Traffic flow prediction is a very challenging task in traffic networks. Travel order quantity prediction is of great value to the analysis of traffic flow. However, the number of travel orders is ...closely related to time, leading to its short-term proximity and long-term cyclical dependence. Therefore, it is difficult to predict travel order quantity using traditional methods. To capture both the long-term and short-term correlations, in this paper, we propose an attention-based bidirectional long short-term memory network (AT-BLSTM) model, which consists of bidirectional long short-term memory (BLSTM) layer and attention layer. The BLSTM layer contains forward and backward long short-term memory (LSTM), which uses a novel method to combine forward and backward output. The attention layer utilizes novel self-attention algorithm to assign different weights according to the correlation between features. In this case, AT-BLSTM can predict travel order quantity more accurately than other time series models. For example, AT-BLSTM achieves the lowest MAE (0.0647), the lowest RMSE (0.0836) and the lowest MAPE (0.1239) among all the methods on dataset Xiuying. Extensive experiments on real-world travel datasets offer evidence that the proposed approach matches or outperforms state-of-the-art methods.
Full text
Available for:
DOBA, EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, IZUM, KILJ, KISLJ, MFDPS, NLZOH, NUK, OILJ, PILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, SIK, UILJ, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
Graph neural networks (GNNs) have many variants for graph representation learning. Several works introduce PageRank into GNNs to improve its neighborhood aggregation capabilities. However, these ...methods leverage the general PageRank to perform complex neighborhood aggregation to obtain the final feature representation, which leads to high computational cost and oversmoothing. In this paper, we propose simple hierarchical PageRank graph neural networks (SHP-GNNs), which first utilize the simple PageRank to aggregate different neighborhood ranges of each node and then leverage a jumping architecture to combine these aggregated features to enable hierarchical structure-aware representation. In this case, first, the simple PageRank turns the neighborhood aggregation process to no-learning, thereby reducing the computational complexity of the model. Then, the jumping structure combines the aggregation features of each node’s different hierarchy (neighborhood range) to learn more informative feature representation. Finally, the successful combination of the above methods alleviates the oversmoothing problem of deep GNNs. Our experimental evaluation demonstrates that SHP-GNNs achieve or match state-of-the-art results in node classification tasks, text classification tasks, and community prediction tasks. Moreover, since SHP-GNNs’ neighborhood aggregation is a no-learning process, SHP-GNNs are more suitable for node clustering tasks.
Full text
Available for:
EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, KILJ, KISLJ, MFDPS, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
Online ride-hailing order forecasting is a very important part of the intelligent traffic dispatch system. Accurate order forecasting can reduce the flow of invalid vehicles and improve the user ...experience of online ride-hailing. We propose a multi-view deep long short-term memory (LSTM) network architecture (MultiView deep LSTM framework), which uses convolutional neural network and graph convolutional network to extract the temporal and spatial characteristics of online ride-hailing orders, obtains the correlation information between regional orders through the order view, regional speed view, and weather factor view, and then uses LSTM unit and attention unit to predict the order volume in real time. We use Didi Haikou, China’s online ride-hailing dataset for training, compare it with the prediction algorithms of other articles, and experiment with different choices of the contrast framework. The experimental results show that our deep learning framework can effectively capture comprehensive spatio-temporal correlation and obtain better results. The model maintained good performance at 15 min, 30 min, and 1 h. Experiments conducted on the actual demand data onto ride-hailing from Didi Haikou data prove that our method is better than the latest method.
Full text
Available for:
DOBA, EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, IZUM, KILJ, KISLJ, MFDPS, NLZOH, NUK, OILJ, PILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, SIK, UILJ, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
Graph neural networks (GNNs) leverage graph convolutions or their approximations to extract features of nodes from graph-structured data. Nevertheless, these methods only combine information from ...nodes’ neighborhoods, without taking into account the impact of other nodes outside neighborhoods. To address the shortcomings, we present full-graph attention neural networks (FGANNs), novel neural network architectures that consider the impact of all nodes when performing self-attention, leveraging masked attention to enable (implicitly) specifying different weights to different nodes in a neighborhood. Under such circumstances, we address several important challenges of spectral-based graph neural networks simultaneously, and make FGANN readily available to semi-supervised classification problems. Extensive experiments on citation networks offer evidence that the proposed approach outperforms state-of-the-art methods.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Graph convolutional networks (GCNs) and their variants are excellent deep learning methods for graph-structured data. Moreover, multilayer GCNs can perform feature smoothing repeatedly, which creates ...considerable performance improvements. However, they may inherit unnecessary complexity and redundant computation; to make matters worse, they introduce overfitting as the number of layers increases. In this paper, we present simplified multilayer graph convolutional networks with dropout (DGCs), novel neural network architectures that successively perform nonlinearity removal and weight matrix merging between graph conventional layers, leveraging a dropout layer to achieve feature augmentation and effectively reduce overfitting. Under such circumstances, first, we extend a shallow GCN to a multilayer GCN. Then, we reduce the complexity and redundant calculations of the multilayer GCN, while improving its classification performance. Finally, we make DGCs readily applicable to inductive and transductive tasks. Extensive experiments on citation networks and social networks offer evidence that the proposed model matches or outperforms state-of-the-art methods.
