroughly exploring node-wise similarity. For instance, some hypothesize that the node similarity matrix within the representation space is identical, ignoring the inherent semantic relationships among nodes. Given the fundamental role of instance similarity in clustering, our research investigates contrastive graph clustering from the perspective of the node similarity matrix. We argue that an ideal node similarity matrix within the representation space should accurately reflect the inherent semantic relationships among nodes, ensuring the preservation of semantic similarities in the learned representations. In response to this, we introduce a new framework, Reliable Node Similarity Matrix Guided Contrastive Graph Clustering (NS4GC), which estimates an approximately ideal node similarity matrix within the representation space to guide representation learning. Our method introduces node-neighbor alignment and semantic-aware sparsification, ensuring the node similarity matrix is both accurate and efficiently sparse. Comprehensive experiments conducted on 8 real-world datasets affirm the efficacy of learning the node similarity matrix and the superior performance of NS4GC.
{"title":"Reliable Node Similarity Matrix Guided Contrastive Graph Clustering","authors":"Yunhui Liu;Xinyi Gao;Tieke He;Tao Zheng;Jianhua Zhao;Hongzhi Yin","doi":"10.1109/TKDE.2024.3435887","DOIUrl":"10.1109/TKDE.2024.3435887","url":null,"abstract":"Graph clustering, which involves the partitioning of nodes within a graph into disjoint clusters, holds significant importance for numerous subsequent applications. Recently, contrastive learning, known for utilizing supervisory information, has demonstrated encouraging results in deep graph clustering. This methodology facilitates the learning of favorable node representations for clustering by attracting positively correlated node pairs and distancing negatively correlated pairs within the representation space. Nevertheless, a significant limitation of existing methods is their inadequacy in thoroughly exploring node-wise similarity. For instance, some hypothesize that the node similarity matrix within the representation space is identical, ignoring the inherent semantic relationships among nodes. Given the fundamental role of instance similarity in clustering, our research investigates contrastive graph clustering from the perspective of the node similarity matrix. We argue that an ideal node similarity matrix within the representation space should accurately reflect the inherent semantic relationships among nodes, ensuring the preservation of semantic similarities in the learned representations. In response to this, we introduce a new framework, Reliable Node Similarity Matrix Guided Contrastive Graph Clustering (NS4GC), which estimates an approximately ideal node similarity matrix within the representation space to guide representation learning. Our method introduces node-neighbor alignment and semantic-aware sparsification, ensuring the node similarity matrix is both accurate and efficiently sparse. Comprehensive experiments conducted on 8 real-world datasets affirm the efficacy of learning the node similarity matrix and the superior performance of NS4GC.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"36 12","pages":"9123-9135"},"PeriodicalIF":8.9,"publicationDate":"2024-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141871193","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Relation extraction is an important task in the field of natural language processing. Previous works mainly focus on adopting pipeline methods or joint methods to model relation extraction in general scenarios. However, these existing methods face challenges when adapting to complex relation extraction scenarios, such as handling overlapped triplets, multiple triplets, and cross-sentence triplets. In this paper, we revisit the advantages and disadvantages of the aforementioned methods in complex relation extraction. Based on the in-depth analysis, we propose a novel two-stage bidirectional extract-then-evaluate framework named BeeRe . In the extraction stage, we first obtain the subject set, relation set, and object set. Then, we design subject- and object-oriented triplet extractors to iteratively recurrent obtain candidate triplets, ensuring high recall. In the evaluation stage, we adopt a relation-oriented triplet filter to determine subject-object pairs based on relations in triplets obtained in the first stage, ensuring high precision. We conduct extensive experiments on three public datasets to show that BeeRe achieves state-of-the-art performance in both complex and general relation extraction scenarios. Even when compared to large language models like closed-source/open-source LLMs, BeeRe still has significant performance gains.
{"title":"A Bidirectional Extraction-Then-Evaluation Framework for Complex Relation Extraction","authors":"Weiyan Zhang;Jiacheng Wang;Chuang Chen;Wanpeng Lu;Wen Du;Haofen Wang;Jingping Liu;Tong Ruan","doi":"10.1109/TKDE.2024.3435765","DOIUrl":"10.1109/TKDE.2024.3435765","url":null,"abstract":"Relation extraction is an important task in the field of natural language processing. Previous works mainly focus on adopting pipeline methods or joint methods to model relation extraction in general scenarios. However, these existing methods face challenges when adapting to complex relation extraction scenarios, such as handling overlapped triplets, multiple triplets, and cross-sentence triplets. In this paper, we revisit the advantages and disadvantages of the aforementioned methods in complex relation extraction. Based on the in-depth analysis, we propose a novel two-stage bidirectional extract-then-evaluate framework named \u0000<sc>BeeRe</small>\u0000. In the extraction stage, we first obtain the subject set, relation set, and object set. Then, we design subject- and object-oriented triplet extractors to iteratively recurrent obtain candidate triplets, ensuring high recall. In the evaluation stage, we adopt a relation-oriented triplet filter to determine subject-object pairs based on relations in triplets obtained in the first stage, ensuring high precision. We conduct extensive experiments on three public datasets to show that \u0000<sc>BeeRe</small>\u0000 achieves state-of-the-art performance in both complex and general relation extraction scenarios. Even when compared to large language models like closed-source/open-source LLMs, \u0000<sc>BeeRe</small>\u0000 still has significant performance gains.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"36 12","pages":"7442-7454"},"PeriodicalIF":8.9,"publicationDate":"2024-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141871192","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-29DOI: 10.1109/TKDE.2024.3434567
Wangze Ni;Peng Cheng;Lei Chen;Shiyu Yang
Recently, transportation-as-a-service (TaaS) becomes an increasing trend, and online taxi platforms start to apply electric vehicles to serve passengers. Since the recharging time of an electric vehicle is long and non-negligible, it is necessary to smartly arrange the recharging schedules of electric vehicles in working schedules. In order to maximize the number of served taxi-calling tasks, online taxi platforms assign electric vehicles whose remaining electric power is enough to serve the dynamically arriving taxi-calling tasks and schedule suitable idle vehicles to recharging piles to recharge. We formally define the power-aware electric vehicle assignment (PAEVA) problem to serve as many taxi-calling tasks as possible under the constraints of remaining electric power and deadline. We prove that the PAEVA problem is NP-hard. To solve PAEVA, we design a novel strategy to help arrange the schedules of electric vehicles. Specifically, the strategy requires that, in a time slot and an area gird, the ratio of the number of electric vehicles whose remaining electric power is higher than a threshold $alpha$ to the number of predicted taxi-calling tasks should be higher than a threshold $beta$ . We propose two approximation approaches with theoretical guarantees to adaptively determine the values of the two thresholds of the strategy. We evaluate our solutions’ effectiveness and efficiency by comprehensive experiments on real datasets.
{"title":"Task Assignment Framework for Online Car-Hailing Systems With Electric Vehicles","authors":"Wangze Ni;Peng Cheng;Lei Chen;Shiyu Yang","doi":"10.1109/TKDE.2024.3434567","DOIUrl":"10.1109/TKDE.2024.3434567","url":null,"abstract":"Recently, transportation-as-a-service (TaaS) becomes an increasing trend, and online taxi platforms start to apply electric vehicles to serve passengers. Since the recharging time of an electric vehicle is long and non-negligible, it is necessary to smartly arrange the recharging schedules of electric vehicles in working schedules. In order to maximize the number of served taxi-calling tasks, online taxi platforms assign electric vehicles whose remaining electric power is enough to serve the dynamically arriving taxi-calling tasks and schedule suitable idle vehicles to recharging piles to recharge. We formally define the power-aware electric vehicle assignment (PAEVA) problem to serve as many taxi-calling tasks as possible under the constraints of remaining electric power and deadline. We prove that the PAEVA problem is NP-hard. To solve PAEVA, we design a novel strategy to help arrange the schedules of electric vehicles. Specifically, the strategy requires that, in a time slot and an area gird, the ratio of the number of electric vehicles whose remaining electric power is higher than a threshold \u0000<inline-formula><tex-math>$alpha$</tex-math></inline-formula>\u0000 to the number of predicted taxi-calling tasks should be higher than a threshold \u0000<inline-formula><tex-math>$beta$</tex-math></inline-formula>\u0000. We propose two approximation approaches with theoretical guarantees to adaptively determine the values of the two thresholds of the strategy. We evaluate our solutions’ effectiveness and efficiency by comprehensive experiments on real datasets.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"36 12","pages":"9361-9373"},"PeriodicalIF":8.9,"publicationDate":"2024-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141871194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-29DOI: 10.1109/TKDE.2024.3434565
Letian Gong;Huaiyu Wan;Shengnan Guo;Xiucheng Li;Yan Lin;Erwen Zheng;Tianyi Wang;Zeyu Zhou;Youfang Lin
The rapid growth of location-based services (LBS) has yielded massive amounts of data on human mobility. Effectively extracting meaningful representations for user-generated check-in sequences is pivotal for facilitating various downstream services. However, the user-generated check-in data are simultaneously influenced by the surrounding objective circumstances and the user's subjective intention. Specifically, the temporal uncertainty and spatial diversity exhibited in check-in data make it difficult to capture the macroscopic spatial-temporal patterns of users and to understand the semantics of user mobility activities. Furthermore, the distinct characteristics of the temporal and spatial information in check-in sequences call for an effective fusion method to incorporate these two types of information. In this paper, we propose a novel Spatial-Temporal Cross-view Contrastive Representation (STCCR) framework for check-in sequence representation learning. Specifically, STCCR addresses the above challenges by employing self-supervision from “spatial topic” and “temporal intention” views, facilitating effective fusion of spatial and temporal information at the semantic level. Besides, STCCR leverages contrastive clustering to uncover users’ shared spatial topics from diverse mobility activities, while employing angular momentum contrast to mitigate the impact of temporal uncertainty and noise. We extensively evaluate STCCR on three real-world datasets and demonstrate its superior performance across three downstream tasks.