Full text
Available for:
CEKLJ, EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, KILJ, KISLJ, MFDPS, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
Traffic prediction (forecasting) is a key problem in intelligent transportation. It helps engineers to obtain traffic trends in advance so that they can make favorable decisions quickly and ...accurately, or even improve the working mode of existing systems. However, it is very challenging to design a model for such problem that fully utilize the factors related to traffic. This paper investigates machine learning in traffic prediction and proposes Multiple Information Spatial–Temporal Attention based Graph Convolution Networks (MISTAGCN). The model consists of two parts. The first part utilizes combinations of different inputs and graph structures to compute the corresponding latent variables. In the second part of the model, the latent variables are comprehensively integrated and deeply mined. In particular, as the basic component of the model, a STBlock integrates mechanisms such as temporal attention, spatial attention, graph convolution, ordinary convolution and residual connections to fully explore the potential information contained in the data. Experiments on Haikou online car-hailing dataset and New York yellow taxi trip dataset illustrate that the proposed model outperforms state-of-art baseline models.
•A novel deep learning model for multi–steps traffic prediction problems is proposed.•A unified STBlock structure is devised to process multi channel information.•The same STBlock structure is utilized for information fusion.•Novel formulas are devised to calculate spatial and temporal attentions.•Extensive experiments are conducted to verify the competitiveness of this model.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
7.
Hybrid deep graph convolutional networks Yang, Fei; Zhang, Huyin; Tao, Shiming
International journal of machine learning and cybernetics,
08/2022, Volume:
13, Issue:
8
Journal Article
Peer reviewed
Graph neural networks (GNNs) leverage graph convolutions or their approximations to cope with graph-structured data. According to whether convolution is applied to the spectral domain or spatial ...domain, advances in this direction are generally divided into spectral approaches and spatial approaches. However, current study generally utilizes one of the approaches, while ignoring the combination of the two approaches. In this paper, we propose hybrid deep graph convolutional networks (HDGCNs), novel neural network architectures that combine the spectral approach and the spatial approach to calculate the adjacency matrix, leveraging the deep graph convolutional networks (GCNs) to release the advantages of this combination. In this way, the gap between spectral approaches and spatial approaches is eliminated, and the characteristics of the two approaches are merged together. Extensive experiments on citation networks and web networks offer evidence that the proposed models outperform state-of-the-art methods.
Full text
Available for:
EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, KILJ, KISLJ, MFDPS, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
In this study we find a global minimizer of a concave function over a sphere. By introducing a differential equation, we obtain the invariant characteristics for a given optimization problem by ...constructing a canonical dual function. We present two theorems concerning the global optimality of an extrema of the optimization problem.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Urban online ride-hailing demand forecasting is an important component of smart city transportation systems. An accurate online ride-hailing demand prediction model can help cities allocate online ...ride-hailing resources reasonably, reduce energy waste, and reduce traffic congestion. With the massive popularity of online ride-hailing, we can collect a large amount of order data, and how to use deep learning models for improving order prediction accuracy has become a hot research topic. Most of the urban online taxi demand forecasting methods do not sufficiently consider the influencing factors and cannot model the complex nonlinear spatio-temporal relationships. Therefore, we propose a multi-view deep spatio-temporal network framework (MVDSTN) architecture to obtain the spatio-temporal relationships for online ride-hailing demand prediction. Our proposed model includes five views,up-passenger order view, down-passenger order view, POI view, spatial GCN view, POI view and weather view, applies LSTM with attention mechanism to achieve demand prediction for urban online taxi bodies. Experiments Haikou Didi Taxi datasets and Wuhan Taxi datasets prove that our model has good robustness and the prediction method outperforms current methods.
Full text
Available for:
CEKLJ, EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, KILJ, KISLJ, MFDPS, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
Recent graph neural networks for graph representation learning depend on a neighborhood aggregation process. Several works focus on simplifying the neighborhood aggregation process and model ...structures. However, as the depth of the models increases, the simplified models will encounter oversmoothing, resulting in a decrease in model performance. Several works leverage sophisticated learnable neighborhood aggregation algorithms to learn more accurate graph representations. However, the high computational cost limits the depth of these models and the ability to tackle large graphs. In this paper, we propose simple jumping knowledge networks (SJK-Nets), which first leverage a simple no-learning method to complete the neighborhood aggregation process, and then utilize a jumping architecture to combine the different neighborhood ranges of each node to achieve a better structure-aware representation. Under such circumstances, first, we use a simple neighborhood aggregation algorithm to reduce computational complexity of the model. Then, we aggregate the features of high-order neighboring nodes to learn more informative node feature representations. Finally, by combining the above methods, the oversmoothing problem of the deep graph neural networks is alleviated. Our experimental evaluation demonstrates that SJK-Nets achieve or match state-of-the-art results in node classification tasks, text classification tasks, and community prediction tasks. Moreover, since SJK-Nets’ neighborhood aggregation is a no-learning process, SJK-Nets are successfully extended to node clustering tasks.
Full text
Available for:
CEKLJ, EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, KILJ, KISLJ, MFDPS, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
PDF