{"title":"Spatial-Temporal Cross-View Contrastive Pre-Training for Check-in Sequence Representation Learning","authors":"Letian Gong;Huaiyu Wan;Shengnan Guo;Xiucheng Li;Yan Lin;Erwen Zheng;Tianyi Wang;Zeyu Zhou;Youfang Lin","doi":"10.1109/TKDE.2024.3434565","DOIUrl":"10.1109/TKDE.2024.3434565","url":null,"abstract":"The rapid growth of location-based services (LBS) has yielded massive amounts of data on human mobility. Effectively extracting meaningful representations for user-generated check-in sequences is pivotal for facilitating various downstream services. However, the user-generated check-in data are simultaneously influenced by the surrounding objective circumstances and the user's subjective intention. Specifically, the temporal uncertainty and spatial diversity exhibited in check-in data make it difficult to capture the macroscopic spatial-temporal patterns of users and to understand the semantics of user mobility activities. Furthermore, the distinct characteristics of the temporal and spatial information in check-in sequences call for an effective fusion method to incorporate these two types of information. In this paper, we propose a novel Spatial-Temporal Cross-view Contrastive Representation (STCCR) framework for check-in sequence representation learning. Specifically, STCCR addresses the above challenges by employing self-supervision from “spatial topic” and “temporal intention” views, facilitating effective fusion of spatial and temporal information at the semantic level. Besides, STCCR leverages contrastive clustering to uncover users’ shared spatial topics from diverse mobility activities, while employing angular momentum contrast to mitigate the impact of temporal uncertainty and noise. We extensively evaluate STCCR on three real-world datasets and demonstrate its superior performance across three downstream tasks.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"36 12","pages":"9308-9321"},"PeriodicalIF":8.9,"publicationDate":"2024-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141871195","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-25DOI: 10.1109/TKDE.2024.3433587
Fan Yun;Zhiwen Yu;Kaixiang Yang;C. L. Philip Chen
Due to the advantages of fast training speed and competitive performance, Broad Learning System (BLS) has been widely used for classification tasks across various domains. However, the random weight generation mechanism in BLS makes the model unstable, and the performance of BLS may be limited when dealing with some complex datasets. On the other hand, the instability of BLS brings diversity to ensemble learning, and ensemble methods can also reduce the variance and bias of the single BLS. Therefore, we propose an ensemble learning algorithm based on BLS, which includes three modules. To improve the stability and generalization ability of BLS, we utilize BLS as the base classifier in an AdaBoost framework first. Taking advantage of the incremental learning mechanism of BLS, we then propose a selective ensemble method to raise the accuracy and diversity of the BLS ensemble method. In addition, based on the former selective Adaboost framework, we suggest a hierarchical ensemble algorithm, which combines sample and feature dimensions to further improve the fitting ability of the ensemble BLS. Extensive experiments have demonstrated that the proposed method performs better than the original BLS and other state-of-the-art models, proving the effectiveness and versatility of our proposed approaches.
{"title":"AdaBoost-Stacking Based on Incremental Broad Learning System","authors":"Fan Yun;Zhiwen Yu;Kaixiang Yang;C. L. Philip Chen","doi":"10.1109/TKDE.2024.3433587","DOIUrl":"10.1109/TKDE.2024.3433587","url":null,"abstract":"Due to the advantages of fast training speed and competitive performance, Broad Learning System (BLS) has been widely used for classification tasks across various domains. However, the random weight generation mechanism in BLS makes the model unstable, and the performance of BLS may be limited when dealing with some complex datasets. On the other hand, the instability of BLS brings diversity to ensemble learning, and ensemble methods can also reduce the variance and bias of the single BLS. Therefore, we propose an ensemble learning algorithm based on BLS, which includes three modules. To improve the stability and generalization ability of BLS, we utilize BLS as the base classifier in an AdaBoost framework first. Taking advantage of the incremental learning mechanism of BLS, we then propose a selective ensemble method to raise the accuracy and diversity of the BLS ensemble method. In addition, based on the former selective Adaboost framework, we suggest a hierarchical ensemble algorithm, which combines sample and feature dimensions to further improve the fitting ability of the ensemble BLS. Extensive experiments have demonstrated that the proposed method performs better than the original BLS and other state-of-the-art models, proving the effectiveness and versatility of our proposed approaches.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"36 12","pages":"7585-7599"},"PeriodicalIF":8.9,"publicationDate":"2024-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141772311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：